Why is it so hard to pick the best embryo?

Let’s think of an illustration for one of the smaller species!  If you were to look at mouse embryos, you would see they all look alike at the blastocyst stage and only < 2% have the wrong number of chromosomes (aneuploid). Look at blastocyst human embryos and no two look alike. Furthermore, over two-thirds of the embryos are aneuploid: they have the wrong number of chromosome and therefore cannot result in the birth of a child. What gives! Blame mother nature and human evolution. The reason comes from the need for humans to be able to reproduce sufficiently to provide enough people for the next generation.


Nature approached this problem for humans by making then very adaptable to their environment. Because of this, humans can inhabit almost every place on the planet. But since the conditions are so varied around the globe, creating all humans alike would prevent them from existing in many habitats. One size definitely does not fit all. So, nature designed humans to reproduce with a great deal of adaptability. Now if you were mother nature and you started with a number of different people to fit into a certain place, would you decide to make the choice of who lives and who dies only after birth. No! Carrying a child to term is a tremendous task for a woman. So, if you want to start to select which embryo survives, you start with a large number of different embryos and only the ones that fit the conditions survive. The variability is introduced at the formation of the embryo, but that creates chromosomal instability, and thus a large number of human embryos have the wrong number of chromosomes and those embryos are eliminated from the race. Humans have 23 sets of chromosomes (one from mom and one from dad) and the error is random. So, one embryo may have three number 22 chromosomes while another may have only one number 16. The embryo will develop normally until it reaches the point where the abnormal chromosome prevents further development. Some embryos never really get to day three while others can get to 8-10 weeks. Thus, the variability in how embryos look.

For a pregnancy to be successful, there needs to be a normal embryo, the lining of the uterus (endometrium) needs to be ready to accept the embryo, and the embryo needs to be at the endometrium at the right time. Normal embryos develop in a normal, tightly timed sequence of development. Its like building a house. You don’t build the roof first and then the foundation. So many abnormal embryos develop in the wrong time sequence. Mother nature, in her infinite wisdom, decided that in order for an embryo to be selected to continue to develop, it had to be at the correct stage of development when the uterus was ready to accept the embryo. Nature created a narrow window of time when an embryo could implant so that many, but not all, of the abnormally developing embryos would simply perish. This optimal time for implantation is called the implantation window.

Shouldn’t all embryos look alike when they are ready to implant?
Do all people look alike? There is considerable physical variability in human embryos because they are genetically different. Some embryos can produce a child, and some cannot. Since the embryos vary physically, it becomes difficult to look at human embryos under the light microscope and see which one is normal and which one is abnormal. But what does being abnormal mean? In this context, abnormal means having the wrong number of chromosomes. Human embryos can look perfectly normal and yet have the wrong number of chromosomes. Conversely, an embryo can be downright ugly and yet have a normal number of chromosomes and create a beautiful child.

Until recently, it has not been possible to test embryos to determine if they have the correct number of chromosomes (euploid) or have the wrong number of chromosomes (aneuploid). Before this technical advancement, the appearance of the embryo was used to select the best embryo to transfer. Further technical advances permit observing an embryo developing without taking it out of its culture conditions (morphokinetic). Recent and future advances combine AI and molecular biology in an attempt to pick the best embryo. But for the most part, selecting the correct embryo today relies upon looking at the embryo on the day of transfer and choosing the embryo based upon standard guidelines for embryo selection. The ability to determine the chromosome status of the embryo (ploidy) has allowed comparison between the ploidy of the embryo and the embryo score with morphological assessment or with morphokinetic scoring.

So how good are the morphology base methods at picking the euploid embryo?
Lukewarm at best.
Given that most human embryos created in IVF have the wrong number of chromosomes (estimated to be a high as 70%), what can embryologists do to improve the odds of picking a good embryo? Currently, there are three main techniques used by embryologists to pick the best embryo and a number of other techniques being developed. Most embryologist employ an early method based upon how the embryo looks using the light microscope.[i] The appearance of an embryo is called the morphology of the embryo. There are two general techniques based upon morphology, one complex and one simplified. The simple method categorizes embryos as either good, fair, or poor. The more complex method uses both a number which refers to the stage of a blastocyst embryo and a double grading system for blast embryos for the inner cell mass and the outer cells (trophectoderm) using and A,B,C system. So, backing up a bit, what does this mean? Embryos are transferred either on day three of their development or day 5/6. Day 3 embryos are called cleavage stage embryos and day 5/6 are called blastocyst embryos. Day three embryos have 8-10 cells. Embryos at this stage are graded upon how many cells are present, what percentage of the cells have broken down and fragmented, and what the shape of the cells looks like. The ideal number of cells on day 3 is 7-9 and strangely when a day 3 embryo has more than 7-9 cells, the chance for a successful pregnancy goes down. Depending upon the embryologist, the scoring for day three embryos may be a simple as good, fair, poor or may report the score based upon the number of cells, the percent fragmentation, and the shape where each category is ranked good, fair, poor.

However, today most IVF programs are extending the culture period to 5/6 days before transferring the embryos. This results in a greater success rate per transfer since embryos of poor quality cannot continue to develop beyond the day 3 stage. The embryo’s own genetic makeup determines if an embryo can make the transition from a cleavage stage embryo to a blastocyst embryo. A blastocyst embryo looks like a basketball with a glop of cells inside of it and has a total of over 100 cells. The glop of cells is called the inner cell mass and forms the fetus while the outer cells form the placenta and are called the trophectoderm. The trophectoderm cells make the pregnancy hormone HCG. So, a woman can have a positive pregnancy test because the trophectoderm makes the HCG, but have an abnormal pregnancy because there is no inner cell mass to form the child. This is called a blighted ovum and usually results from a pregnancy with the wrong number of chromosomes.

