Preimplantation Genetic Screening (PGS)

Almost 15 years ago I  with my associate, Levent Keskintepe PhD were the first to introduce full chromosome Preimplantation Genetic Sampling/Screening (PGS) into the IVF  clinical realm to try and identify euploid embryos whose cells contained the required 46 chromosomes (23 pairs), necessary to render them  potentially “competent” to propagate viable pregnancies.  Aneuploid embryos (those that have more or less than a total of 46 chromosomes) are by and large considered to be “incompetent”,  far less likely to propagate a viable pregnancy and thus largely unworthy of being transferred to the uterus. While sperm karyotype can certainly impact the chance of embryo aneuploidy, it is egg karyotype that by far have exerts the most significant influence. And the older the woman, the higher the incidence of egg aneuploidy. Up to the mid-30’s, the incidence of egg aneuploidy is about 50%. By age 40y it increases to about 75% and by age 45 it is >90%.  In younger women, a euploid embryo has better than a 45% chance of propagating a pregnancy while for a women of 40y the baby rate per transferred euploid embryo decreases to approximately  40%. For those women who conceive using such euploid embryos, the miscarriage rate falls to below 10%.

Genetic Testing Embryos for IVFInitially the standard method used for PGS was, comparative genomic hybridization (CGH). About 8-10 years ago, a new and improved technology known as next generation gene sequencing (NGS) was introduced. This method determines the precise order of nucleotides within a DNA molecule. NGS has now virtually replace all other methods used for PGS.

In the vast majority of cases PGS involves the testing of advanced, day 5-6 embryos (blastocysts). These are biopsied and several cells are microsurgically removed from the trophectoderm-TE (the outer cellular layer of the blastocyst) and subjected to PGS analysis.  The biopsied blastocysts are then ultra-rapidly frozen (vitrified) and banked (vitribanked) while awaiting the PGS report, whereupon patients are scheduled for] frozen embryo transfer (FET) using euploid embryos.

When selectively used, PGS can be a valuable tool to improve implantation potential and IVF outcome. However, as stated, PGS is not a panacea. There are factors other than numerical chromosomal integrity (karyotype) that also influence embryo “competency”, profoundly.  In my opinion, PGS is currently being over-used, particularly when used in younger women who have normal ovarian reserve.  Unless this remarkable technology is properly applied, it is my belief that we will soon see the popularity and utility of PGS starting to decline.

Embryo aneuploidy and “mosaicism”.

Human embryo development occurs through a process that encompasses reprogramming, sequential cleavage divisions and mitotic chromosome segregation and embryonic genome activation. Chromosomal abnormalities may arise during germ cell and/or preimplantation embryo development and represents a major cause of early pregnancy loss. About a decade ago, I and my associate, Levent Keskintepe PhD were the first to introduce full embryo karyotyping (identification of all 46 chromosomes) through preimplantation genetic sampling (PGS) as  a method by which to selectively transfer only euploid embryos (i.e. those that have a full component of chromosomes) to the uterus. We subsequently reported on a 2-3-fold improvement in implantation and birth rates as well as a significant reduction in early pregnancy loss, following IVF. Since then PGS has grown dramatically in popularity such that it is now widely used throughout the world.

Most IVF programs that offer PGS services, require that all participating patients consent to all their aneuploid embryos (i.e. those with an irregular quota of chromosomes) be disposed of. However, there is  now growing evidence to suggest  that following embryo transfer, some aneuploid embryos will in the process of ongoing development,  convert to the euploid state (i.e. “autocorrection”) and then go on to develop into chromosomally normal offspring. In fact, I am personally aware of several such cases occurring within our IVF network. So clearly, summarily discarding  all aneuploid embryos as a matter of routine  we are sometimes destroying  some embryos that might otherwise have “autocorrected” and gone on to develop into  normal offspring.

Thus by discarding aneuploid embryos the possibility exists that we could be denying some women the opportunity of having a baby. This creates a major ethical and moral dilemma for those of us that provide the option of PGS to our patients. On the one hand, we strive “to avoid knowingly doing harm” (the Hippocratic Oath) and as such would prefer to avoid or minimize the risk of miscarriage and/or chromosomal birth defects and on the other hand we would not wish to deny patients with aneuploid embryos, the opportunity to have a baby.

The basis for such embryo “autocorrection” lies in the fact that some embryos found through PGS-karyotyping to harbor one or more aneuploid cells (blastomeres) will often also harbor chromosomally normal (euploid) cells (blastomeres). The coexistence of both aneuploid and euploid cells coexisting in the same embryo is referred to as “mosaicism.”

