ORIGINAL ARTICLES: ASSISTED REPRODUCTION

Diagnostic efficacy of blastocoel fluid

and spent media as sources of DNA
for preimplantation genetic testing in
standard clinical conditions
Antonio Capalbo, Ph.D.,a,b Valeria Romanelli, Ph.D.,a Cristina Patassini, M.Sc.,a Maurizio Poli, D.Phil.,a,c
Laura Girardi, M.Sc.,a Adriano Giancani, M.Sc.,d Marta Stoppa, M.Sc.,d Danilo Cimadomo, Ph.D.,d
Filippo Maria Ubaldi, M.D.,d and Laura Rienzi, M.Sc.d
a
IGENOMIX, Marostica, Italy; b University of Rome, Sapienza, Dipartimento di Scienze Anatomiche, Istologiche, Medico-
Legali e dell’Apparato Locomotore, Sezione Istologia ed Embriologia Medica, Rome, Italy; c REPROOMICS, Amsterdam,
the Netherlands; d GENERA Centers for Reproductive Medicine, Clinica Valle Giulia, Rome, Italy

Objective: To determine whether blastocoel fluid (BF) or spent blastocyst medium (SBM) is a suitable template for genotype and/or
karyotype assessment of in vitro fertilization–generated embryos.
Design: Prospective blinded study.
Setting: Genetic laboratory.
Patient(s): From 26 patients undergoing preimplantation genetic testing (PGT) treatments, 103 trophectoderms (TE), 92 BF samples,
and 72 SBM samples.
Intervention(s): The BF and SBM were retrieved at the time of TE biopsy. Two DNA extraction strategies were evaluated on independent
BF and SBM samples. Further enrolled samples were processed using next-generation sequencing and quantitative polymerase chain
reaction for assessment of monogenic disorders (PGT-M) or aneuploidy (PGT-A).
Main Outcome Measure(s): DNA amplification and concordance rates across BF, SBM, and TE to assess diagnostic efficiency.
Result(s): No differences were detected among the DNA extraction methods tested. In PGT-M tests, for BF and SBM, 2.9% and 20.8% of
all samples, respectively, produced a diagnosis concordant with the corresponding TE (n ¼ 2 of 69 and 15 of 72, respectively). The SBM
samples were associated with higher discordance rates and higher artifacts/contamination detection compared with BF. In multiple
occasions, the maternal mutated variant was detected in the SBM of homozygous wild-type embryos, showing evidence of maternal
DNA persistence in culture medium. In PGT-A tests, BF analysis showed high amplification failure rates (65.2%) and an overall
concordance rate of 37.5% among amplified samples.
Conclusion(s): Based on current methodologies, BF and SBM genetic analyses do not provide sufficiently reliable results to be
employed clinically. Until the risk of maternal contamination can be properly prevented, SBM should not be used for PGT-M
purposes. (Fertil SterilÒ 2018;110:870–9. Ó2018 by American Society for Reproductive Medicine.)
El resumen está disponible en Español al final del artículo.
Key Words: Blastocentesis, blastocoel, noninvasive, PGT, spent blastocyst media

Discuss: You can discuss this article with its authors and other readers at https://www.fertstertdialog.com/users/16110-fertility-
and-sterility/posts/33666-26013

I
n the last decade, the number of ploidies (PGT-A) has rapidly increased This trend is mainly due to the
preimplantation genetic testing worldwide, and PGT is now widely em- development and introduction into
(PGT) cases for monogenic condi- ployed as an adjunct diagnostic/testing clinical work of new molecular and cy-
tions (PGT-M) or for de novo aneu- tool in in vitro fertilization (IVF) cycles. togenetic methodologies able to produce
a comprehensive assessment of the
Received March 24, 2018; revised May 1, 2018; accepted May 28, 2018. whole embryonic chromosomal comple-
A.C. has nothing to disclose. V.R. has nothing to disclose. C.P. has nothing to disclose. M.P. has nothing
to disclose. L.G. has nothing to disclose. A.G. has nothing to disclose. M.S. has nothing to disclose. ment, resulting in improved clinical out-
D.C. has nothing to disclose. F.M.U. has nothing to disclose. L.R. has nothing to disclose. comes (1, 2). Additional improvements
Supported by iGenomix and Genera Centers for Reproductive Medicine.
Reprint requests: Antonio Capalbo, Ph.D., IGENOMIX, via Fermi 1, Marostica (VI), Italy (E-mail: in IVF laboratories, including extended
antcapalbo@gmail.com). embryo culture to the blastocyst stage
and enhanced cryopreservation
Fertility and Sterility® Vol. 110, No. 5, October 2018 0015-0282/$36.00
Copyright ©2018 American Society for Reproductive Medicine, Published by Elsevier Inc. protocols (3, 4), have favored the
https://doi.org/10.1016/j.fertnstert.2018.05.031

