**3. Studies of cfDNA in blastocyst fluid (BF)**

#### **3.1 Presence of cfDNA in in blastocyst fluid (BF)**

The presence of cfDNA in blastocyst fluid (BF) was revealed as early as 2013. For the first time, Palini et al. [17] reported that genomic DNA has identified in 90% (26/29) BF by qPCR, with an average amount of gDNA at 9.9 pg. TBC1D3 on chr17 and TSPY1Y on the Y chromosome were successfully amplified, which confirmed that the sex of embryos could be determined by qPCR using BF. In addition, genome-wide amplification (WGA) was demonstrated for blastocyst fluid amplification, and aneuploidy detection of WGA products was performed using microarray comparative genomic hybridization (aCGH). The aCGH results revealed the presence of the Y chromosome in two BF samples, which was consistent with previous data obtained by PCR analysis. It also showed multiple chromosomal abnormalities that agreed with the karyotype, suggesting that the WGA product DNA was indeed of embryonic origin. Subsequently, Zhang et al. [46] performed next-generation sequencing on WGA-amplified BF DNAs and compared the data with paired blastomere biopsy samples. The results showed decent concordance in genomic coverage and pattern regions between the two DNA sources. Further analysis of the gene annotation results suggested that cfDNA in the BF contained the sequences of the majority of genes, indicating that cfDNA in the BF could be used for solving monogenic diseases.

#### **3.2 Sampling method of blastocyst fluid**

BF can be isolated from embryos for downstream analysis using blastocyst puncture, which is a microscopy-assisted puncture made with an ICSI injection pipette on the trophoblast side, opposite of the inner cell mass. The fluid is then taken until the blastocyst completely collapses around the pipette [54]. This method can only isolate minute amounts (0.01 μL) of BF for analysis [55, 56].

#### **3.3 Study of PGT-A using blastocyst fluid**

Recently, BF has been studied as a source for minimally invasive PGT-A. In 2014, Gianaroli et al. [57] isolated BF from 51 blastocysts, and the average concentration of WGA-amplified DNA detected in 39 BF samples (76.5%) was 900.38 ng/ μL (ranging between 876.3–939.5 ng/μL). The results of BF detected in 38 cases (38/ 39, 97.4%) were consistent with the ploidy of TE biopsy results, and 9 cases (9/9, 100%) were consistent with the ploidy of the blastomere biopsy, and the coincidence rate of PB ploidy was 93.3% (28/30). Magli et al. [55] of the same group

#### *Non-Invasive Chromosome Screening for Embryo Preimplantation Using Cell-Free DNA DOI: http://dx.doi.org/10.5772/intechopen.98416*

improved the sampling and amplification of BF samples to a success rate of 82% (95/116). aCGH was performed on 87 of these samples, of which 82 samples (94%) were used for subsequent analysis. The ploidy consistency of BF and TE samples was 97.1% (67/69), 94.4% (34/36) with blastomere biopsy and 94.1% (32/34) with polar body biopsy. In 2019, Magli et al. [12] used the same approach to amplify blastocyst fluid extracted from 256 blastocysts with a 71% (182/256) WGA amplification rate. Compared with the results of the corresponding TE cells, the overall ploidy consistency was 93.6% (161/172). The above results suggested that the cfDNA of the BF was highly predictive of embryo ploidy; however, the results from other research groups have been less satisfactory.

Tobler et al. [58] obtained a WGA amplification success rate of 63% (60/96) after thawing and culturing donated embryos to extract BF. The concordance between the BF karyotype and the whole embryo was 48.3% (29/60) by aCGH analysis, suggesting that BF may not be suitable for the PGT test. In 2018, Tsuiko performed chromosome analysis using NGS on 16 donated blastocysts, and the success rate of BF amplification was 87.5% (14/16). However, only 10 samples (10/16, 62.5%) passed sequencing and quality control for the subsequent analysis. The results showed that only 40% (4/10) of the BF samples completely agreed with the TE or ICM results. In addition, significantly more mosaicisms were found in BF samples than in TE and ICM samples. These results together suggest that although BF-DNA can be successfully amplified for NGS, one would not recommend using BF as a single DNA source for preimplantation genetic screening due to its low concordance with ICM and TE results. Capalbo et al. [59] (2018) performed PGT-A on 23 BF samples and compared the results with TE. Only 8 cases (34.8%) were successfully amplified, and only 3 cases (37.5%) were consistent with the ploidy of TE detection results.