It is against this background, that an ever-increasing number of IVF practitioners, rather than summarily discard PGS-identified aneuploid embryos are now choosing to cryobanking (freeze-store) certain of them, to leave open the possibility of ultimately transferring them to the uterus. In order to best understand the complexity of the factors involved in such decision making, it is essential to understand the causes of embryo aneuploidy of which there are two varieties:

  1. Meiotic aneuploidy” results from aberrations in chromosomal numerical configuration that originate in either the egg (most commonly) and/or in sperm, during preconceptual maturational division (meiosis). Since meiosis occurs in the pre-fertilized egg or in and sperm, it follows that when aneuploidy occurs due to defective meiosis, all subsequent cells in the developing embryo/blastocyst/conceptus inevitably will be aneuploid, precluding subsequent “autocorrection”. Meiotic aneuploidy will thus invariably be perpetuated in all the cells of the embryo as they replicate. It is a permanent phenomenon and is irreversible. All embryos so affected are thus fatally damaged. Most will fail to implant and those that do implant will either be lost in early pregnancy or develop into chromosomally defective offspring (e.g. Down syndrome, Edward syndrome, Turner syndrome).
  2. “Mitotic aneuploidy” occurs when following fertilization and subsequent cell replication (cleavage), some cells (blastomeres) of a meiotically normal (euploid) early embryo mutate and become aneuploid. This is referred to as “mosaicism”. Thereupon, with continued subsequent cell replication (mitosis) the chromosomal make-up (karyotype) of the embryo might either comprise of predominantly aneuploid cells or euploid cells. The subsequent viability or competency of the conceptus will thereupon depend on whether euploid or aneuploid cells predominate. If in such mosaic embryos aneuploid cells predominate, the embryo will be “incompetent”). If (as is frequently the case) euploid cells prevail, the mosaic embryo will likely be “competent” and capable of propagating a normal conceptus.

Since some mitotically aneuploid (“mosaic”) embryos can, and indeed do “autocorrect’ while meiotically aneuploid embryos cannot, it follows that an ability to reliably differentiate between these two varieties of aneuploidy would potentially be of considerable clinical value. The recent introduction of a variety of preimplantation genetic screening (PGS) known as next generation gene sequencing (NGS) has vastly improved the ability to reliably and accurately karyotype embryos and thus to diagnose embryo “mosaicism”.

The ability of mosaic embryos to autocorrect is influenced by the stage at which the condition is diagnosed as well as the percentage of mosaic cells. Many embryos diagnosed as being mosaic while in the earlier cleaved state of development, subsequently undergo autocorrection to the euploid state (normal numerical chromosomal configuration) during the process of undergoing subsequent mitotic cell to the blastocyst stage. Similarly, mosaic blastocysts can also undergo autocorrection after being transferred to the uterus. The lower the percentage of mosaic cells in the blastocyst the greater the propensity to autocorrect and propagate chromosomally normal (euploid) offspring. By comparison, a blastocyst with 10% mosaicism could yield a 30% healthy baby rate with 10-15% miscarriage rate, while with >50% mosaicism the baby rate is roughly halved and the miscarriage rate double.

Aneuploidy involves the addition (trisomy) or subtraction (monosomy) of one or part of one chromosome in any given pair.  As previously stated, some aneuploidies are meiotic in origin while others are mitotic “mosaics”. Certain aneuploidies involve only a single, chromosome pair (simple aneuploidy) while others involve several pairs (i.e. complex aneuploidy). Aside from monosomy involving the absence of the y-sex chromosome (i.e. XO) which can result in a live birth (Turner syndrome) of a compromised baby, virtually all monosomies involving autosomes (non-sex chromosomes) are likely to be lethal and will rarely result in viable offspring. Some autosomal meiotic aneuploidies, especially trisomies 13, 18, 21, can propagate viable and severely chromosomally defective babies. Other meiotic autosomal trisomies will almost invariably, either not attach to the uterine lining or upon attachment, will soon be rejected. All forms of meiotic aneuploidy are irreversible while as stated, mitotic aneuploidy (“mosaicism) can autocorrect, yielding healthy offspring. Most complex aneuploidies are meiotic in origin and will thus almost invariably fail to propagate viable pregnancies.

Since certain “mosaic” meiotic aneuploid trisomy embryos (e.g. trisomies 13, 18, & 21) can potentially result in aneuploid concepti. For this reason, it is my opinion that unless the woman/couple receiving such embryos is willing to commit to terminating a resulting pregnancy found through amniocentesis or chorionic villus sampling (CVS) to be so affected, she/they are probably best advised not to transfer have them transferred to the uterus. Embryos harboring other autosomal mosaic trisomic embryos, should they not autocorrect in-utero will hardly ever produce a baby and as such there is hardly any risk at all…in transferring such embryos. However, it is my opinion that in the event of an ongoing pregnancy, amniocentesis or CVS should be performed to make certain that the baby is euploid.  Conversely, when it comes to mosaic autosomal monosomy, given that virtually no autosomal monosomy embryos are likely to propagate viable pregnancies, the transfer of such mosaic embryos is virtually risk free.  Needless to say,  in any such cases , it is absolutely essential to make full disclosure to the patient (s) , and to insure the completion of a detailed informed consent agreement which would include a commitment by the patient (s) to undergo prenatal genetic testing (amniocentesis/CVS) aimed at excluding a chromosomal defect in the developing baby and/or a willingness to terminate the pregnancy should a serious birth defect be diagnosed.

