Benchmarks of egg quality
In the vast majority of cases, oocyte quality is the most important variable determining the success of reproduction. Yet, there is no single clinically useful marker that would predict it. Neither the level of reproductive hormones, the size of the follicles, nor even a visual observation of the oocyte after it is retrieved during IVF, can provide guidance to its developmental potential. Therefore today, oocyte quality can only be defined as a “probability of an oocyte to become a child, once fertilized”.
Oocyte chromosomal maturity
Being unable to evaluate oocyte’s quality prospectively, we can only say whether it is mature, meaning that it has completed the first meiotic division and is at the expected stage of “chromosomal configuration”. This stage can be easily discerned by an observation of the so-called first polar body. At the IVF retrieval, once an embryologist finds that an oocyte retrieved from the follicle has a polar body she will record it as “mature”. The physician considers ovarian stimulation successful if most of the harvested eggs are mature and thus oocyte chromosomal maturity became a synonym of oocyte quality. Yet, we knew for a long time, that the ability of the egg to extrude the polar body is one of the earliest developmental competency benchmarks. This benchmark is reached before the ability to develop after fertilization is attained. Thus the oocyte’s “maturity”, as it is defined today, is not the same as seeing the red color of the tomato and deciding it is ripe. It is more like trying to choose a ripe tomato by looking only at size and shape.
In order to fully appreciate the value of an oocyte’s ability to extrude the first polar body as a competency benchmark, you may think about a newborn developing a sucking reflex. It appears as early as 32 weeks, in the otherwise still immature fetus. Now, imagine that a sucking reflex would be the only benchmark of a fetus’s developmental competency.
Follicle size
The reproductive endocrinologist will monitor the size of the follicles during ovarian stimulation and when the largest follicle reaches about 22-25 mm in diameter, she will trigger the ovulation so that the oocytes can be retrieved. This follicular size was chosen because it is close to the follicle’s size just before ovulation in the natural cycle with an underlying assumption that it signifies the oocyte’s competency. However, the size of the follicle and the pace of its expansion is determined by the amount of circulating FSH and the activity of the mural granulosa. It is not influenced by an oocyte and therefore can not tell us anything about an oocyte. That is why, for example, the largest follicle has the same chance to contain the best egg as a smaller follicle.
The misconception that follicular size is a reflection of the oocyte’s properties can be traced to the early stages when the oocyte does indeed controls the follicle (by secreting several specific growth factors). But once the cavity is formed and expanded, the gradient of oocyte-derived growth factors becomes too diluted to reach distal granulose cells. This allows those granulosa cells to escape the oocyte’s control, differentiate into mural granulosa, and begin expressing receptors to FSH. Once this happens, the corona-cumulus complex will remain under the control of the oocyte-produced growth factors. However, the enlargement of the follicle, the result of mural granulosa proliferation and fluid accumulation, will be controlled initially by FSH and later by both FSH and LH. Full recognition of this dual control is very important, because it helps to appreciate that the processes responsible for the egg acquiring competency (nursing with corona-cumulus), and the process responsible for the ultimate egg release from the follicle (ovulation), are driven by completely different, independent, and uncoordinated mechanisms.
While the follicle is small, the gradient of growth factors (produced by the egg) prevents the induction of FSH receptors in granulosa. In yellow: a gradient of oocyte produced growth factors: CDF-9, BMP-15, SMAD
Reproduced from Dozortsev and Diamond, Fert Stert 2020, open access publication
As an egg is losing control over the follicle to FSH, its time for maturation is at the mercy of the follicle (or an REI). In yellow: a gradient of oocyte-produced growth factors: CDF-9, BMP-15, SMAD
Reproduced from Dozortsev and Diamond, Fert Stert 2020, open access publication
This is why just like an obstetrician cannot use the size of the belly to predict a fetus’ maturity, it is not possible to use the size of the follicle to predict an egg’s competence.
Furthermore, during controlled ovarian stimulation, a woman receives additional amounts of FSH, which makes the follicle “grow” about 1.2 times faster than during the natural cycle (in many cases the pace is even higher). At the same time, the pace of the oocyte’s acquisition of its developmental competence is unaffected. This creates a potential for asynchrony between the follicular growth and oocyte acquisition of developmental competence – “term maturation”.
Estradiol
During the follicular phase, estradiol plays multiple roles, some of which we may still not know.