870 VOL. 110 NO. 5 / OCTOBER 2018


implementation of PGT methodologies in IVF treatments as
well.
Despite its extensive application, one of the biggest chal-
lenges of the PGT methodology is the requirement of biopsy
samples collected from an embryonic specimen, a procedure
that entails both technical and economic challenges. In this
regard, trophectoderm (TE) biopsy is considered generally
safe, and it is the technique with the least impact on the
embryo (5, 6). Nonetheless, the high degree of technical skill
required to obtain these samples and the costs associated
with both the equipment (i.e., laser) and operator training
remain a bottleneck for wider implementation of this
strategy. In fact, although TE biopsy has been proven safe
(5), suboptimal embryo biopsies may undermine embryo
developmental and reproductive competence (7).
Furthermore, although reassuring data have been obtained
so far (8), conclusive evidence about the potential long-term
effects of embryo biopsies has yet to be reported.
During the last few years, to reduce both laboratory work-
load and the potential detrimental effect of suboptimal bi-
opsies, IVF and PGT laboratories have been investigating
new sources of embryo-derived DNA. Indeed, the develop-
ment of a successful minimally or noninvasive methodology
would also facilitate the implementation of PGT strategies in
smaller IVF laboratories, where resources for completing an
adequately thorough training on embryo biopsy are limited.
Recently, blastocoel fluid (BF) was suggested as a source
of embryonic DNA for PGT purposes (9, 10). The fluid can be
aspirated from the embryo using an injection needle through
a procedure described by Poli et al. (11) as blastocentesis.
Using this technique, the sampling of embryo-derived com-
pounds is less invasive compared with biopsies, avoiding
the removal of trophectoderm cells from the developing
embryo. Besides, blastocyst collapse, whether through blasto-
centesis, laser pulse, or other means is often employed before
cryopreservation to maximize postwarming survival rates
(12). Finally, it was shown that the active aspiration of the
blastocoelic content does not impact embryo architecture,
leading to high survival rates for both good and poor
morphology embryos (13, 14), thus suggesting its
compatibility with conventional IVF laboratory workflow.
Nonetheless, blastocentesis entails some degree of
invasiveness to the embryo, although it is minimal. To
develop a completely noninvasive, inexpensive, and easily
implementable methodology for PGT purposes, several
investigators proposed the use of spent culture medium
collected at blastocyst stage (SBM) as a source of embryonic
DNA (15–17). In particular, Xu et al. (18) have shown high
ploidy concordance rate between SBM samples and their
corresponding embryos (>80%). This team achieved healthy
live births by employing noninvasive aneuploidy screening
in couples undergoing IVF for reciprocal translocation and
repeated pregnancy loss (18). If sufficiently accurate and
reproducible, both these methodologies open the doors to a
minimally or noninvasive approach to PGT of IVF embryos.
In this study, we assess the suitability of BF and SBM spec-
imens as PGT analysis templates. First, we evaluated the most
efficient way to treat BF and SBM samples for DNA extraction
before amplification. Then we determined the diagnostic rate
VOL. 110 NO. 5 / OCTOBER 2018
Fertility and Sterility®
of PGT-M when applied to TE, BF, and SBM samples, also
providing details on amplification failure and allele dropout
(ADO) rates of alleles of maternal and paternal origin. In this
context, genotyping analysis offered the unique possibility
to obtain important information about the nature and origin
of DNA amplified from BF and SBM samples in standard
clinical conditions. Finally, we assessed the diagnostic
reliability of BF DNA analysis for PGT-A application.
MATERIALS AND METHODS
Study Design
Institutional review board approval for this study was
obtained from the Clinica Valle Giulia Ethics Committee.
Patients undergoing IVF treatment at the G.EN.E.R.A. center
between April 2015 and December 2016 were approached to
take part in the study at the time they were starting ovarian
stimulation. After consultation about the methodologies
investigated, they signed the informed consent for the study.
Initially, we compared amplification failure rates between
two DNA extraction methods: EXTþAMP (DNA extraction
before amplification) and directAMP (loci amplification
without prior DNA extraction). In this experiment, 9 BF and
12 SBM samples were enrolled from four couples
(two consenting for BF donation and two for SBM samples)
and split into the two experimental arms, for a total of 18
BF (9 EXTþAMP and 9 directAMP) and 24 SBM (12
EXTþAMP and 12 directAMP) samples (Supplemental
Table 1, available online). These samples were used only to
set the best conditions for DNA extraction and amplification
and were not further considered.
A total of 267 samples were collected from 26 patients
and enrolled in two separate studies focused on [1] the
application of PGT-M protocols on TE, SBM, and BF
specimens (221 samples from 14 consenting couples who
were referred to the clinic as carriers of an inheritable genetic
condition); [2] the application of PGT-A protocols on TE and
BF specimens (46 samples, 23 TE and 23 BF from 12 patients
undergoing PGT-A cycles) (Fig. 1). The samples included in
the studies were obtained from patients undergoing PGT
(PGT-M or PGT-A) and consenting to the use of each specific
specimen (BF and/or SBM). Each consenting patient was
included in only one branch of the study (the clinical
data related to the patients enrolled are shown in
Supplemental Table 2, available online).
In the PGT-M branch of study 1 of the 221 samples,
80 were TE biopsies, 69 were BF samples, and 72 were SBM
samples (specifically, 61 triads BF-TE-SBM, 8 pairs BF-TE
and 11 pairs TE-SBM; Fig. 1). All 221 samples were
subsequently analyzed using PGT-M protocols specific for
the parental mutation in order to genotype the embryo (19).
Additionally, a subset of these samples (70 loci carrying
the pathologic mutation, 39 of paternal origin, and 31 of
maternal origin) were analyzed to assess the relationship
between test concordance and origin of the investigated
locus. Because SNPs are common genomic variations in the
population, their exclusion from the analysis rules out
the potential interference of exogenous/environmental
contamination. Instead, the exclusive analysis of the genetic
871
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
FIGURE 1
Study design and sample size. BF ¼ blastocoel fluid; PGT-A ¼
preimplantation genetic testing for aneuploidies; PGT-M ¼
preimplantation genetic testing for monogenic diseases; SBM ¼ spent
blastocyst media; TE ¼ trophectoderm.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.
mutations that are rare variants and specific for each trio
considered, allows the precise assessment of parental
contribution in fluidic samples.
In the PGT-A study branch 2, 46 samples were enrolled
from 12 patients (23 pairs BF-TE; Fig. 1) quantitative
polymerase chain reaction (qPCR)-based comprehensive
chromosomal testing was used as standard method to
process clinical TE biopsies. Instead, because of the lower
quantity and quality of the starting DNA template
obtainable from fluidic samples, BF specimens were
processed by whole-genome analysis/next-generation
sequencing (WGA-NGS) (Fig. 1). Nevertheless, the two
methods employed were previously reported to generate
consistent outcomes for whole chromosome aneuploidy
detection on five-cell preparations of euploid and aneuploid
fibroblast lines (20).
Outcome Measures
In PGT-M cases the amplification rates were calculated
based on the detection of the target loci sequences tested.
In PGT-M cases the results were deemed concordant across
specimens (BF/SBM/TE) if they showed the same allele
variant. Discordant results were considered artifacts if the
allele showed an unexpected genotype, or allele drop-in
(ADI) if they showed the alternative parental allele variant,
not present in the embryo. Finally, the presence of a single
allele in a sample expected to be heterozygous for that locus
was counted as ADO.
When the relationship between the efficiency in detecting
the correct locus variant and its parental inheritance was
investigated, all SNPs used for linkage analysis were removed
from the data set to minimize biases potentially introduced by
exogenous DNA contamination (including those cases where
872
diagnosis was not possible via direct mutation and linkage
analysis was used to produce a diagnosis). Also, to precisely
assess allelic origin (maternal or paternal), those cases where
the same mutation was present in both parents were removed,
and only triads derived from embryos presenting mutation
sites in heterozygosis or compound heterozygotes were
analyzed.
In PGT-A cases, standard parameters for either qPCR (21)
or VeriSeq NGS protocol (22) were used to evaluate DNA
amplification, depending on the technology employed. Only
patterns consistent with a uniform whole chromosome aneu-
ploidy were reported for this study, as segmental and mosai-
cism diagnoses are still controversial topics in clinical PGT-A
cycles.
Generation of IVF Embryos
Follicular stimulation was performed as described else-
where (23), and all metaphase II oocytes used in this study
were subjected to intracytoplasmic sperm injection (ICSI)
36 to 38 hours after human chorionic gonadotropin (hCG)
administration. Zygotes showing two pronuclei were
further individually cultured in 25 mL of culture medium
(Quinn's Advantage, Cleavage Medium, Cooper Surgical,
supplemented with 5% Quinn's Advantage human serum
albumin [HAS], or continuous single culture medium; Ir-
vine Scientific, Australia) covered by pre-equilibrated min-
eral oil and incubated in 6% CO2 and 5% O2 tension into
MINC bench-top incubators (Cook Medical). When sequen-
tial culture was used, the medium changeover was per-
formed on day 3 (Quinn's Advantage blastocyst medium
supplemented with 5% Quinn's Advantage HSA; Cooper
Surgical). Each blastocyst's morphology was graded imme-
diately before BF aspiration or TE biopsy as described else-
where (24). The TE biopsy and 24-chromosome aneuploidy
testing were performed using standard protocols described
elsewhere (21, 25).
BF and SBM Specimen Collection
Blastocoel. Fully expanded blastocysts were placed in a 10-
mL drop of HEPES buffered medium (either Cooper Surgical
or Irvine, both supplemented with 5% HSA) and overlaid
with warmed mineral oil (either Cooper Surgical or Irvine).
A conventional ICSI needle was gently pushed through the
zona pellucida and TE cell wall to reach the center of the cav-
ity. Light negative pressure was applied to allow the aspira-
tion of the fluid. Once the blastocoelic cavity reached
around 10% of its initial volume, the needle was retracted
from the embryo, and the retrieved specimen was released
in a 5-mL drop of the same medium plated next to the biopsy
drop used. The medium drop containing the BF was then
collected and placed in a 0.2-mL sterile PCR tube stored at
80 C to be processed as described in the genetic analysis
section. The TE biopsy was performed immediately after the
BF aspiration. Both these procedures were performed by se-
nior embryologists who had at least 2 years of experience
with TE biopsy.
Spent blastocyst media. Once the embryos reached the blas-
tocyst stage, they were moved to a biopsy/blastocentesis dish,
VOL. 110 NO. 5 / OCTOBER 2018

