Richard J. Paulson

Professor, University of Southern California
  • University of Southern California
  • United States of America

Recent Comments

May 10, 2018
Replying to Robert Wagner

I am a molecular geneticist and am somewhat confused by your paper.  You seem to assume that PGS finds "abnormal" embryos only from the population of embryos that will not implant in any event (which would appear to make PGS completely useless).  I am unaware of any evidence supporting that assumption.  Is there any?  In fact, isn't the finding that the implantation rate of screened vs unscreened embryos is basically equal as likely to mean that PGS takes equally from both of your red and blue groups?  In addition, don't we already know that embryos with severe genetic abnormalities (e.g., trisomy 21 - Down syndrome) successfully implant and lead to full term pregnancies? And aren't those (red group embryos) precisely the kinds of embryos that PGS is intended to detect?

Thanks so much for your comments! The one non-selection study performed to date (Scott et al, Fertil Steril 2012;97:870) showed a false positive rate of about 10% (41% implantation rate in "euploid" embryos and 4% implantation rate in "aneuploid" embryos. That's not where the lost embryos are going to. The point is that when you test embryos to see which are less likely to implant and then remove those from the population, the remaining embryos should logically implant at a much higher rate. The fact that they don't implies that the biopsy process itself causes sufficient damage to decrease the implantation rate back to the baseline observed before the testing was done.

May 10, 2018

Thanks, Micah! It definitely adds information. Whether a lab adds that information may depend on whether it costs $1,000, $10,000 or $100,000.

Aug 15, 2017
Replying to Alexander Quaas

This thought-provoking inklings piece provides an interesting perspective on the efficiency of PGS, including the potential discarding of normal embryos.

While the widespread use of PGS is a relatively new phenomenon, choosing embryos for transfer by other means was always necessary. Could the same model presented not also be applied to the selection and discarding process used based on embryo morphology? 

Or do you believe that the embryo loss rate with morphology based selection is 0%? Meaning no embryos with the potential to implant are ever discarded?

Otherwise wouldn't it be conceivable that PGS could "rescue" embryos that would have been discarded by morphology (for example CC grade) but turn out to be euploid? If that were the case then this phenomenon would have to be factored into the equations and figures presented.

In the field of clinical medicine, mathematical modeling will always be just an approximation. Your point is well taken, there are indeed other criteria that are being used to decide when to discard embryos.

But in the clinical practice of PGS, I would argue that morphology is still being used to decide about the fate of embryos, in conjunction with PGS, rather than PGS being used to "rescue" bad-looking embryos. For example, a poor quality blastocyst may not be "good enough for biopsy." Additionally, when multiple euploid embryos are available for transfer, the morphologically best embryo is generally transferred. If multiple euploid blastocysts are cryopreserved, 2 or 3 may be thawed, and then the best one transferred to maximize the chance of implantation.

This is a strategy that maximizes the per-embryo implantation rate, and the argument for this approach is that it minimizes the time to pregnancy. It is the opposite of the "every last baby out of every last egg" approach that I prefer, especially in women over 40.

Most practitioners think that they are losing 5-10% of potential implantations when they use PGS. The point of my inklings piece was to point out that even a very good increase in implantation rate would be associated with a 20% loss of potential implantations. Since most programs are not seeing an appreciable increase in implantations among women under 35, this represents approximately a 40% loss of potential implantations.