During controlled ovarian stimulation, estradiol is also used as an indirect measure of an egg’s competence. It is usually assumed that a certain level of E2 per follicle (about 200 pg/ml) predicts an egg’s competence. However, the level of E2 is just another way to measure the follicle, because its main source is mural granulosa, which has everything to do with the follicle and nothing to do with the oocyte.
Follicular stimulating hormone (FSH)
The level of FSH is a mirror reflection of the level of estradiol and therefore cannot predict oocyte quality
Anti-mullerian hormone (AMH)
AMH is produced mainly by mural granulosa of antral follicles and correlates very well with the number of antral follicles and number of oocytes retrieved during ovarian stimulation, but its level is not predictive of oocyte quality.
To sum it up, neither E2, FSH, or AMH nor the size of the follicle cannot be used to predict an oocyte’s competence. Figuratively speaking, they represent different parameters of the box, not its content.
The current paradigm is that patients with PCOS have intrinsically poor-quality eggs. This conclusion is drawn from the observation that PCOS patients have low fecundity. However, low fecundity in these patients can also be attributed to an irregular cycle, which does not control by eggs and therefore cannot possibly reflect their properties.
Another line of evidence for the poor quality of the oocytes in IVF patients comes from IVF experience where we see that despite yielding more eggs than an average IVF patient, the percentage of excellent quality embryos is relatively low.
However, the stimulation and the respective follicular phase in these patients is usually shorter than average because a very large cohort of eggs becomes recruited. Rapidly rising levels of E2 pose a risk of potentially life-threatening hyperstimulation, forcing a physician’s hand to trigger earlier. As the result ,the vast majority of the oocytes have deprived of the opportunity to achieve term stimulation, similarly to very young patients.
About 75% of preimplantation (in vitro) embryos are chromosomally abnormal. By the time they reach a blastocyst stage, about 40% of all surviving embryos are chromosomally abnormal. On average, oocytes contribute about 27% of numerical errors, sperm about 7% and the remaining about 3 % appear as an embryo divides (the numbers for sperm, oocyte, and embryo errors are probabilistic and can’t simply be added together to get 100%). With age, an oocyte’s relative contribution to overall chromosomal errors increases.
Chromosomal errors in oocytes
It is very important to keep in mind that ALL immature oocytes in the ovary, at any age, are chromosomally normal. An oocyte remains chromosomally normal until after a natural LH surge or an artificial trigger is administered. It is only at this moment that chromosomal errors begin to take place. More than half of all errors happen at this time. Most of the remaining errors take place after fertilization before pronuclei are formed in a zygote.
The percentage of mature oocytes with numerical chromosomal errors can be estimated in two ways. An indirect approach, based on the number of chromosomal errors in sperm and embryos, gives an expected percentage of chromosomally equal to 27%. This is in astonishingly close concordance with a direct method of observing chromosomal errors in oocytes – 26.5%.
Chromosomal errors in sperm
There are only 3 publications that exist directly evaluate the rate of chromosomal errors in individual human spermatozoa by injecting them into mouse oocytes. They evaluated both, numerical and structural aberrations in sperm chromosomes. Because we cannot be certain that the fidelity of replication in mouse oocytes for human chromosomes is 100%, the number of structural aberrations may be higher than in reality. However, the number of numerical aberrations can be accepted without any reservations.
My friend and colleague Dr. Andrew Rybouchkin was the first to use this experimental model for evaluating chromosomes of human spermatozoa. He has also shown that so-called globospermia, when sperm cells are lacking acrosome and have round heads) do not have any increase in the number of chromosomal aberrations, compared to sperm donors.
Dr. Lee (from Dr. Yanagimachi) group has demonstrated using the same approach that the rate of numerical chromosomal aberrations in sperm is low, 1.3%. They also show that in grossly chromosomally abnormal sperm cells the rate of aberrations can be as high as 26%. It is very important to keep in mind that the criteria they used to classify a sperm cell as “abnormal” are very different from those used in standard sperm morphology. First, they were judged as abnormal using the inverted microscope, live, without any fixation and staining. Second, they were truly grossly abnormal and clearly represent an outlier of extremely poor morphology, which contributes only a tiny percentage point to morphologically abnormal sperm cells as judged after fixation and staining using conventional techniques.
From the above studies, we can conclude that no more than 7% of the sperm cells are chromosomally abnormal.