and the spent culture medium was collected immediately af-
ter, as described by Capalbo et al. (26). A remarkable strength
of this study is that the TE biopsy strategy used in this study
does not entail any kind of hatching or extra embryonic
manipulation before the biopsy stage, thus providing culture
media samples more representative of those obtained in stan-
dard IVF treatments. Blank culture media drops were collected
from the stock and after 3 and 5 days of culture as negative
controls.
Genetic Analysis
DNA extraction procedure (TE, BF, SBM). The TE biopsies
were processed by alkaline lysis. The DNA extraction from
BF and SBM samples was performed on the total volume of
retrieved sample, using QIAamp DNA Micro Kit (QIAGEN Ger-
many) according to the manufacturer's protocol.
PGT-M by qPCR (TE, BF, SBM). The embryonic origin of the
DNA sourced from BF and SBM, and the possibility of geno-
typing the embryo from these specimens was assessed using
protocols of qPCR analysis customized for the specific disor-
der investigated. Quantitative PCR is based on the use of Taq-
Man allelic discrimination assays (ThermoFisher Scientific)
for direct mutation and indirect linked marker PGT-M anal-
ysis as described elsewhere (19). Each protocol was previously
validated on individually isolated cells from each parent and
relatives.
PGT-A by qPCR (TE) and WGA-NGS VeriSeq protocol
(BF). The chromosomal content was assessed on TE bi-
opsies using a qPCR-based protocol consisting of 96
different TaqMan Copy Number Assays (four on each chro-
mosome) that was previously described in detail and exten-
sively validated for whole chromosome copy number
aneuploidies detection on TE samples (25, 27). Because
TaqMan copy number variation assays employed in the
qPCR-based PGT-A protocol do not work properly on sin-
gle cells or fluidic samples of low quality, DNA quantity BF
samples were first treated for WGA using the SurePlex
DNA Amplification System (Illumina), according to the
manufacturer's protocol. Amplified samples for NGS were
processed with VeriSeq PGS kit (Illumina), on a MiSeq de-
vice. Blinded computational prediction of uniform whole
chromosome aneuploidies was performed as recommended
by the supplier using the automatic aneuploidy detection
algorithm ‘‘default VeriSeq’’ by Bluefuse Multi software
(BlueFuse, Illumina, version 4.2, 20289). To assess the reli-
ability of BF samples for genotyping purposes, we
compared the results obtained from TEs (by qPCR) with
those from corresponding BFs (by WGA-NGS VeriSeq
protocol).
Statistical Analysis
Either two-tailed chi-square or two-tailed Fisher's exact tests
were employed to assess statistical significance. All confi-
dence intervals (CI) were calculated at 95%. Two logistic
regression analyses were performed to assess the correlation
between the outcomes obtained from BF or SBM samples
and both the day of preimplantation development and the
VOL. 110 NO. 5 / OCTOBER 2018
Fertility and Sterility®
blastocyst's morphologic quality. P< .05 was considered
statistically significant.
RESULTS
DNA Amplification Rates in Extracted and
Nonextracted BF and SBM Samples
PGT-M protocols were initially employed to assess the
difference in amplification rates of treated (subjected to
DNA extraction before PCR amplification – EXTþAMP) and
untreated (no DNA extraction; DirectAMP) BF and SBM
specimens. Differences in amplification failure rates between
treated and untreated BF and SBM were not statistically
significant (Supplemental Table 1). We therefore proceeded
without DNA extraction step for the following BF and SBM
samples.
PGT-M Amplification Rates for TE Biopsy and BF
and SBM Samples
In this set of experiments, we assessed the embryonic origin of
the DNA detected in specimens of TE, BF, and SBM. We also
estimated the diagnostic concordance across specimens
derived from the same embryo. To do this, we accurately
genotyped each specimen using a qPCR-based approach
(TaqMan Genotype Assay) targeted at the combined detection
of both mutation sites and multiple informative single-
nucleotide polymorphisms (SNPs) surrounding the mutation.
In total, 405 loci were tested in 80 TE samples, 347 loci in 69
BF samples, and 378 loci in 72 SBM samples (3–14 loci per
sample, average 5 loci per sample; Supplemental Table 3,
available online). Amplification failure occurred in 0, 252
(72.6%), and 39 loci (10.3%) for TE, BF, and SBM samples,
respectively (P< .001) (Table 1). In TE samples, 99.8% of the
tests produced concordant results, showing the presence of
the expected allele; a single ADO occurred in one sample.
Overall, loci amplification in SBM samples performed
statistically significantly better than in the BF samples
(P< .001) (Table 1). Additionally, among successfully
amplified loci, the SBM samples were more frequently
concordant with the control compared with the BF samples
(n ¼ 225 of 339, 66.4% vs. 46 of 95, 48.4%; P¼ .002). Among
the successfully amplified loci, ADO was more frequently
detected for alleles of paternal origin compared with the
ones of maternal origin in both BF (n ¼ 30 of 95, 31.6% vs.
n ¼ 14 of 95, 14.7%, respectively; P¼ .01) and SBM samples
(n ¼ 48 of 339, 14.2% vs. n ¼ 28 of 339, 8.3%, respectively;
P¼ .02). This finding is consistent with the lower levels of
paternally derived DNA in the BF and SBM samples.
Loci Amplification and Concordance Rates Based
on Maternal and Paternal Inheritance
Data from mutation loci analysis were also investigated in
regards of parental inheritance (Table 2). In this branch of
the study, to precisely assess the allelic origin only triads
where embryos showed mutation sites in heterozygosis were
analyzed, as explained in the Materials and Methods section.
In the selected cases, the mutation was present in only one of
the parents, and it was possible to perform direct mutation
873
ORIGINAL ARTICLE: ASSISTED REPRODUCTION