The grading of a blastocyst embryo assigns a number to the expansion of the blastocyst (1-6), the quality of the trophectoderm (A.B.C), and the quality of the inner cell mass (A,B,C). Expansion of the blast refers to the size of the blastocyst cavity and can be either early blast (1-2), expanding blast (3-4), hatching blast (5), or hatched blast (6). A hatched blast has shed its shell, the zona pellucida. Theoretically, a normal embryo should form a blastocoel in the proper timeframe. If the embryo does not do this, then an argument can be made that the embryo is not physiologically normal, and thus has a lower chance of success. The fluid cavity in the blastocyst embryo (blastocoel) is formed by the cells of the trophectoderm. The cavity cells of the trophectoderm creates a concentration gradient of sodium/ potassium and this process requires considerable energy. An abnormal embryo will not have the ability to perform the necessary steps to create the blastocoel in the proper time frame. Blastocysts expand and collapse continuously. Each expansion affects the shell (zona pellucida) of the embryo so that each embryo has an expansion history resulting in a thinning of the shell. The trophectoderm creates the blastocoel, but it also interfaces the endometrial lining. Thus, the trophectoderm is a critical player in the proper development of the embryo and in the proper implantation of the embryo. A normal trophectoderm has a number of cells tightly packed together. A poorly formed trophectoderm suggests a poorly performing embryo with a lower chance of success. While the trophectoderm is critical, the inner cell mass which forms the child directs the trophectoderm. An inner cell mass with a large number of uniform cells tightly packed suggests proper functioning of the embryo and thus a higher chance of success. There is controversy about which score is the most important, so each embryologist will determine what the best embryo is based upon their experience.

Furthermore, there is considerable evidence that there is significant variation between embryologists scoring embryos. A study by Khosravi et al[ii] had 394 embryos evaluated by 5 embryologists. The five embryologists agreed on only 89 out of the 394 embryos graded! As I said…lukewarm at best.

A study by Minasi et al[iii] compared the results from preimplantation genetic testing, standard morphological criteria and morphokinetic data to determine if one technique was superior to the others. The results demonstrated that 56% of the embryos implanted and of these, 19% were biochemical pregnancies and three were ectopic pregnancies. Like many studies, this study found a higher pregnancy rate for frozen embryo transfers than fresh embryo transfers (49% vs 36%). When using the standard morphologic scoring, 6.5% never expanded, so that blastocyst scoring was not possible. When evaluating cleavage stage embryos (day 3), there was no correlation between the number of chromosomes (ploidy state) or the morphological. Up through day three, anything needed for the embryo to develop is provided by the oocyte. After that, the genetic code of the embryo becomes responsible for further development. If that code is wrong, as for example aneuploidy, then the embryo be able to develop to day three and look normal. For blastocyst embryos, the state of expansion, the quality of the TE and the ICM all correlated with ploidy state of the embryos. For embryos with low scores on these attributes, there was an increase in aneuploidy, thus the worse the embryo looked the more likely it was to be an aneuploid embryo. When evaluating the embryos using morphometric (dynamic) testing, the 4- cell stage of embryo development was reached earlier in euploid embryos. This faster rate of expansion for blastocyst embryos also was associated with a higher euploid state. Other morphometric parameters did not correlate with the ploidy of the embryos. In summary, for day 3 embryos, looking at the embryo did not help in identifying the euploid embryo. However, for the blastocyst embryo, morphological characteristics did correlate with the ploidy of the embryo. But two things emerged with this study which impact the decision about which embryo to transfer: some aneuploid embryos look perfect, not all poor-quality embryos were aneuploid. Thus, while trying to decide which embryo to transfer is aided by looking at the embryo and watching it develop, the system is not as good as knowing the ploidy status of the embryo.

Conclusion:
The best method for determining which embryo to transfer is to do preimplantation embryo karyotyping. When that is not available, the standard morphological evaluation seems comparable to morphometrics. Furthermore, most lVF centers will not be able to afford the equipment needed for morphometric studies or will not have the space and time to do this. Future advances in technology may use the field of “omics” using metabolites, RNA, DNA, from the embryo to increase the ability to pick the best embryo.

[i] Gardner, DK, Lane M, Stevens J, Schlenker T, Schoolcraft WB Blastocyst score affects implantation and pregnancy outcome: towards single blastocyst transfer. 2000 Fertil. Steril. 73: 1155[ii] Khosravi et al Deep learning enables robust assessment and selection of human blastocysts after in vitro fertilization 2019 Digital Medicine 2:21[iii] Minasi MG et al Correlation between aneuploidy, standard morphology evaluation and morphokinetic development in 1730 biopsied blastocysts: a consecutive case study 2016 Hum. Reprod. 31:2245

Embarking on a new adventure is always stressful but – keeping an eye on the “good” – makes it all worthwhile. Now that we have had a chance to see the success of our new solo practice and all of the changes we have made. I am happy to report that we are all excited about the results. Along with upgrading our protocols and our laboratory, my personal interaction with each patient by doing all of their ultrasounds and the ability to more closely monitor their ART cycles has been personally gratifying. But none of this matters if the success rates are not there. While it is early days yet, we are excited that over 70% of the patients in our recent ART cycle celebrated positive pregnancy tests. Together, our committed patient base and the exceptional team at the Rinehart Fertility Center, set our new adventure on a successful path. Thank you for your support and know that we are here to take care of all fertility patients with the same individual care – and, hopefully, create more pregnancies!

Leave a reply