On the possibility of selectively transferring embryos, by preimplantation genetic diagnosis (PGD/PGS) determined to be chromosomally abnormal


On the possibility of selectively transferring embryos, by preimplantation genetic diagnosis (PGD/PGS) determined to be chromosomally abnormal


If embryos are investigated for chromosomal abnormalities, CHR and other IVF centers transfer currently only embryos confirmed to be chromosomally normal. We in this OPINION, however, discuss the rationale why, under carefully considered exceptional circumstances, embryos, by preimplantation genetic diagnosis (PGD, also called preimplantation genetic screening, PGS) determined to be chromosomally abnormal, may still be transferrable. Since the risk of all embryos being chromosomally abnormal (aneuploid) increases with declining embryo numbers, such circumstances may mostly present in women with low functional ovarian reserve (LFOR) who routinely produce only small egg/embryo numbers. Whether embryos are correctly diagnosed as aneuploid, in such patients, therefore, is of particular importance because any “false-positive” diagnosis of an embryo may be the difference between having or not having an embryo transfer and between having or not having a pregnancy chance. Women with large embryo numbers usually do not face this problem since they still have many other normal embryos left for transfer even if they lose an embryo because of a “false-positive” diagnosis.

CHR Explains

As previously discussed in OPINION 002, CHR usually does not support the concept of PGD/PGS to improve IVF outcomes in women with LFOR, but once a patient has undergone a full IVF cycle with PGD/PGS and faces the option of having no chromosomally normal embryos available for transfer, the question arises whether the PGD/PGS diagnosis of aneuploidy in all tested embryos was correct. As we discuss in detail in this OPINION, new evidence suggests that PGD/PGS, whether performed at cleavage (day-3) or blastocyst stage (days 5/6), results in a significant number of “false-positive” diagnoses of aneuploidy. We in this OPINION argue that in so-affected patients it is ethical to consider transfer of embryos, which by PGD/PGS were reported as aneuploid, as long as reported aneuploidies are known to be lethal. Potentially non-lethal aneuploidies should never be transferred. Moreover, patients have to receive careful counseling prior to giving informed consent to such transfers about the small but relevant risk of establishing an ongoing aneuploid pregnancy, which may require medical termination.

Practically all IVF centers, CHR included, currently maintain a policy of transferring embryos after genetic testing (PGD/PGS) only if testing reveals the embryos to be chromosomally normal. Such testing has in principle three potential outcomes: An embryo can be normal, can be unequivocally abnormal or cannot be determined with certainty in its normality. Generally accepted standards of care currently require that only normal embryos be transferred and unequivocally abnormal as well as undeterminable embryos not be transferred.

As the recent literature suggests, a diagnosis of “abnormal” is not always correct.1-3 In fact, recent research suggests that current PGD methods for testing for chromosomal abnormalities may not be as reliable as once thought. These tests, while highly accurate in determining normal and clearly abnormal embryos, are prone to a still to be determined degree of “false-positive” diagnoses.

A principal reason for misdiagnoses is that embryos are often “mosaic,” meaning that they are made up of more than one line of cells. However, developing embryos grow from only one line of cells and often segregate out the line(s) of abnormal cells. Those cells are usually segregated into the so-called trophectoderm of the embryo, and away from the core of the embryo. The trophectoderm later develops into the placenta, and if the placenta later contains an abnormal cell line, it is of little consequence, as long as the embryo/fetus contains only cells from an initially normal cell line.

Determinations of chromosomal abnormalities of an embryo can be performed at one of two stages during embryo development: at 6-8 cells, the so-called cleavage-stage (three days after fertilization) or on days-5/6, the so-called blastocyst stage, when the embryo already has more cells than can be counted.

Day-3 embryos usually have only one cell removed and tested. If this cell is chromosomally abnormal and the other cells of the embryo are normal, the embryo as a whole would be classified as “abnormal.” Mosaic embryos, however, frequently self-correct. This is often the consequence of the embryo “segregating” all daughter cells from abnormal cells toward the trophectoderm. If this occurs, this normal embryo would therefore be inaccurately diagnosed as abnormal and discarded for no reason, creating a so-called “false-positive” diagnosis.

Increasingly, embryos are tested on day-5/6 because biopsies at this stage allow for the removal of multiple cells from the trophectoderm and are considered more accurate than single cell biopsies on day-3. In these cases, however, biopsies are made from the trophectoderm where embryos segregate their abnormal cell lines. Therefore, the chance of biopsying a segregated abnormal cell line is higher. This would result in the embryo being discarded even though the embryo may be normal.

In a recent study by colleagues in Toronto, Canada, physicians took trophectoderm biopsies from consenting patients’ embryos and sent them to different genetic laboratories for analysis. Though all three were highly reputable genetic laboratories that utilized state-of-the-art technologies, the results demonstrated significant differences,4 supporting the concept that where the trophectoderm is biopsied matters.

Additionally, normal stem cells have been obtained from embryos even when trophectoderm biopsies defined the embryos as “abnormal” (Shoukhrat Mitalipov, PhD, personal communications). Finally, we are aware of at least two healthy births in infertile women who had only embryos transferred, which after PGD/PGS were classified as monosomies (Barad DH, Vidali A, Kushnir VA, Gleicher N; unpublished data).