TABLE 1

Preimplantation genetic testing for monogenic diseases results in trophectoderm (TE), blastocoel fluid (BF), and spent blastocyst media (SBM)
samples.
TE BF SBM
Probes N % (95% CI) N % (95% CI) N % (95% CI)
All 405 347 378
Amplification failure 0/405 0.0 (0.0–9.1) 252/347 72.6 (67.6–77.3) 39/378 10.3 (7.4–13.8)
Concordant genotype 404/405 99.8 (98.6–99.9) 46/347 13.3 (9.9–17.3) 225/378 59.5 (54.4–64.5)
Discordant genotype 1/405 0.25 (0.01–1.4) 49/347 14.1 (10.6–18.2) 114/378 30.2 (25.6–35.1)
ADO 1/405 0.25 (0.01–1.4) 44/347 12.7 (9.4–16.7) 76/378 20.1 (16.2–24.5)
Materna 1/1 100.0 (2.5–100.0) 14/44 31.8 (19.9–46.6) 28/76 36.8 (26.1–48.7)
Paternal 0/1 0.0 (0.0–97.5) 30/44 68.2 (52.4–81.4) 48/76 63.2 (51.3–73.9)
Artifact 0/405 0.0 (0.0–9.1) 5/347 1.4 (0.5–3.3) 38/378 10.1 (7.2–13.5)
Mutation-specific only 76 65 69
Amplification failure 0/76 0.0 (0.0–4.7) 49/65 75.4 (63.1–85.2) 11/69 15.9 (8.2–26.7)
Concordant genotype 75/76 98.7 (92.9–100.0) 8/65 12.3 (5.5–22.8) 40/69 58.0 (45.5–69.8)
Discordant genotype 1/76 1.3 (0.03–7.1) 8/65 12.3 (5.5–22.8) 18/69 26.1 (16.3–38.1)
ADO 1/76 1.3 (0.03–7.1) 8/65 12.3 (5.5–22.8) 12/69 17.4 (9.3–28.4)
Maternal 1/1 100.0 (2.5–100.0) 0/8 0.0 (0.0–36.9) 5/12 41.7 (15.2–72.3)
Paternal 0/1 0.0 (0.0–97.5) 8/8 100.0 (63.1–100.0) 7/12 58.3 (27.7–84.8)
Artifact 0/76 0.0 (0.0–4.7) 0/65 0.0 (0.0–5.5) 6/69 8.7 (3.3–18.0)
Note: The results are shown for all the probes analyzed and only for mutation-specific ones. ADO ¼ allele dropout; CI ¼ confidence interval.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

detection (no linkage analysis). Examples of qPCR diagnostic
plots are shown in Supplemental Figure 1 (available online).
Identification of the mutated genes of paternal origin in BF
and SBM samples. In this selected cohort, BF and SBM were
characterized by high paternal ADO rate (Table 2). In the BF
samples, eight out of the nine successfully amplified loci
(amplification failure: n ¼ 10 of 19, 52.6%) (Table 2) produced
discordant genotypes, all due to paternal ADO (89%; 95% CI,
51.8–99.7) (Fig. 2A). Of the 14 successfully amplified SBM
samples (amplification failure: n ¼ 2 of 16, 12.5%)
(Table 2), 7 (50.0%) produced discordant genotypes due to
ADO, 6 of which involved paternal alleles (85.7% of the
ADO events and 43% of successfully amplified loci; 95%CI,
17.7–71.1) (Fig. 2A). No mutation was detected in either BF
or SBM samples of the wild-type embryos. This is consistent
with the absence of paternal genome contamination in the
blastocyst fluidic samples.
Identification of mutated genes of maternal origin in BF and
SBM samples. The analysis of maternally inherited
mutations showed the presence of extra-embryonic DNA

TABLE 2

Preimplantation genetic testing for monogenic disease results in trophectoderm (TE), blastocoel fluid (BF), and spent blastocyst media (SBM)
samples divided between paternal- and maternal-specific probes.
TE BF SBM
Parent-specific mutation probes n % (95% CI) n % (95% CI) n % (95% CI)
Paternal
Total 39 35 32
Heterozygous (informative) 20 19 16
Amplification failure 0/20 0.0 (0.0–16.8) 10/19 52.6 (28.9–75.5) 2/16 12.5 (1.6–38.4)
Concordant genotype 20/20 100.0 (83.2–100.0) 1/19 5.3 (0.1–26.0) 7/16 43.8 (19.8–70.1)
Disconcordant genotype 0/20 0.0 (0.0–16.8) 8/19 42.1 (20.3–66.5) 7/16 43.8 (19.8–70.1)
ADO 0/20 0.0 (0.0–16.8) 8/19 42.1 (20.3–66.5) 7/16 43.8 (19.8–70.1)
Maternal 0/0 0.0 0/8 0.0 (0.0–36.9) 1/7 14.3 (0.36–57.9)
Paternal 0/0 0.0 8/8 100.0 (63.1–100.0) 6/7 85.7 (42.1–99.6)
Artifact 0/20 0.0 (0.0–16.8) 0/19 0.0 (0.0–17.6) 0/0 0.0
Maternal
Total 31 27 31
Heterozygous (informative) 19 18 19
Amplification failure 0/19 0.0 (0.0–17.7) 16/18 88.9 (65.3–98.6) 3/19 15.8 (3.4–39.6)
Concordant genotype 18/19 94.7 (74.0–99.9) 2/18 11.1 (1.4–34.7) 11/19 57.9 (33.5–79.8)
Disconcordant genotype 1/19 5.3 (0.1–26.0) 0/18 0.0 (0.0–18.5) 5/19 26.3 (9.2–51.2)
ADO 1/19 5.3 (0.1–26.0) 0/18 0.0 (0.0–18.5) 5/19 26.3 (9.2–51.2)
Maternal 1/1 100.0 (2.5–100) 0/0 0.0 (0.0–0.0) 4/5 80.0 (28.4–99.5)
Paternal 0/1 0.0 (0.0–97.5) 0/0 0.0 (0.0–0.0) 1/5 20. (0.5–71.6)
Artifact 0/19 0.0 (0.0–17.7) 0/18 0.0 (0.0–18.5) 0/19 0.0 (0.0–17.7)
Note: ADO ¼ allele dropout; CI ¼ confidence interval.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

