A recent paper by a major proponent of PGS/PGT-A ironically may be the last straw on the genetic testing saga (we hope)
No subject in fertility practice has likely more often been a subject of these pages, a lead article or just a mention in regard to other issues in the clinical practice of IVF than the chromosomal investigation of embryos in association with IVF cycles. This procedure was initially called preimplantation genetic screening (PGS) and was renamed preimplantation genetic testing for aneuploidy (PGT-A) in 2016. The principal reason why this procedure/test was subject to so many comments in the VOICE was the strong opposition of CHR’s clinical investigators to the increasing utilization of the procedure all around the world for almost two decades.
When first proposed, even CHR was smitten by the PGS hypothesis that eliminating chromosomally abnormal (aneuploid) embryos prior to transfer should improve implantation rates for remaining supposedly normal embryos. In those days still exclusively in Chicago, CHR, indeed, established a PGS laboratory. CHR’s enthusiasm for the procedure/test, however, quickly waned, as Belgian investigators were unable to demonstrate the promised outcome benefits in IVF cycles in a series of small prospectively randomized studies. Reanalyzing some of the Belgian data, CHR investigators, indeed, started to suspect that PGS was, likely, even harmful to IVF outcomes in older women.
This suspicion was shortly thereafter confirmed, when investigators from the Netherlands, based on another prospectively randomized study, confirmed adverse outcome effects in older women. What, however, appalled CHR the most, was the reaction of prominent proponents of PGS to this study: Rather than attempting to support the utilization of PGS based on demonstration of validated outcome improvements in IVF, they, simply, criticized and denigrated the Dutch study and other researchers who were increasingly skeptical of PGS, including, of course, CHR’s own investigators.
We are referring readers of this newsletter to previous articles in previous issues of the VOICE because we do not wish to be repetitive with the historical development PGS underwent, including its renaming as PGT-A. Only so much: Over approximately 20 years, the procedure/test underwent three incarnations, which we and others have described as PGS 1.0 – PGS 3.0, and with a fourth (PGS 4.0) already lurking in the background and threatening to become the next unvalidated introduction to the marketplace. A commentary by CHR investigators, Norbert Gleicher, MD and David H. Barad, MD, MS, was just published in the highly prestigious PNAS journal (Not even non-invasive cell-free DNA can rescue preimplantation genetic testing, 2019-11710).
What all of these incarnations have in common is that they were introduced into clinical practice without any evidence that they, indeed, improved IVF outcomes. Proponents of the procedure who, obviously quite accidentally, derived significant financial benefits from promoting use of the procedure/test, simply, made up claims and, rather than proving those, challenged opponents of the procedure to disprove them. Though this is NOT how science usually works, proponents of PGS/PGT-A, nevertheless, for all of those years got away with this approach.
With evidence starting to accumulate from increasing numbers of sources that the procedure/test did not only NOT improve IVF outcomes but, indeed, was harmful to outcomes of at least poorer prognosis patients who produce only small numbers of embryos in IVF, one, of course, was left wondering why the academic establishment tolerated such obvious abuse of standard scientific norms for such a long time.
A final answer is still pending but increasing evidence points toward economic incentives created by the genetic testing industry for the IVF practice community as the ultimate inducement.
Evidence also started accumulating that the basic biology of the preimplantation-stage human embryo, simply, did not allow for the reliable determination whether an embryo was chromosomally normal (euploid) or abnormal (aneuploid). Three main discoveries led to this conclusion: First, it was noted that when human embryos demonstrate aneuploidies (chromosomal abnormalities), they in a majority of cases did not involve all cells of the embryo but only the so-called cell clones (small islands of abnormal cells in a much larger sea of chromosomally entirely normal cells). If an embryo biopsy revealed aneuploid DNA, this, therefore, did NOT suggest, as claimed by the genetic testing industry, that the whole embryo was aneuploid; it only suggested that, more likely than not, only a small island of chromosomally abnormal cells had been accidentally biopsied, with no determinable meaning for the chromosomal status of the complete embryo.
