About a decade ago, we, (Sher, and Keskintepe, L; et al) introduced full chromosomal karyotyping for preimplantation genetic sampling – (PGS) to assess egg/embryo “competency” for IVF. We performed and reported on 3 separate studies that served to herald transformation in the field and provided an opportunity to positively impact egg cryostorage for fertility preservation (FP) and donor egg banking.
- 2007: Using metaphase comparative genomic hybridization (mCGH) we assessed the chromosomal integrity of 1st polar bodies (PB-1) biopsied from human donor eggs. We then fertilized normal (euploid) eggs with donor sperm and transferred up to two resulting blastocysts to recipients. We reported a viable pregnancy rate of > 50% per embryo transfer following the transfer of up to 2 advanced embryos (blastocysts). These rates were significantly higher than national average results for comparable patients, reported by SART for the transfer of non-tested embryos.
- 2008: We biopsied single cells (blastomeres) from 5-9 cell day-3, cleaved embryos for mCGH testing and then allowed the embryos to continue in culture to day 5-6 and vitrified those that survived to the blastocyst stage. The transfer of up to two (2) euploid blastocysts in a subsequent cycle (“Staggered IVF”) resulted in a similar viable pregnancy rate per embryo transfer as that which we reported in our first study in 2008:
- 2008: Finally, we improved the IVF birth rate per frozen egg from a reported 4-6% (when no PB-1 karyotyping was done) to 28%, when only mCGH-normal/euploid frozen eggs were thawed, fertilized and up to two resulting blastocysts were transferred.
This break-through soon spawned rapidly growing interest in PGS with full egg/embryo karyotyping and heralded the onset of an era where this form of embryo selection would become inculcated into the IVF diagnostic armamentarium.
At first, virtually everyone performed PGS full karyotyping on single cells derived from eggs or day-3 cleaved embryos. However, since the performance of mCGH is a labor-and cost-intensive exercise which requires several days to perform, it soon became apparent that the use of this methodology was cost prohibitive. A more cost-effective approach was sought, leading to the replacement of mCGH with array-CGH (aCGH) where the process took only a few hours and the analytics was abbreviated and much easier.
We had been reporting excellent results with mCGH, performed onsite both on PB-1 and on day-3 blastomeres and where the amplification process was conducted onsite, ASAP following performance of the biopsies. So, being firm believers in the adage that “if it ain’t broke, don’t fix it”, we elected to continue using mCGH for PB-1 and day 3 blastomere embryo karyotyping. However, given that performance of mCGH testing, onsite in our own genetics laboratory soon became cost-prohibitive, we negotiated an exclusive arrangement with a leading genetics center of excellence in Turkey for a lower fee and thereupon began outsourcing all biopsied specimens (derived from eggs and embryos) for amplification, hybridization and analysis.
An initial study using aCGH reported from a large and reputable center in Colorado where lower cost aCGH analysis was used for testing day-3 embryos, produced results and a degree of reliability that was comparable to what we had reported on using mCGH. It resulted in this more cost-effective aCGH process becoming recognized as the PGS “gold standard”. Soon more and more IVF centers began sending day-3 biopsy specimens to remotely located genetics laboratories that sprung up all over. But unfortunately reported results did not match the favorable results previously reported by us in Las Vegas and from Colorado, leading geneticists to believe that the this discrepancy was likely due to the fact that a single cell (blastomere or PB biopsy) did not generate enough DNA (about 20pg per cell) to permit reliable CGH analysis and that much more DNA (2-3 times the amount) was required to perform the assay reliably It was proposed that this would require 4-10 cells and thus that the material would need to be obtained through biopsying the trophectoderm (TE) of blastocysts..
Soon TE, blastocyst biopsies started to replace day 3 biopsies. However, it soon became apparent that even multicellular TE biopsies failed to significantly enhance reliability of the PGS results reported. IVF outcome statistics simply did not improve across the board. Moreover, the examination of miscarried concepti resulting from the transfer of PGS-tested embryos often showed discordance between result reported through aCGH and those found through other tried and tested standard genetic methods. All this soon led to a growing lack of confidence in the reliability of PGS testing using aCGH performed on TE.
Citing a possible flaw in the aCGH process as being responsible for the discordance, experts began to advocate alternative PGS testing methods such as SNP-array, and next generation gene sequencing –NGS (among others). But even this failed to significantly improve the predictive reliability of PGS testing, continued to cast suspicion on the validity of previous claims made regarding the reliability and value of PGS embryo testing. To make matters worse, recent reports suggest that younger women might not benefit from PGS testing, proposing that such testing be limited to older women, and those with repeated IVF failure or recurrent pregnancy loss.
The fact that we in Nevada had noted a drop off in our own success using mCGH following the outsourcing of all mCGH testing that coincided with our outsourcing of all testing to a laboratory in Europe, alerted us to the possibility that the belief that the deteriorating results with PGS on the basis of the technology being flawed, was probably wrong. Was it possible that differences in performance and reliability could be due to the biopsied DNA specimens becoming corrupted prior to the analytic process? Could it be that embryo DNA biopsies do not travel well and are damaged in transit to a remotely located genetics laboratory? In addition, could the freezing of biopsied DNA in order to prepare them for remote transportation in and of itself have a damaging effect and could a delay in initiating amplification also lead to corruption of the DNA?
Notwithstanding the above, it is a fact that accurate PGS is probably best performed using NGS. The question however, is whether it is preferable to perform the test on a single cell biopsy on CD 3 or is it preferable to biopsy TE derived from a day 5-6 blastocyst. Clearly biopsying a multicellular (>100cells) blastocysts by allowing access to several cells at a time, has the advantage of providing access to much more DNA than when a single cell (blastomere) is extracted from a 5-10 cell day 3 embryo. But does this enhance the reliability of PGS. I know that most would say that it does …but I am unconvinced. Consider the fact most numerical chromosomal aberrations (aneuploidy) involving day- embryos originate during maturational division (meiosis) in the egg and that many such meiotic aneuploid embryos do not survive to the blastocyst stage. This is because all the embryo’s cells are aneuploid. Thus, embryos affected by meiotic aneuploidy are permanently “incompetent”. They cannot recover. In contrast, aneuploidy can also result from replication errors during repeated cell division that follows fertilization (i.e. during mitosis). In such cases some of the embryo’s cells will be aneuploid while others will be normal. This is referred to as “mosaicism”. Mosaicism is potentially reversible (autocorrected) while meiotic aneuploidy is not and there is no known method whereby karyotyping of single cells can differentiate between meiotic and mitotic aneuploidy (mosaicism).
I simply do not buy into the argument that mosaicism is more prevalent in early embryos than in blastocysts. After all, mitotic cell replication is far more advanced in blastocysts than in day 3 embryos. Thus the opportunity is greater for mitotic errors to occur in blastocysts than in a day-3 embryos. Thus in my opinion, the closer to fertilization that the embryo biopsied is performed, the lower the likelihood of encountering mitotically aneuploid cells.
Clearly, unless the above mentioned issues are investigated and resolved soon, the entire concept of PGS embryo selection will likely soon fall into disrepute. That would indeed be a pity because we know that if conducted responsibly and selectively, such testing can have great benefit. We MUST act fast and act NOW to avoid this from happening.
In summary, while I do concur that blastocyst biopsy is a reliable method by which to access DNA material for PGS testing, I strongly suspect that the reliability of PGS embryo karyotyping has less to do with the method of analytics employed (mCGH, aCGH, SNParray, NGS) and more to do with preservation and preparation of the biopsy specimen for analytical processing.