874 VOL. 110 NO. 5 / OCTOBER 2018

 Fertility and Sterility®

FIGURE 2

Summary of preimplantation genetic testing for monogenic disorders (PGT-M) results is indicative of contamination derived from maternal DNA in
spent blastocyst medium (SBM) and/or blastocoel fluid (BF). (A) Excess of maternal wild type (WT) DNA causes paternal allele dropout (ADO) in BF
and SBM in cases of heterozygosis for paternal mutation in the trophectoderm (TE). (B) Maternal WT DNA causes ADO of mutated allele in the SBM
in cases of heterozygosis for maternal mutation in the TE. (C) Maternal DNA carrying the mutation detected in the SBM in cases of homozygosis of
maternal WT in the TE. The detection rates of these events are reported below each section of the picture. ve, negative; þve, positive.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

contamination in BF samples and of maternal DNA
contamination in the SBM samples. In the SBM samples
collected from heterozygote embryos (amplification failure:
n ¼ 3 of 19, 15.8%) (Table 2), 5 of the 16 successfully
amplified loci produced discordant genotypes due to ADO
(31.3%, 95% CI, 11.0–58.7), 4 of which involving the
maternal allele (80% of the ADO events, and 25.0% of the
successfully amplified loci; 95% CI, 7.3–52.4). The mutated
alleles were potentially masked by the excess of the wild-
type variant deriving from the degenerating first polar body
or by extra embryonic DNA (Fig. 2B). Importantly, in four
cases where the embryo showed homozygosis for the wild-
type allele, the mutated allele was detected in the SBM sample
(n ¼ 4 of 12, 33% of the cases; 95% CI, 9.9–65.1) showing
direct evidence of maternal DNA contamination (Fig. 2C).
Results from combined maternal and paternal mutation
loci data sets confirmed the higher amplification rates in
SBM compared with BF samples (n ¼ 30 of 35, 85.7% vs.
n ¼ 11 of 37, 29.7%; P< .001) (Table 2). However, in this
subset of targets there was no difference in concordance
rate between the SBM and BF results when the TE results
were used as the standard (n ¼ 18 of 30, 60.0% vs. n ¼ 3 of
11, 27.3%; P¼ not statistically significant) (Table 2).
Additionally, among successfully amplified loci in both BF
and SBM, ADO occurred more frequently for alleles of
paternal origin compared with maternal ones (n ¼ 14 of 23,
60.9% vs. n ¼ 4 of 18, 22.2%; P¼ .03) (Table 2), suggesting
a higher amount of maternal genomic DNA compared to
paternal genomic DNA.
PGT-M Diagnostic Rates for TE Biopsy, BF and SBM
Samples
Monogenic disorders diagnosis was performed using both
probes for the mutation site and for informative SNP sites
flanking the mutation site (Supplemental Table 3). Overall,
it was possible to generate consistent embryonic haplotypes
in 100% of the TE samples (Supplemental Table 4, available
online). In the BF and SBM groups, respectively, only 2.9%
and 20.8% of samples allowed whole haplotype generation
that was fully concordant with the corresponding TE sample
(P¼ .001; Supplemental Table 4). In the BF experiments, no
correlation was observed between either the DNA
amplification rate or diagnostic concordance and the day
of embryo biopsy (day 5, 6, or 7). Conversely, the blastocyst's
morphologic quality showed a statistically significant
correlation with DNA amplification (odds ratio [OR] 0.734;
P¼ .027). Among the samples that amplified, a statistically
significant correlation was observed between the
blastocyst's morphologic quality and the diagnostic
concordance with the related TE sample (OR 0.705;
P¼ .048) (Supplemental Fig. 2, available online).
In the SBM experiments, among the samples that ampli-
fied, the day of preimplantation development was highly

VOL. 110 NO. 5 / OCTOBER 2018 875

ORIGINAL ARTICLE: ASSISTED REPRODUCTION
correlated with both DNA amplification and diagnostic
concordance (OR 3.224 and 2.064, respectively; P¼ .03 and
.02, respectively). By contrast, the blastocyst's morphologic
quality did not show any statistically significant correlation
(Supplemental Fig. 3, available online).
PGT-A Diagnostic Rates in TE Biopsy and BF
Samples
Of the 23 BF samples submitted to chromosomal analysis by
WGA-NGS (VeriSeq; Illumina), 15 (65.2%) failed to amplify,
and the other eight produced a signal able to generate a full
karyotype (Supplemental Table 5, available online). Of these
eight specimens, five gave a result discordant with the
corresponding TE sample (62.5%, of the successfully
amplified specimens and 21.7% of all BF samples analyzed).
The remaining three BF samples produced a karyotype
concordant with the control (37.5%; 95% CI, 8.5–75.5%;
corresponding to 13.0% of all BF samples).
Supplemental Table 5 also reports the data outlining the
quality of both the TE biopsy qPCR-based analyses and the
related BF NGS-based analyses. Examples of discordant and
concordant diagnoses and amplification failure plots are
shown in Supplemental Figure 4 (available online). Calculated
on chromosomes pairs identification, the sensitivity and spec-
ificity of PGT-A on BF in this study are 66.7% (n ¼ 4 of 6) and
94.4% (n ¼ 169 of 179), respectively (Supplemental Table 5).
Additionally, a statistically nonsignificant trend toward a
lower amplification failure but higher discordance rates was
reported as the time required for the blastocyst to reach full
expansion increased and as morphologic quality decreased
(Supplemental Fig. 5, available online).
DISCUSSION
In the first part of this study, the embryos’ genotypes were
generated during PGT-M cycles and compared across the
different specimens. Although inferior compared with the
TE samples, the SBM samples performed statistically signifi-
cantly better than the BF samples in terms of diagnostic
rate. Despite its diagnostic potential, SBM showed high
detection of artifacts or ADI, which occurred in 10.1% of all
loci investigated. This notable rate of detectable nonem-
bryonic DNA material has several potential origins including
the presence of exogenous DNA in the culture media.
In fact, small amounts of DNA were reported in different
brands of fresh, unused media (15, 28). These contaminants
are most probably introduced through the supplementation
of nonpure DNA-binding proteins to the media (i.e., albumin).
Although this type of DNA contamination is negligible and
possibly harmless in conventional IVF treatments, it can be
easily detected when PCR is employed on culture media
samples (15).
It is interesting that the ADO rates were statistically
significantly higher for paternal alleles than maternal ones
in both BF and SBM samples. This preliminary observation
suggested an imbalance in DNA representation in favor of
maternal DNA over paternal DNA. It is thus probable that ge-
netic material from the cumulus complex or polar bodies is
still present in the culture system and that it is collected and
876
analyzed together with embryonic DNA. In the SBM samples,
this hypothesis was further supported by the detection of the
mutated allele of maternal origin where the corresponding TE
showed homozygosity for the wild-type allele. These data
provided compelling evidence about the substantial presence
of maternal DNA in SBM samples.
Recent studies have suggested the clinical use of SBM for
aneuploidy testing (16, 18) or PGT-M (29), but our results
highlight the needs for further investigations before this diag-
nostic approach can be considered in clinical settings. Several
strategies could be adopted to increase the fraction of embry-
onic DNA and prevent nonembryonic DNA carryover. Among
these, we propose pretreatment depletion of DNA in culture
media, complete oocyte denudation, and standardization of
timings (related to blastocyst expansion) for BF and SBM
collection. Amplification of embryo-specific DNA could be
performed using a customized version of the second-
generation noninvasive fetal DNA analysis proposed by
Chan et al. (30). Isolation and targeted analysis of exosomes
released by the blastocyst in the culture media might repre-
sent an alternative effective way to selectively obtain embry-
onic DNA for genetic testing (31).
Our PGT-A results showed that only a portion of BF sam-
ples (34.8%) could generate a signal to be used for embryo
karyotyping. In previous large studies on the use of BF for
PGT purposes (32–34), BFs were collected after the embryos
had been cryopreserved or biopsied (experimental
conditions). By contrast, in our study blastocoel sample
collection was performed on fresh, untreated blastocysts,
thus simulating a real PGT cycle treatment (clinical
conditions). The outcomes under experimental conditions
were higher amplification rates [i.e., 62% (32), 82% (33) and
96% (34)] compared with those we obtained under clinical
conditions. It is possible that the higher DNA amplification
rates observed in those studies were due to larger DNA
availability in the cavity caused by micromanipulation- or
cryopreservation-induced cell lysis.
In our data set, only a fraction (37.5%) of the amplified
samples led to results concordant with those generated by
TE biopsies, in agreement with published data by Tobler's
and Werner's groups (53% and 72% of amplified samples
matched the original embryo diagnosis, respectively). Because
qPCR and VeriSeq PGT-A protocols provided concordant re-
sults for uniform aneuploidies when tested on cell lines sam-
ples (20), only a minority of discordant cases observed in our
study may be attributed to the use of different technologies to
conduct aneuploidy assessment.
The biological bases of such discordances are extremely
difficult to define in the absence of functional studies on
the biological mechanisms of embryonic DNA release in the
extracellular environment. Membrane-encapsulated DNA
can derive from DNA-containing fragments originating
from cells undergoing apoptosis, or defective (or corrective)
chromosome segregation processes during cell division or se-
lective degeneration of abnormal cells in mosaic diploid/
aneuploid embryos (32). This type of nonrepresentative
DNA can provide serious contamination to the analytical
sample, critically jeopardizing the diagnostic accuracy. How-
ever, at present, convincing evidence about the existence
VOL. 110 NO. 5 / OCTOBER 2018