All of this is of little consequence for women with large numbers of embryos. Because such women, even after losses of “false-positive” embryos, will still have “normal” embryos available for transfer, their “false-positive” losses will not significantly affect their pregnancy chances with IVF. For women with small embryo numbers, loss of one or two “false-positive” embryos can mean the difference between having or not having an embryo transfer and, therefore, between having or not having a pregnancy chance in their IVF cycle.

It therefore appears ethically appropriate to offer patients with LFOR and apparently no chromosomally normal embryos in an IVF cycle in carefully selected circumstances the opportunity to transfer PGD/PGS defined chromosomally abnormal embryos under the assumption that some of them may have been ”false-positively” designated as “abnormal.” Offering such an option is, however, ethically contingent upon absence of “non-lethal” chromosomal abnormalities in so transferred embryos, which would have the potential of implantation and delivery. These “non-lethal” chromosomal abnormalities can lead to genetic disorders like Down’s syndrome and Turner’s syndrome (and others) if delivered. Therefore, only reported aneuploidies known to be “lethal” should be transferred in such circumstances.

Most chromosomally abnormal embryos do not implant, and if they do, they frequently result in early miscarriages. Chromosomal abnormalities, which never result in live births, are therefore considered “lethal.” Ongoing pregnancies with “lethal” chromosomal abnormalities are almost impossible. The transfer of embryos with “lethal” chromosomal abnormalities can, therefore, be performed under the assumption that either a normal cell line will take over during further growth of the embryos or that if the “lethal” cell line becomes dominant, the embryo will with great likelihood either fail to implant or be quickly miscarried.

Because of the obvious complexities of here outlined ethical process, comprehensive counseling of patients is essential prior to obtaining written informed consent from patients for the transfer of embryos, by PGD/PGS reported to be aneuploid. Aside from above outlined considerations, such counseling also has to discuss the possibility that, though highly unlikely, the establishment of an ongoing chromosomally abnormal pregnancy can never completely be ruled out. Therefore, patients who are considering transfer of such embryos should be advised that all pregnancies established from such embryos should undergo early prenatal genetic testing, and that in chromosomally abnormal pregnancies a medically induced pregnancy termination may become necessary.

Considering all in this OPINION outlined issues, CHR has developed a policy, which allows for individualized transfer of embryos, by PGD/PGS reported to be aneuploid. CHR will make this policy gladly available to colleagues and IVF centers.


  1. Barbash-Hazan S, Frumkin T, Malcov M, Yaron Y, Cohen T, Azem F, Amit A, Ben-Yosef D. Preimplantation aneuploid embryos undergo self-correction in correlation with their developmental potential. Fertil Steril 2009;92:890-896
  2. Van Echten-Arends J, Mastenbroek S, Sikkema-Raddatz B, Korevaar JC, Heineman MJ, van der Veen F, Repping S. Chromosomal mosaicism in human preimplantation embryos: a systemic review. Hum Reprod Update 2011;17:620-627
  3. Bazrgar M, Gourabi H, Valojerdi MR, Yazdi PE, Beharvand H. Self-correction of chromosomal abnormalities in human preimplantation embryos and embryonic stem cells. Stem Cells Dev 2013; 22:2449-2456
  4. Esfandiari N, Bentov Y, Casper RF. Trophectoderm biopsy for aneuploidy screening using different platforms and conflicting test results. ESHRE O-034, Munich, Germany, June 29-July 2, 2014


We hereby expand from monosomies  to trisomies the OPINION above, originally expressed on October 27, 2014, that it is reasonable to offer embryo transfer to patients with appropriate informed consent who have no or too few euploid embryos for transfer after PGD/PGS.

This expansion of indication is based on a number of factors:

  1. The recent recognition by CHR investigators that, likely, ca. 50% of blastocyst-stage embryos are mosaic (other investigators recently suggested a ca. 30% rate; Grifo et al., ASRM 2015);
  2. This high rate of mosaicism does not permit reliable diagnosis of ploidy by trophectoderm biopsy even with most advanced diagnostic platforms;
  3. The establishment of so far 3 known chromosomally normal pregnancies after transfer of aneuploid embryos at 3 different IVF centers (Gleicher et al, ASRM 2015);
  4. Studies performed at CHR and one other center (Esfandiari et al., ESHRE 2014), which demonstrated very poor correlation between results reported from 2 reputable PGD/PGS laboratories;
  5. Studies reported at another IVF center, which reported poor correlation between biopsies of trophectoderm and inner cell mass (Orvieto et al, Personal communication, reported at ASRM 2015);
  6. Studies performed at CHR, which demonstrated very high outcome discrepancy between multiple biopsies of the same embryos.

CHR, therefore, concludes that the high prevalence of mosaicism in blastocyst-stage embryos renders PGS/PGD ineffective. Moreover, especially in women with few embryos, false positive embryo diagnoses will adversely affect pregnancy chances, unless such embryos are transferred.

Norbert Gleicher, MD, leads CHR’s clinical and research efforts as Medical Director and Chief Scientist. A world-renowned specialist in reproductive endocrinology, Dr. Gleicher has published hundreds of peer-reviewed papers and lectured globally while keeping an active clinical career focused on ovarian aging, immunological issues and other difficult cases of infertility.

Addendum published: November 4, 2015


Center for Human Reproduction
21 E 69th Street
Upper East Side

New York, NY 10021
Phone: 626-385-7918
Fax: 212-994-4499

Office Hours

Get in touch