Should PGS be done routinely in IVF?

Initially, after Dr Keskintepe and I introduced PGS testing into the clinical IVF arena (2005) initial results were very encouraging. Pregnancy rates of >50% and birth rates of 50-60 were reported. In addition, the reported incidence of miscarriages and chromosomal birth defects was likewise greatly reduced.  We were so enthusiastic that we envisioned a time when full embryo karyotyping might become routine in IVF.  As it has turned out, we were disillusioned when following the widespread introduction of PGS testing success rates started declining. This was especially true in cases where the technology was applied to the embryos of older women and women who regardless of age had diminished ovarian reserve (DOR).  Further investigation led to the following conclusions:

  1. Chromosomal numerical integrity, is not the only factor that impacts embryo “competency”. It is probably that non-chromosomal, genetic and metabolomic factors might also be age-related.
  2. Many variables, that also determine IVF outcome and which are often outside the control of the embryology/genetic laboratory also affect embryo competency. These include selection and implementation of individualized protocols for controlled ovarian stimulation (COS), anatomical and immunologic implantation factors as well as the technical skill of the physician performing embryo transfer etc.
  3. Not all PGS-aneuploid embryos are “incompetent”. As described above, some aneuploid embryos are “mosaic” (see elsewhere) and often capable of “autocorrecting” upon being transferred to the uterus,

It is now my considered opinion that PGS-embryo selection should only be considered in the following circumstances:

  1. Older women (over 35y) whose eggs are much more likely than in younger women, to be aneuploid and incompetent.
  2. Women who, regardless of age have significant diminution of ovarian reserve (DOR) and need to “make hay while the sun shines”!
  3. Recurrent unexplained IVF failure.
  4. Recurrent pregnancy loss (RPL).
  5. Embryo gender selection for family gender balancing
  6. Women who have alloimmune implantation dysfunction (IID) with activation of uterine natural killer cells (NKa).
  7. Where karyotyping reveals one or other partner to have a balanced chromosomal translocation
  8. Known or anticipated specific genetic abnormalities


PGS for Gender Selection and Family balancing.

Nevertheless, it is an inescapable reality that the very idea of medical sex selection challenges moral and ethical beliefs at their very foundation. Many hold that the growing popularity of gender selection solely for the convenience of altering a family’s gender balance represents an unwanted example of how assisted reproductive technology is subject to abuse…and thus it should be outlawed. They also see it as an example of a disturbing trend towards “designer babies” where genetic engineering could be used to manipulate the intellect, body configuration, build, height, and the talents of future offspring. This assertion is commonly followed by the tantalizing question as to where all this would end and whether we as a society “would really want to live in such a world.” There is, however, one clear exception to the apparent across-the-board opposition to sex selection that is well worthy of mention. This applies in cases where sex selection is used to avoid the occurrence of a serious medical disorder that selectively affects one gender or the other (e.g., Hemophilia, a life threatening bleeding disorder that selectively affects male offspring).PGD using comparative genomic hybridization (CGH) next generation gene sequencing (NGS) which assesses all the embryo’s chromosomes can be used for both detecting all the embryo’s chromosomes and thus can determine embryo “competency” reliably. It also reliably identifies gender.

Sex selection done purely for family balancing is somewhat controversial, raising concern that if widely accessible and freely available, such practice could distort the natural sex ratio, leading to a population gender imbalance. However, for this to happen, there would have to be a significant population preference for sex selection. In reality, the contrary seems to apply, since studies conducted in western societies discount these concerns. In fact, the relatively high cost of IVF with the added cost of gender selection in the United States makes it unlikely that the demand would ever become large enough to impact overall population gender balance. In addition, several studies done in Western countries have shown that the majority of people do not seem to be concerned about the gender of their offspring, and that with a few notable exceptions, gender preference does not appear to be slanted in the direction of either male or female. Thus, from a practical standpoint, such concerns are overstated.

Given that in the United States most couples do not care about the gender of their offspring, and only a minority are interested in selecting the sex of their children there is currently no risk that IVF sex-selection will impact the population gender balance. Thus, in my opinion by and large, freedom of choice should prevail and a service for sex selection should be freely available

So, I absolutely do offer gender selection in the following circumstances.

  • Medical Indications for Gender Selection:
    • For cases associated with
      • Sex-linked genetic disorders or,
      • Serious genetic disorders that are more likely to occur in one gender or the other.
    • Family balancing
      • For couples who have at least one child of the opposite gender to that which they choose for their IVF embryo transfer and,
      • For those women who do not have any children at all but prefer to have a child of one or the other gender.