aneuploidy corrective mechanisms in human preimplantation
embryos is still lacking. Intriguingly, a uniform release of
DNA from all embryonic cells might result in fluidic samples
being even more reliable than TE biopsies for predicting the
genetic complement of the blastocyst.
In an extensive investigation of noninvasive chromo-
some screening (NICS), Xu et al. (18) reported that the test
was indeed validated under experimental conditions by
comparing ploidy results obtained from culture media and
vitrified/warmed embryos. The diagnostic consistency at the
embryo level was observed in 38 of the 42 pairs of samples,
but a complete karyotype concordance was achieved in
only 62% of the samples. Moreover, the investigators did
not report any genotyping data, thus failing to assess the
presence and origin of nonembryonic DNA contaminants in
the samples. Furthermore, the six live births achieved by
seven couples undergoing NICS-based embryo selection did
not represent enough evidence to support the clinical effec-
tiveness of NICS intervention, especially considering the
lack of a control arm in the study.
CONCLUSION
Based on the standard embryologic and genetic methodolo-
gies adopted in this study, neither BF nor SBM genetic anal-
ysis offers consistent and sufficiently reliable diagnostic rates
to justify their use in clinical PGT treatments. Due to the high
risk of maternal contamination and subsequent misdiagnosis
directly observed here, SBM should not be used as specimens
for the detection of single-gene mutations until these risk fac-
tors are properly assessed and prevented. Nonetheless, future
efforts to develop novel strategies for the enrichment of the
embryonic DNA fraction in SBM samples are needed; if
they become effective, they will be milestone achievements
in the field of preimplantation genetics.
REFERENCES
1. Dahdouh EM, Balayla J, Garcia-Velasco JA. Comprehensive chromosome
screening improves embryo selection: a meta-analysis. Fertil Steril 2015;
104:1503–12.
2. Rubio C, Bellver J, Rodrigo L, Castillon G, Guillen A, Vidal C, et al. In vitro
fertilization with preimplantation genetic diagnosis for aneuploidies in
advanced maternal age: a randomized, controlled study. Fertil Steril 2017;
107:1122–9.
3. Glujovsky D, Blake D, Farquhar C, Bardach A. Cleavage stage versus blasto-
cyst stage embryo transfer in assisted reproductive technology. Cochrane
Database Syst Rev 2012;7:CD002118.
4. Rienzi L, Gracia C, Maggiulli R, LaBarbera AR, Kaser DJ, Ubaldi FM, et al.
Oocyte, embryo and blastocyst cryopreservation in ART: systematic review
and meta-analysis comparing slow-freezing versus vitrification to produce
evidence for the development of global guidance. Hum Reprod Update
2017;23:139–55.
5. Scott RT, Upham KM, Forman EJ, Zhao T, Treff NR. Cleavage-stage biopsy
significantly impairs human embryonic implantation potential while blasto-
cyst biopsy does not: a randomized and paired clinical trial. Fertil Steril 2013;
100:624–30.
6. Cimadomo D, Capalbo A, Ubaldi FM, Scarica C, Palagiano A, Canipari R,
et al. The impact of biopsy on human embryo developmental potential dur-
ing preimplantation genetic diagnosis. Biomed Res Int 2016;2016:7193075.
7. Neal SA, Franasiak JM, Forman EJ, Werner MD, Morin SJ, Tao X, et al. High
relative deoxyribonucleic acid content of trophectoderm biopsy adversely af-
fects pregnancy outcomes. Fertil Steril 2017;107:731–6.e1.
VOL. 110 NO. 5 / OCTOBER 2018
Fertility and Sterility®
8. Sunkara SK, Antonisamy B, Selliah HY, Kamath MS. Pre-term birth and low birth
weight following preimplantation genetic diagnosis: analysis of 88 010 singleton
live births following PGD and IVF cycles. Hum Reprod 2017;32:432–8.
9. Poli M, Jaroudi S, Sarasa J, Spath K, Child T, Wells D. The blastocoel fluid as a
source of DNA for preimplantation genetic diagnosis and screening. Fertil
Steril 2013;100(Suppl):S37.
10. Gianaroli L, Magli MC, Pomante A, Crivello AM, Cafueri G, Valerio M, et al.
Blastocentesis: a source of DNA for preimplantation genetic testing: results
from a pilot study. Fertil Steril 2014;102:1692–9.e6.
11. Poli M, Ori A, Turner K, Child T, Wells D. Defining the biochemical content of
the human blastocoel using mass spectrometry: a novel tool for identifying
biomarkers of embryo competence. Fertil Steril 2012;98(Suppl):S110.
12. Kader A a, Choi A, Orief Y, Agarwal A. Factors affecting the outcome of hu-
man blastocyst vitrification. Reprod Biol Endocrinol 2009;7:99.
13. Chen SU, Lee TH, Lien YR, Tsai YY, Chang LJ, Yang YS. Microsuction of blas-
tocoelic fluid before vitrification increased survival and pregnancy of mouse
expanded blastocysts, but pretreatment with the cytoskeletal stabilizer did
not increase blastocyst survival. Fertil Steril 2005;84:1156–62.
14. Poli M, Ori A, Child T, Jaroudi S, Spath K, Beck M, et al. Characterization and
quantification of proteins secreted by single human embryos prior to im-
plantation. EMBO Mol Med 2015;7:1465–79.
15. Hammond ER, Mcgillivray BC, Wicker SM, Peek JC, Shelling AN, Stone P,
