To PGS or not to PGS?

Preimplantation genetic screening (PGS) is the technique which involves the screen of the DNA of a preimplantation embryo in order to identify genetic alterations, mostly involving ploidy, in order to avoid transferring these embryos back into the mother uterus.

After e few years in which PGS had lost interest within the medical community due to the failure to show an improvement in pregnancy rates (Masterbroek and Repping 2011, Human Reproduction), we have seen its resurgence in the last 2-3 years, with new claim of efficacy which are based on the new and improved detection techniques now in place. In fact, the old and partial FISH analysis has been recently substituted by molecular biology techniques such as competitive genomic hybridization arrays (CGH), the PCR based comprehensive chromosome screening (CCS), and the newly validated next generation sequencing (NGS). All these techniques share the same background process, by which an embryo is biopsied, the genetic material is extracted and amplified in order to obtain the amount needed to run the test, and the amplified genomic DNA is then analyzed in order to find alteration which are mostly at the chromosomal level, and which for most patients are related to altered ploidy.

When you attend a scientific meeting where PGS is discussed in the context of ART, and talk with people in the ART field, you will notice that more and more clinics and hospitals are offering PGS for certain categories of patients, some going as far as including PGS in their standard protocol of care. According to the available literature, PGS is currently proposed as a usueful tool to improve pregnancy rates in three categories of patients: 1) Advanced maternal age (AMA) patients, meaning usually women older than 37-38 years old, although the limit seems to change from report to report, and it can be as low as 35. 2) Repeated implantation failures, and 3) Repeated pregnancy loss.

Although the indications reported are 3, more than 90% of patients with these indications will fall into the AMA category. According to the last ESHRE registry report (ESHRE 2009), across Europe, about 50% of all ART cycles are performed in women older than 36, which translate in several thousand treatments annually. It is therefore important that, faced with an indication for an invasive procedure which might potentially affect thousands of women annually in Europe alone, robust clinical data are collected and evaluated critically.

Unfortunately, very few randomized clinical trials are available assessing the value of PGS. In the study by Forman and colleagues, women were randomized to receive either 1 euploid blastocyst, as assessed by CCS, or 2 non screened blastocysts (Forman EJ et al, Fertility and Sterility 2013). The study showed that, although pregnancy rates were equal in the two groups, the transfer of 1 euploid blastocyst ensured high pregnancy rates with no twins. If we look closely at the inclusion criteria of this study, a few things need to be pointed out: although the abstract talks about including women younger than 42 years old, the average age of women included was 35, with almost 80% of women included younger than 37. The authors, moreover, implicitly understand the limitation of the blastocyst biopsy and transfer system even in this relatively young population, as they calculate to include about 25% more women than actually needed in the trial since they expected that many of them would not have the 2 good quality blastocysts needed for inclusion. This point should give us pause to think when we evaluate offering PGS to AMA patients, whatever the cutoff for AMA we might chose.

A second randomized clinical trial, from the same group which needs to be applauded for their effort to put rigorous testing in this controversial field, is the one of Scott and colleagues, again with CCS (Scott RT Jr et al, Fertility and Sterility 2013). In this study the researchers randomized women to receive in a fresh ET either 1 euploid blastocyst or 1 untested blastocyst. They find that the group receiving the 1 euploid blastocyst had higher pregnancy rates. This is of course good news, however, if again we look at the inclusion criteria, the women in this trail were on average 32 years old, and 53% of the candidates were excluded from the trial because of “low” ovarian reserve, meaning that they were not likely to provide the required 6-7 good quality blastocyst needed for inclusion.

No other well conducted randomized clinical trial were available to the public at the time of me writing, and therefore we must conclude that we are currently lacking the most basic evidence necessary to propose an invasive procedure to women older than 35 years old. So, when we see one of these women older than 35 in our practice, and we are asked if they would benefit from PGS, we should in all conscience say that we do not know, and as a precautionary measure, abstain from offering an invasive procedure of no recognized value for the population at hand.

Let me continue with a few more considerations. PGS, like many other diagnostic tools, does not, by its own nature, improve pregnancy rates in the sense that it will not change the nature of the embryos that we have. What PGS does is ordering the embryos from good to bad, so that we can transfer the good ones first. So, what we are really doing is shortening the time to pregnancy, nothing more. When we suggest PGS as an option, we should think that the people who stand to benefit more from PGS are those with a projected long time to pregnancy, for instance those that have lots of embryos but also a high proportion of aneuploidy embryos, such as in some cases repeat implantation failures and repeat pregnancy loss patients.

For AMA patients the issue is another, these women usually have high aneuploidy rates but few embryos. First of all, we’ll lose some embryos which will not develop to blastocyst just by waiting so that we can biopsy them, and, as numerous studies have reported, some of those embryos that we lose are indeed viable and would have produced a pregnancy (Glujovski et al, Cochrane coll 2011). Of course, biopsy-ing the embryo at day 3 is also not a good option, as we know now that this harms the embryo and diminishes its implantation potential (Scott RT Jr et al, Fertility and Sterility 2013). Then we need to have a very good freezing program, because if we lose to freeze/thaw even 6-7% of embryos, then we’ll lose 1 transfer every 15, which is not acceptable especially in patients with only 1-2 embryos available.

With the information that we have now, it might well be that we are causing more harm than good in AMA patients by proposing PGS, and as medical professionals committed to maximizing the chances of a healthy pregnancy for our patients, we should refrain from indicating this technique for them.

Rita Vassena
Scientific Director, Clinica EUGIN