In summary, studies assessing the reliability of BF DNA as a template for PGT-A have yielded conflicting conclusions, with 37.5%–97.4% concordance with TE samples (**Table 3**). Differences in consistency levels between studies may be related to differences in embryo handling. Tsuiko and Tobler used frozen embryos donated by patients after treatment, and BF was absorbed after blastocyst resuscitation and expansion; Tobler's study [58] specifically suggests that they are not suitable for clinical use. In contrast, the BF samples of Gianaroli, Magli and colleagues were obtained from fresh culture cycles and underwent many procedures (PB biopsies, assisted incubation and blastomere biopsies) that were not performed or relatively few procedures were performed in other studies. Thus, the observed increased rates of amplification and concordance may be due to the superior quality of BF obtained from freshly cultured embryos over frozen embryos or to an increase in the amount of DNA in the blastocyst cavity resulting from unintentional cell lysis or death during the procedure. It is also important to point out that the sample size in the Capalbo and Tsuiko studies was much smaller than that of Gianaroli, Magli and colleagues, so different results might have been obtained if a larger sample size had been studied.

According to the results of available studies collectively, BF DNA presented a high percentage of DNA amplification failure. The percentage of blastocyst fluid samples that successfully underwent whole genome amplification and produced detectable levels of DNA ranged from 34.8% to 82% [46, 58–60]. The difficulty in the successful amplification of BF-DNA lies in the small amount of fluid obtained from the cystic cavity, and the BF volume reported in various studies ranged from 0.3 nl to 1 μl. The volume difference would significantly impact the concentration of BF-DNA and may impose a negative effect on the efficiency of subsequent amplification. Another major reason for failure in BF-DNA amplification would be degradation. The success rate of whole genome amplification was generally much higher (≥98%) [58, 61] in TE biopsy, which makes BF DNA less suitable as an alternative source for clinical application [19].


*Summary of BF cfDNA study.*

**100**

*Non-Invasive Chromosome Screening for Embryo Preimplantation Using Cell-Free DNA DOI: http://dx.doi.org/10.5772/intechopen.98416*

In addition, blastocyst puncture is still a minimally invasive procedure, as an ICSI needle is inserted into the blastocyst cavity to extract the fluid. However, with the presence of amplifiable cfDNA in the BF, the low consistency of the test results to TE or other gold standards suggests that technical variability may still exist. The process may accidentally acquire loose cells trapped in the lumen or shed cell material, resulting in inconsistent test outcomes. Thus, further optimization is required before using BF DNA for PGT.

#### **3.4 Correlation between BF test results and clinical outcomes**

The DNA concentration in the BF was very low, and the total amount varied greatly between samples, implying that the BF DNA amount could be related to the developmental status of embryos. In 2019, Magli et al. [60] showed a significantly higher WGA amplification success rate of BF in aneuploid blastocysts (n = 150, 81%) versus euploid blastocysts (n = 32, 45%), suggesting differences in the quality or total amount of DNA in the BF of euploidy and aneuploidy embryos. Moreover, the clinical pregnancy rate was 77% in the group with failed BF amplification and 37% in the group with successful BF amplification when following the clinical outcome of 53 TE euploid blastocysts. The same trend was found in the rate of persistent pregnancies in failed and successful amplified BF samples, which were 68% and 31.5%, respectively. These clinical results suggest that the success of BF amplification may have predictive value for the viability of the corresponding blastocyst; nevertheless, further validation is required to draw conclude. Given that the total amount of DNA in the BF is relatively low, combining blastocyst fluid and culture medium may increase the amount of cfDNA, therefore improving the amplification success rate and consistency of detection.

#### **4. Problems in the application of cfDNA to PGT-A**

Before the large-scale use of SCM in PGT-A, further investigations are needed to trace the origin of cfDNA and confirm whether cfDNA represents the euploidy of embryos.