et al. Characterizing nuclear and mitochondrial DNA in spent embryo culture
media: genetic contamination identified. Fertil Steril 2017;107:220–8.e5.
16. Shamonki MI, Jin H, Haimowitz Z, Liu L. Proof of concept: preimplantation
genetic screening without embryo biopsy through analysis of cell-free
DNA in spent embryo culture media. Fertil Steril 2017;106:1312–8.
17. Feichtinger M, Vaccari E, Carli L, Wallner E, Madel U, Figl K, et al. Non-inva-
sive preimplantation genetic screening using array comparative genomic hy-
bridization on spent culture media: a proof-of-concept pilot study. Reprod
Biomed Online 2017;34:583–9.
18. Xu J, Fang R, Chen L, Chen D, Xiao J-P, Yang W, et al. Noninvasive chromo-
some screening of human embryos by genome sequencing of embryo cul-
ture medium for in vitro fertilization. Proc Natl Acad Sci USA 2016;113:
11907–12.
19. Zimmerman RS, Jalas C, Tao X, Fedick AM, Kim JG, Pepe RJ, et al. Devel-
opment and validation of concurrent preimplantation genetic diagnosis
for single gene disorders and comprehensive chromosomal aneuploidy
screening without whole genome amplification. Fertil Steril 2016;105:
286–94.
20. Goodrich D, Tao X, Bohrer C, Lonczak A, Xing T, Zimmerman R, et al. A ran-
domized and blinded comparison of qPCR and NGS-based detection of
aneuploidy in a cell line mixture model of blastocyst biopsy mosaicism. J
Assist Reprod Genet 2016;33:1473–80.
21. Capalbo A, Ubaldi FM, Cimadomo D, Maggiulli R, Patassini C, Dusi L, et al.
Consistent and reproducible outcomes of blastocyst biopsy and aneuploidy
screening across different biopsy practitioners: a multicentre study involving
2586 embryo biopsies. Hum Reprod 2016;31:199–208.
22. Vera-Rodriguez M, Michel C-E, Mercader A, Bladon AJ, Rodrigo L,
Kokocinski F, et al. Distribution patterns of segmental aneuploidies in hu-
man blastocysts identified by next-generation sequencing. Fertil Steril
2016;105:1047–55.e2.
23. Ubaldi FM, Capalbo A, Colamaria S, Ferrero S, Maggiulli R, Vajta G, et al.
Reduction of multiple pregnancies in the advanced maternal age population
after implementation of an elective single embryo transfer policy coupled
with enhanced embryo selection: pre- and post-intervention study. Hum Re-
prod 2015;30:2097–106.
24. Capalbo A, Rienzi L, Cimadomo D, Maggiulli R, Elliott T, Wright G, et al. Cor-
relation between standard blastocyst morphology, euploidy and implanta-
tion: an observational study in two centers involving 956 screened
blastocysts. Hum Reprod 2014;29:1173–81.
25. Treff NR, Tao X, Ferry KM, Su J, Taylor D, Scott RT Jr. Development and vali-
dation of an accurate quantitative real-time polymerase chain reaction-
based assay for human blastocyst comprehensive chromosomal aneuploidy
screening. Fertil Steril 2012;97:819–24.
26. Capalbo A, Ubaldi FM, Cimadomo D, Noli L, Khalaf Y, Farcomeni A, et al.
MicroRNAs in spent blastocyst culture medium are derived from
877
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
trophectoderm cells and can be explored for human embryo reproductive
competence assessment. Fertil Steril 2016;105:225–35.e3.
27. Capalbo A, Treff NR, Cimadomo D, Tao X, Upham K, Ubaldi FM, et al. Com-
parison of array comparative genomic hybridization and quantitative real-
time PCR-based aneuploidy screening of blastocyst biopsies. Eur J Hum
Genet 2015;23:901–6.
28. Vera-Rodriguez M, Diez-Juan A, Jimenez-Almazan J, Martinez S, Navarro R,
Peinado V, et al. Origin and composition of cell-free DNA in spent medium
from human embryo culture during preimplantation development. Hum Re-
prod 2018;33:745–56.
29. Wu H, Ding C, Shen X, Wang J, Li R, Cai B, et al. Medium-based noninvasive
preimplantation genetic diagnosis for human a-thalassemias-SEA. Medicine
(Baltimore) 2015;94:e669.
30. Chan KCA, Jiang P, Sun K, Cheng YKY, Tong YK, Cheng SH, et al. Second
generation noninvasive fetal genome analysis reveals de novo mutations,
single-base parental inheritance, and preferred DNA ends. Proc Natl Acad
Sci USA 2016;113:E8159–68.
878
31. Giacomini E, Vago R, Sanchez AM, Podini P, Zarovni N, Murdica V,
et al. Secretome of in vitro cultured human embryos contains extracel-
lular vesicles that are uptaken by the maternal side. Sci Rep 2017;7:
5210.
32. Tobler KJ, Zhao Y, Ross R, Benner AT, Xu X, Du L, et al. Blastocoel fluid
from differentiated blastocysts harbors embryonic genomic material
capable of a whole-genome deoxyribonucleic acid amplification and
comprehensive chromosome microarray analysis. Fertil Steril 2015;104:
418–25.
33. Magli MC, Pomante A, Cafueri G, Valerio M, Crippa A, Ferraretti AP, et al.
Preimplantation genetic testing: polar bodies, blastomeres, trophectoderm
cells, or blastocoelic fluid? Fertil Steril 2015;105:676–83.
34. Werner MD, Scott K, Bohrer C, Gabriele D, Tao X, Hong KH, et al.
The accuracy of blastocoel fluid comprehensive chromosomal
screening (CCS) is dependent on amplification yield and sequencing
depth when using nextgen sequencing. Fertil Steril 2017;102(Suppl):
e308–9.
VOL. 110 NO. 5 / OCTOBER 2018
 Fertility and Sterility®

