The PGS Debate Continues

Because potentially most damaging to many IVF patients, likely the most important issue is the subject of preimplantation genetic screening (PGS), which has been very frequently addressed in these pages.  CHR has, of course, for years argued that the effectiveness of PGS in improving IVF outcomes (i.e. live birth rates) and decreasing miscarriage rates, claimed by some, is a mirage based on inappropriate statistical outcome assessments. More recently, we also have been able to explain why PGS, likely, can’t deliver on these promised outcome improvements by demonstrating that the technique of trophectoderm biopsy at blastocyst stage of embryos is highly inaccurate, leading to many false-positive diagnoses and, therefore, to the mistaken discarding of many embryos with normal pregnancy potential.

voice_1511_blastocystAfter reporting the first chromosomally normal births from transfer of allegedly chromosomally abnormal embryos here at CHR, at last count 17 healthy children have so far been delivered from transfers of such embryos here in the U.S. and in Europe. More importantly, not a single chromosomally abnormal pregnancy has been reported after such transfers. There is no stronger evidence for the inadequacy of PGS and, indeed, for the potential damage to many patients who undergo PGS because, by discarding false-positive, allegedly chromosomally abnormal embryos, they may be discarding the patients’ last pregnancy chances.

Damages to such patients are further aggravated when, as is often the case, colleagues then conclude that such patients should advance into egg donation since they no longer are able to produce chromosomally normal embryos. Such patients, therefore, not only were deprived of transfers of perfectly normal embryos but, because of that false-positive PGS diagnosis, also end up prematurely being pushed into egg donation.

New recommendations from the PGDIS

With so many healthy children born in a number of centers around the world after transfer of allegedly chromosomally abnormal embryos, even the most ardent proponents of PGS had to acknowledge that their prior claims no longer held water. The so-called Preimplantation Genetic Diagnosis International Society (PGDIS), a “club” of PGS laboratories and PGS proponents, therefore, on July 19, 2016 issued in their newsletter a “Position Statement on Chromosome Mosaicism and Preimplantation Aneuploidy Testing at the Blastocyst Stage.” By issuing brand new “recommendations for PGS laboratories,” new “recommendations for clinicians” and “guidelines to prioritize mosaic embryo transfer,” the society basically implicitly refuted all of their own prior recommendations how PGS should be performed.

The only aspect of this announcement more surprising than that it was even published was its content. It, even more blatantly than in earlier misrepresentations from this society, demonstrated to what degree the PGS industry has been driven by economic rather than scientific considerations. And here is why: In their revised recommendations for laboratories, the society, suddenly, discovered that “only validated Next Generation Sequencing (NGS) that can quantitatively measure copy numbers should be used for measurement of mosaicism in biopsy samples. Ideally, that should be NGS methodology that can accurately and reproducibly measure 20% mosaicism in a known sample.”

Though, of course, never spelled out in the newsletter, this implies that PGS laboratories, which up to now used other platforms than NGS, or utilized NGS platforms with even higher than 20% sensitivity levels for mosaicism, have for years been reporting highly inaccurate results, likely leading to the discarding of large numbers of perfectly normal embryos. In the U.S. a majority of PGS laboratories for the longest time did not use NGS platforms, including some of the largest-volume laboratories. Some of these laboratories also generated the most widely cited papers supportive of PGS in the literature.

In the next paragraph, the new PGDIS recommendation for laboratories also explains why the organization, suddenly, recommends the exclusive use of a NGS platform with minimal 20% sensitivity threshold for mosaicism. The answer is that the PGDIS, simply, decided to establish brand-new criteria for what represents chromosomally normal and abnormal embryos: Under these new criteria, an embryo is chromosomally “normal” as long as a single trophectoderm biopsy (of at least 5 cells) demonstrates less than 20% aneuploidy (chromosomally abnormal) cells. Following this definition, a 5-cell biopsy really would have to be entirely normal because even only 1 abnormal cell would exceed that threshold. But if that biopsy contains 6 or 7 cells, then one abnormal cell in that biopsy would also still be considered “normal.