#### **4.1 The origin of cfDNA**

At present, there are several opinions about the origin of cfDNA. It is believed that during embryonic development, cfDNA may be released into the culture media through cell lysis, apoptosis, cell debris or other mechanisms [19]. In apoptotic cells, DNA is cleaved to form fragments. These fragment sizes are multiples of approximately 180-bp oligomers [62]. Zhang's [46] research shows that the fragments have two peaks. The first peak had a range of 160–220 bp, whereas the second peak was broader, ranging from 300 to 400 bp. Bolton et al. [63] found that apoptosis was frequently observed within the ICM and TE between euploid cells and aneuploid cells in a mouse model. The results also showed that in aneuploid embryos, the percentages of apoptotic cells in the ICM and TE were 41.4% and 3.3%, respectively. In euploid embryos, the percentage of apoptotic cells in euploid cells was 19.5% and 0.6%, respectively. If cfDNA mainly comes from apoptotic cells, Bolton's and other studies demonstrate that cfDNA mainly originates from the ICM.

Moreover, Victor et al. [51] tracked the development of embryos dynamically using immunofluorescence technology, especially cell proliferation and apoptosis in euploid embryos, aneuploid embryos and mosaic embryos. The TE and ICM in euploid cells showed lower level of activities in both not only in cell proliferation

but also in apoptosis. Nevertheless, in mosaic and aneuploid embryos, the levels of the two processes were both higher.

#### **4.2 Maternal contamination**

The discordant results in SCM and control embryos were mainly attributable to a high percentage of maternal DNA in the spent culture media. The Vera Rodriguez team [41] analysed the SCM and TE results from 56 samples. Among them, 17 embryos were detected as aneuploid or aneuploid males (XY) by TE, while all of them were detected as aneuploid females (XX) by SCM. When Feichtinger et al. compared the consistency between the SCM and polar body, the negative control (SCM of fertilization failure oocytes) was also effectively amplified. Oocytes are unlikely to abandon their DNA to the culture media. Therefore, this maternal contamination may come from cumulus cells or other exogenous DNA [36]. However, contamination may be minimized with the degradation of DNA in vitro by changing the solution in the sequential culture at D3 and delaying the sampling time. In sequential culture, if the granular cells before ICSI are not removed entirely, it is recommended to remove them again during the culture medium change on Day 3.

To reduce the interference of human judgement, we should develop a calculation method to eliminate maternal DNA for the purpose of internal quality control. To confirm whether there was maternal DNA contamination in the embryo medium, Hammond et al. carried out a short tandem repeat (STR) analysis on abandoned spent culture media (n = 10), media controls (n = 2) and the corresponding cumulus cells [64]. Maternal DNA from cumulus cells successfully amplified all gene loci, but the amount of DNA in the SCM was too small to identify DNA. STR analysis is a long-range amplification, and highly fragmented SCM DNA seems unsuitable for this detection. Vera Rodriguez et al. conducted SNP sequencing of three groups of samples (TE/follicular fluid DNA/embryonic SCM DNA) from 35 embryos and successfully quantified the proportion of maternal DNA contamination in SCM [41]. This suggests that SNP detection can be used to evaluate maternal contamination in SCM.

In addition, considering that the fragment size of cfDNA may be different from that of maternal contamination DNA, or according to the cleavage characteristics of cfDNA at the restriction site, it can be considered to identify the target fragment size and restriction site characteristics, and only analyse the DNA that conforms to the embryonic DNA. The hypothesis is that there is a DNA amplification and database construction method that can maintain the characteristics of DNA fragments or restriction sites, and the original cfDNA template can be amplified with high fidelity.

#### **4.3 Background cfDNA contamination of SCM**

Some studies have suggested that another source of exogenous DNA contamination is the low level of background DNA in SCM. In 2017, Hammond et al. [64] detected low baseline levels of DNA in the base media and protein supplement components of three commercial culture media. They also detected a copy of nuclear DNA in the culture media with no previous contact with embryos. Vera Rodriguez et al. [41] detected DNA quantification using qPCR in 53 spent culture media and 17 control samples (culture media with no previous contact with embryos). Then, PGT-A analysis was performed in 56 spent culture media and 11 control samples. The SCM was amplified successfully, and 11 controls generated an amplification-failure pattern, causing downstream analysis failure. In addition, Li et al. [35] found that the background DNA of the control culture did not interfere