Eficacia diagn
ostica del fluido del blastocele y del medio usado como fuentes de ADN para el examen gen

etico preimplantacional en
condiciones clínicas est
andar
Objetivo: Determinar si el fluido del blastocele (BF) o el medio usado en el blastocisto (SBM) son modelos adecuados para la evaluaci

on
del genotipo y/o cariotipo de embriones generados por fecundaci
on in vitro.
~o: Estudio prospectivo ciego.
Disen
Entorno: Laboratorio de gen
etica.
Paciente(s): 103 trofoectodermos (TE), 92 muestras BF y 72 muestras SBM de 26 pacientes sometidas a tratamientos de examen gen-

etico preimplantacional (PGT).
Intervenci
on(es): El BF y el SBM se recuperaron en el momento de la biopsia de TE. Se evaluaron dos estrategias para la extracci

on de
ADN en muestras independientes de BF y SBM. Las muestras adicionales incluidas se procesaron usando secuenciaci
on de nueva gen-
eraci

on y cuantificaci
on de la reacci

on en cadena de la polimerasa para la evaluaci
on de los trastornos monog
enicos (PGT-M) o
aneuploidías (PGT-A).
Principales medidas de los resultados: Tasas de amplificaci

on del ADN y concordancia en BF, SBM y TE para evaluar la eficacia del
diagn

ostico.
Resultados: No se detectaron diferencias entre los m
etodos de extracci
on de ADN evaluados. En las pruebas PGT-M para BF y SBM,
2.9% y 20.8% de todas las muestras, respectivamente, produjeron un diagn
ostico concordante con el TE correspondiente (n¼2 de 69 y
15 de 72, respectivamente). Las muestras de SBM se asociaron con mayores tasas de discordancia y m
as artefactos/contaminaciones
comparados con BF. En muchas ocasiones, la variante materna mutada se detect
o en el SBM de embriones homocig
oticos de tipo sil-
vestre, mostrando evidencia de la persistencia del ADN materno en el medio de cultivo. En las pruebas PGT-A, el an
alisis del BF mostr
o
altas tasas de fallo en la amplificaci

on (65.2%) y un total de tasa de concordancia del 37.5% entre las muestras amplificadas.
Conclusi
on(es): Con base en las metodologías actuales, los an
alisis gen
eticos BF y SBM no proporcionan resultados suficientemente
fiables para ser empleados clínicamente. Hasta que no se pueda prevenir adecuadamente el riesgo de contaminaci
on materna, el SBM no
se debe utilizar con fines PGT-M.

VOL. 110 NO. 5 / OCTOBER 2018 879

ORIGINAL ARTICLE: ASSISTED REPRODUCTION

SUPPLEMENTAL FIGURE 1

Examples of quantitative polymerase chain reaction (qPCR) plots for parental, trophectoderm (TE), blastocoel fluid (BF), and spent blastocyst
medium (SBM) specimens and allele ratio cluster diagrams for the three sample types. (A, B) Single-nucleotide polymorphism (SNP) genotyping
with probe for HBB Cod30 (G>C). (A) No detection of paternal mutation in SBM of heterozygous embryo indicates paternal allele dropout
(ADO) or contamination by extraembryonic/maternal wild-type DNA. (B) Allelic discrimination plot. (C, D) SNP genotyping with probe for HBB
IVSI-110 (G>A). (C) Detection of maternal mutation in SBM of homozygous embryo indicates maternal mutated DNA contamination. (D)
Allelic discrimination plot.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

879.e1 VOL. 110 NO. 5 / OCTOBER 2018

 Fertility and Sterility®

SUPPLEMENTAL FIGURE 2

Preimplantation genetic testing for monogenic disorders (PGT-M) results in blastocoel fluid (BF) per day of preimplantation development when the
sample was collected and per blastocyst's morphological quality. (A) Distribution of the probes analyzed according to the day of preimplantation
development when the samples were collected. (B) Distribution of the probes analyzed according to blastocyst's morphologic quality. (C) Probes'
amplification rate according to the day of preimplantation development when the samples were collected. (D) Concordant diagnostic rate
according to the day of preimplantation development when the samples were collected. (E) Probes' amplification rate according to blastocyst's
morphologic quality. (F) Concordant diagnostic rate according to blastocyst's morphologic quality. (G) Logistic regression analysis investigating
the correlation between positive DNA amplification from BF samples and related blastocysts' day of preimplantation development and
morphological quality. (H) Logistic regression analysis investigating the correlation, among samples showing a positive DNA amplification,
between a concordant diagnosis in the BF with its paired trophectoderm (TE) and related blastocysts' day of preimplantation development and
morphological quality. P<.05 was considered significant. ADO ¼ allele dropout; OR ¼ odds ratio.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

VOL. 110 NO. 5 / OCTOBER 2018 879.e2

ORIGINAL ARTICLE: ASSISTED REPRODUCTION

SUPPLEMENTAL FIGURE 3

PGT-M results in spent blastocyst media (SBM) per day of preimplantation development when the sample was collected and per blastocyst's
morphologic quality. (A) Distribution of the probes analyzed according to the day of preimplantation development when the samples were
collected. (B) Distribution of the probes analyzed according to blastocyst's morphologic quality. (C) Probes' amplification rate according to the
day of preimplantation development when the samples were collected. (D) Concordant diagnostic rate according to the day of preimplantation
development when the samples were collected. (E) Probes' amplification rate according to blastocyst's morphological quality. (F) Concordant
diagnostic rate according to blastocyst's morphological quality. (G) Logistic regression analysis investigating the correlation between positive
DNA amplification from SBM samples and related blastocysts' day of reimplantation development and morphologic quality. (H) Logistic
regression analysis investigating the correlation, among samples showing a positive DNA amplification, between a concordant diagnosis in the
SBM with its paired trophectoderm (TE) and related blastocysts' day of preimplantation development and morphological quality. P<.05 was
considered statistically significant. ADO ¼ allele dropout; OR ¼ odds ratio; PGT-M ¼ preimplantation genetic testing for aneuploidies.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

879.e3 VOL. 110 NO. 5 / OCTOBER 2018

 Fertility and Sterility®

SUPPLEMENTAL FIGURE 4

Examples of qPCR results from trophectoderm biopsies and NGS results from corresponding blastocoel fluids: (A) aneuploid diagnosis, (B) euploid
diagnosis, and (C) amplification failure from blastocoel fluid. Plots on the left represent qPCR results obtained from trophectoderm (TE) biopsies.
Each chromosome-specific copy number (CN) is represented as a black bar between the chromosome-specific thresholds (light green bars) for
trisomy, monosomy and nullisomy from top to bottom respectively. The plots on the right represent NGS results obtained from corresponding
blastocoel fluid (BF). The continuous red bar identifies the monosomy threshold, while the green bar represents the trisomy threshold. Embryo
karyotype diagnoses are reported below the graphs.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

VOL. 110 NO. 5 / OCTOBER 2018 879.e4

ORIGINAL ARTICLE: ASSISTED REPRODUCTION

SUPPLEMENTAL FIGURE 5

PGT-A results in blastocoel fluid (BF) (A) per day of preimplantation development when the sample was collected, and (B) per blastocyst's
morphologic quality.
Capalbo. Noninvasive embryo genetic testing. Fertil Steril 2018.

879.e5 VOL. 110 NO. 5 / OCTOBER 2018