In other words, it suddenly became very obvious that many, if not most, embryos were mosaics. The definition of mosaicism is “presence of two or more cell lines in an organism derived from a single cell.” In an embryo, chromosomal abnormalities can occur in two ways: Chromosomal errors can occur during meiosis (i.e., before first cell division), in which case the chromosomal defect will be in every cell of the growing embryo. Or a chromosomal defect may occur during any subsequent cell division (mitosis). Since the number of mitotic cell divisions leading to a ca. 200-cell blastocyst-stage embryo is quite large, it should not surprise that mitotic aneuploidies in embryos are more frequent than meiotic aneuploidies. They, however, are clonal, usually involving only a cluster of cells and, therefore, are not reflective of the whole embryo.
A possibly even more important discovery was that aneuploid cells often “commit suicide” by inducing so-called apoptosis. The ability of embryos to self-correct in this way downstream from blastocyst stage was first demonstrated by British investigators in the mouse. It since has also been demonstrated in human embryos (Rockefeller University investigators in collaboration with CHR investigators, indeed, this month are presenting such evidence at the 2019 Annual Meeting of the ASRM in Philadelphia). Remarkably, both mouse and human data suggest that this ability to self-correct aneuploidies is much more pronounced in the so-called embryonic cell lineage (which produces the fetus) than in the extraembryonic cell lineage that makes up the trophectoderm, from which the placenta arises. Yet, when embryos are biopsied, it is the trophectoderm from where the biopsy is taken.
Performing a diagnostic procedure at blastocyst stage, therefore, no longer makes sense: Biopsying the trophectoderm, from which the placenta arises, makes no sense because the results can be discordant from biopsies of the inner cell mass that create the fetus. Moreover, if really existing aneuploidies can self-correct downstream post-biopsy and, if such self-corrections are more efficient in the embryonic (inner cell mass) than the extraembryonic cell lineage (trophectoderm), what remains the purpose of a blastocyst-stage embryo biopsy?
It is important to point out that all of these important discoveries up to this point have not been made by proponents of the procedure, who one would expect to conduct most research in attempting to validate PGS/PGT-A as a clinically valuable test. Investigators who reported these previously unknown characteristics of preimplantation-stage embryos were, like above noted British group at Cambridge University (who reported the self-correction in mice), scientists more interested in basic biology than clinical results, and opponents of the procedure/test who felt obliged to disprove the clinical validity of PGS/PGT-A because that appeared the only way to convince the field, smitten by the PGS/PGT-A hypothesis, that the procedure/test was not only of NO benefit but, in many patients, actually reduced pregnancy and live birth chances with IVF.
Proponents of PGS/PGT-A during all of these years, in often rather crass ways, have simply obstructed communications of the truth by attempting to prevent publications of critical scientific studies, while attacking the messengers, though, themselves, proving incapable of presenting even an iota of evidence in support of the continuous use of PGS/PGT-A in association with IVF.
They, of course, have tried: Whenever the evidence against PGS/PGT-A became too overwhelming, the testing industry presented a new version of the procedure/test, with new claims, yet again universally unvalidated and, indeed, mostly outright false. The inducement for this article was, however, a recent publication in Fertility & Sterility, the official organ of the ASRM, in which some of the leading proponents of PGS/PGT-A, many with considerable financial interests in the testing industry, finally at least to a degree did the right thing: quite comically calling themselves the “STAR Study Group,” this group of fierce proponents of PGS/PGT-A since its inception, for a change, conducted a properly designed prospectively randomized multicenter study that convincingly demonstrated what, very well, and of course unintentionally, could be the final nail in the coffin of PGS/PGT-A.
The most recent study by Munné et al.
(Preimplantation genetic testing for aneuploidy versus morphology as selection criteria for single frozen-thawed embryo transfer in good-prognosis patients: a multicenter randomized clinical trial. Fertil Steril 2019; in press, available at https://www.fertstert.org/article/S0015-0282(19)31979-X/fulltext).
Probably the world’s leading proponent of PGS/PGT-A after Yuri Verlinsky’s premature death, Santiago Munné, PhD, has in the past been responsible for some of the most outlandish statements in support of PGS/PGT-A, claiming utility for improved IVF outcomes, cost-effectiveness and alleged clinical efficiencies. From the very beginning, he and his consorts were almost uniformly wrong in their oral as well as published representations of how PGS/PGT-A should be performed, what results meant, and in the process, because of statistically incorrect conclusions, often misrepresenting their own data.