Such a definition, however, neither makes logical sense nor is it based on scientific evidence. Indeed, there are no data in the literature to suggest that below 20% cell mosaicism in a single biopsy denotes a chromosomally normal embryo. The reason why the PGDIS established this cut off is simply technical: Even best current NGS platforms do not have the sensitivity to detect mosaicism levels below 20%.

These new recommendations are, however, even more bizarre, because if the NGS platform detects between 20% and 80% abnormal cells, this embryo is now considered “euploid-aneuploid-mosaic.” Imagine that, the same society, which up till now recommended the discarding of embryos with any chromosomal abnormalities, now under these new recommendations, isn’t even ready to declare an 80% chromosomally abnormal biopsy as aneuploid “abnormal.”

We have no intention of suggesting that an 80% abnormal biopsy is suggestive of a chromosomally abnormal embryo; all we want to point out is how radically the PGDIS has shifted position, without even saying mea culpa for misrepresentations to patients as well as physicians in the past.

The most remarkable aspect of this recommendation is, however, that the 20-80% range for definition of a “euploid-aneuploid mosaic” embryos is, once again, being pulled out of thin air. There is no data to support such a range; indeed, CHR investigators just submitted a manuscript for publication in which mathematical models demonstrate that a single trophectoderm biopsy really does not have the statistical power to determine the status of an embryo.

Similarly arbitrary is the new recommendation of the PGDIS of what represents an “aneuploid” embryo, which the society now defines as over 80% aneuploidy in a biopsy. Imagine, at 80%, an embryo is “euploid-aneuploid-mosaic,” and can potentially be transferred, but at 81% mosaicism, an embryo is “aneuploid” and, therefore, should be discarded. Nobody can take such a classification seriously!

In a remarkable turnaround, considering how severely some key members of the PGDIS criticized CHR when we first published the center’s then new policy of offering transfers of selected “aneuploid” embryos to properly counseled patients, the society now formally endorses such transfers. Indeed, it endorses such transfers implicitly at low mosaicism rates (below 20%) by classifying such embryos automatically as “euploid-normal;” but it also endorses potential transfers of “euploid-aneuploid mosaic” embryos up to 80% aneuploidy and, indeed, like CHR did in its initial announcement, suggests a hierarchy of transfer, depending of detailed chromosomal findings in embryos.

While we agree with a hierarchy of transferrable embryos (i.e., best embryos should always be transferred first), we consider PGS, in its ability to determine an embryo’s complete chromosomal status, such a flawed procedure that, as of this point, in absence of completely normal embryos, we, likely, would be ready to transfer a large majority of embryos containing abnormal cell lines.

We were also struck by how self-serving to the PGS industry one of the society’s recommendation to clinicians was: In addressing “the possible transfer of embryos with only mosaic aneuploidies,” the first recommendation the PGDIS suggests for patients in such a case is not transfer of mosaic embryos but “a further cycle of IVF with (repeated) aneuploidy testing to increase the chance of identifying a normal euploid blastocyst for transfer.” The “chutzpah” of this recommendation is really difficult to fathom: After getting a worthless uninterpretable result, let’s do it all over again, incurring not only the cost of yet another PGS procedure but also of a complete second IVF cycle!

In summary, even the most interested proponents of PGS in these new PGDIS recommendations acknowledge how useless and, indeed, misleading their past diagnostic approaches to PGS have been. Yet, still, even in these new recommendations, they fail to acknowledge that, in early stage embryos, a single biopsy of a few trophectoderm cells cannot accurately enough reflect the chromosomal status of an embryo. PGS, therefore, clinically is a worthless procedure because a single biopsy can never determine whether an embryo can be transferred or should be discarded.

The new PGDIS recommendations quite obviously do not go far enough, because, considering what is now known, and assuming that the PGDIS considers the best interest of patients their first priority, their new recommendations should have been that, until further studies demonstrate otherwise, PGS should no longer be utilized in attempts to improve live birth rates in IVF and/or reduce miscarriage rates. And, before we forget, all of those who in the past mistakenly have been pushing hard for PGS, owe their patients a sincere mea culpa. It is one thing not to know better; to continue on the same track once things have become obvious, however, carries a very different connotation!

This is a part of the September 2016 CHR VOICE.