*Non-Invasive Chromosome Screening for Embryo Preimplantation Using Cell-Free DNA DOI: http://dx.doi.org/10.5772/intechopen.98416*

with the detection results, and all samples from the same embryo had the same sex chromosome diagnosis. The results testified that although there was a low baseline of exogenous DNA in the control SCM, the effect could be ignored due to the small amount of exogenous DNA. In addition, some commercial media use human serum albumin (HSA) to improve embryo development. Because HSA can adsorb DNA, the amount that is added should be controlled within a reasonable range. We suggest that the blank culture medium should be set as the parallel culture at the same time as recovering blastocyst culture media for quality control of SCM samples.

#### **5. Prospects**

Using SCM cfDNA for PGT-A has become a hot topic in the assisted reproductive field. The current research focuses on the aneuploidy consistency between cfDNA and embryos. However, the consistency comparison needs to consider many influencing factors, such as the definition of consistency, sampling methods, analysis methods, etc. We need to reach a certain consensus and standard before making an objective evaluation. Through the summary and introduction of this review, we found that a suitable sampling time, WB as the gold standard, suitable amplification methods and NGS platforms may provide the basis for the standardization of non-invasive PGT-A.

In the future, we should develop a standardized and automatic embryo culture and sample collection system based on the operating habits and culture process of the embryo laboratory. Meanwhile, we need to combine the sample downstream detection technology with corresponding software analysis to select the best embryos, which are those with the lowest probability of chromosomal abnormalities, for transplantation to improve the success rate of the whole IVF-ET cycle. It is believed that with the development of technology, the realization of automatic collection and detection platforms, the accumulation of detectable sample sizes, and the average cost of PGT-A detection on each embryo will be greatly reduced, and it is encouraging that SCM sample collection and detection will be developed into the routine process of embryo laboratories.

With the accumulation of data, SCM can be used to develop detection content for different populations. For the low-risk population with chromosomal abnormalities, we can give the risk value of chromosomal abnormalities on chromosome 16, which easily causes abortion, and on chromosomes 13, 15, 18, 21 and X, which easily cause birth defects. Referring to the NIPT method, the influence, caused by a proportion of false positive cfDNA, will be reduced. For high-risk populations with chromosomal abnormalities, such as elderly individuals, the consistency between SCM test results and embryos is high, and the interference of false-positives is relatively low. For this group of people, it is recommended to conduct whole chromosome screening, determine the risk value of chromosomal abnormalities of each embryo and determine the order of embryo transfer according to the risk value. For people with clear indications of PGT-A, we suggest that SCM samples be stored while biopsy samples are collected, especially for the laboratory or biopsy operator who has just performed biopsy. If there is no result or the result of biopsy cannot be used, SCM test results could provide remedial measures for the failure of detection due to sample loss or operation. In addition, for IVF cycles with poor quality or a small number of embryos, we should minimize the damage of biopsy operations to embryos and consider the detection of chromosome aneuploidy by SCM under the premise of full knowledge.

Exogenous contamination of embryos has been a more concerning issue at present. To improve the accuracy of detection, DNA fragments from embryos should be identified by differentiated DNA fragment length or linkage analysis in a bioinformatics platform. In addition, due to the different fertilization methods of IVF embryos, researchers have been worried that sperm will interfere with SCM. If we find a way to eliminate maternal or paternal contamination, it will be supplied to IVF embryos. We can use this technology to eliminate the problem of chromosome aneuploidy through non-invasive PGT-A to achieve ideal clinical outcomes for a wider population.

At the same time, non-invasive PGT detection can try not only for aneuploidy detection but expand to the comprehensive evaluation on the based on morphology and take into account the DNA concentration, chimaeric ratio, resolution, consistency with the gold standard and other factors of embryonic SCM. With the accumulation of clinical outcomes, it can also be combined with the clinical data of patients as an index to predict the clinical outcomes of embryos.

A noninvasive artificial intelligence embryo evaluation model could be established, which not only provides suggestions for clinicians on the order of embryo implantation but also provides patients with the most suitable and economical detection scheme, ultimately saving time for pregnancy and improving the overall success rate of IVF-ET.