He and colleagues now must, however, be given credit for conducting a reasonably well-designed prospectively randomized multicenter study of IVF outcomes in relatively young women up to only age 40, in which they randomized patients to elective single frozen-thawed embryo transfer (eSET) after PGS/PGT-A or eSET after only morphological assessments of embryos.
What they found did not surprise us here at CHR because so much evidence by now is fully predictive of here reported IVF outcome data. Assuming Munné, however, ever really believed any of the hot air he over years sold to the public and IVF practitioners all over the world, these data must have been shocking for him and his colleagues.
- There was NO improvement in pregnancy rates with PGS/PGT-A over standard embryo morphology with reference point cycle start.
- There was no reduction in miscarriage rate with reference point cycle start; indeed miscarriages trended higher with PGS/PGT-A.
- The argument that PGS/PGT-A facilitates eSET was, thus, fully debunked as well. Those who favor eSET, don’t need PGS/PGT-A to maximize IVF outcomes.
- And all of these outcomes were achieved in relatively good-prognosis patients. They were young and reached blastocyst-stage embryo transfer with at least 2 embryos.
- Had poorer-prognosis patients been included in the study, overall PGS/PGT-A effects on IVF outcomes would have been negative; i.e., PGS/PGT-A, on an intention-to-treat (ITT) basis with reference point cycle start, would have reduced pregnancy and live birth rates, as Dutch colleagues already in 2007 noted in a by Munné et al highly criticized study in The New England Journal of Medicine.
Here are how Munné et al. summarized the results of their study in their recent abstract in Fertility & Sterility: “PGT-A did not improve overall pregnancy outcomes in all women, as analyzed per embryo transfer or per ITT. There was a significant increase in ongoing pregnancy rates per embryo transfer with the use of PGT-A in the subgroup of women aged 35-40 years who had two or more embryos that could be biopsied, but this was not significant when analyzed by ITT.”
Following these obviously negative results, in the discussion of the paper, Munné and his co-authors, nevertheless, had still the audacity to conclude that this study warrants continuous utilization of PGS/ PGT-A in clinical IVF practice. With a good-prognosis patient population of 661 women (average age 33.7 ± 3.6 years, when the average age of U.S. IVF patients reported to CDC in 2016 was 36 years) there was NO difference in outcomes between PGT-A and morphology. Yet, in a post-hoc sub-analysis they “discovered” that “women between 35-40 demonstrated a significant increase in ongoing pregnancy rates with reference point embryo transfer, which disappeared once the numbers were statistically correctly analyzed by intent to treat (i.e., reference point cycle start). That they even mention this post-hoc sub-analysis that used inappropriate statistical methodology is shameful (and that the editors of the journal would allow this is really inexcusable) since over 20 years all the false claims made in regard to PGS/PGT-a by Munné and other leading proponents of the procedure/test were, exactly, based on this kind of data manipulation.
It is this repeated and very obvious data manipulation that, finally, establishes this behavior pattern as willful. Giving many of these data manipulators over many years the benefit of the doubt, one now, unfortunately, is left with nothing but certainty that these data manipulations were fully intentional. How else can Munné et al., after in their study establishing beyond reasonable doubt that PGS/PGT-A does neither improve pregnancy nor reduces miscarriage rates as they have claimed for over 20 years, should be given the benefit of continuous clinical utilization?
It is this kind of statistical manipulation of data that is often not understood by IVF practitioners and the lay public and, therefore, has allowed PGS/PGT-A for over two decades to survive and prosper. Here, Munné and his cohort of investigators just are attempting to play the same game for a little bit longer by pointing out a really non-existent allegedly better clinical pregnancy rate in women 35-40 year old undergoing PGS/PGT-A, even though their own data show otherwise. Hopefully, this time the authors have, however, gone a little too far, actually hammering down the final nail in the coffin of PGS/PGT-A practice.
THERE SIMPLY IS NO LONGER ANY EXCUSE TO PERFORM PGS/PGT-A UNDER THE ASSUMPTION OF IMPROVING IVF OUTCOMES!