play button Created with Sketch.
Deep Dive

Criticisms of PGS

PGS is a polarizing subject within the fertility field. By our estimates, it’s being used in 35% of all IVF cycles, but there are a number of clinicians who are uncomfortable with the degree to which it’s being applied. They steer their patients away from its use, and urge their colleagues do the same. Here are some of the arguments we hear brought up most often by PGS critics.

PGS Biopsy Harms the Embryo

The first argument is that embryos are sensitive and when we biopsy an embryo we run the risk of causing harm. Embryos are notoriously fragile, and even under the best conditions laboratories have trouble growing half of them to a healthy state. As one clinician put it, “if we know just exposing these embryos to air for a few moments impacts their welfare, what do you think happens when remove five to seven of their cells at a formative stage?”

When PGS was first introduced, embryologists biopsied embryos earlier in their stage of development on day three, known as the cleavage stage, and this clearly impacted the health of those embryos. As you can see below, PGS biopsies for “cleavage” stage embryos had a negative impact on implantation rates.

In recent times, the entire field has moved towards biopsying more developed embryos that have reached blastocyst stage on day five or day six. Below is data presented by Richard Scott at the American Society of Reproductive Medicine (ASRM) conference in 2016, showing that much of the disparity in implantation rates between biopsied and non-biopsied embryos has been removed. However, implantation is only one measure of an embryo’s health and we have found no long term studies yet that follow up on the offspring resulting from embryos that were biopsied.

The Assumptions PGS Is Based On Don’t Hold Up, So The Test Is Inaccurate

PGS rests on a few assumptions. The first is that there is value in taking a sample of cells from the trophectoderm, or the portion of the embryo that becomes the placenta. This is not the portion of the embryo that becomes the fetus (known as the “inner cell mass”), and so intuitively it seems less important. PGS cynics point out knowing the composition of the inner cell mass is what should matter, and that that there is data showing samples from the trophectoderm doesn’t reflect the inner cell mass.

PGS bulls fire back with two datapoints. The first is that there is data that the trophectoderm and inner cell mass do in fact resemble each other, though this data was collected with an inferior technology nearly four years ago. Second, when investigators discovered that samples of the trophectoderm and inner cell mass were in fact different, the quality of of the trophectoderm was still incredibly predictive of an embryo’s ability to deliver a healthy pregnancy.

A second assumption is that when we biopsy the trophectoderm, the sample we recover is representative of the whole trophectoderm. PGS cynics point out that in many circumstances the samples we recover are “mosaic,” and reflect a trophectoderm that has both normal and abnormal cells. Ultimately this raises the concern that what we are sampling is more complex than we first assumed, and thus our samples are less likely to reflect the true make-up of the trophectoderm. When we take a sample that looks normal, have we merely sampled a patch of the trophectoderm that is normal, and missed patches that are abnormal? Or vice versa?

False Positives Leads to Discarding Embryos That Could Be Healthy Babies

Critics point to mosaicism, and both the field’s historic inability to detect it, and the reluctance to put mosaics in their own category, to say that the results of PGS have led us to discard embryos that could have resulted in healthy pregnancies.

Mosaicism is a complex topic, and we have a more complete article on it here. But high level, while mosaics don’t make babies nearly as often as purely normal, euploid embryos, they do have some potential. About 15 – 25% of moscaic embryos turn into live births, so their potential isn’t nothing, in contrast with purely abnormal, aneuploid embryos.

Historically, the technologies used to run PGS analysis did not detect mosaics, so embryos were only classified as euploid or aneuploid. In that case, there may have been false positives, where embryos were called aneuploid, and discarded, when they may have really been mosaic and had some potential. By some estimates, 20% of embryos are mosaic.

Still, today, some labs do not report which embryos are mosaic, and some doctors do not tell patients if they have mosaic embryos, instead classifying many mosaics as aneuploid. This raises concerns for many experts, who believe that patients have the right to know, and that black-and-white classifications lack precision and lead to discarding embryos that could be a patient’s only chance of a healthy pregnancy in some cases.

This may be especially true in older, poor-prognosis patients who ultimately placed faith in PGS readings, decided not to transfer any of their apparently abnormal embyros and missed out on the chance to become a biological mother. How frequently might that be happening? At the moment, we cannot say, but clearly after PGS testing, over half of older patients are electing not to transfer (though remember, aneuploidy increases significantly with age, so the chances of a viable embryo decreases with age).

PGS bulls disagree with this for two reasons. First, some clinics who do PGS most often suggest that mosaicism rates are closer to 4 – 10%, though the data on that is sparse. Second, they point out that using PGS for this older, poorer prognosis population is important because it clearly helps lower failed transfer and miscarriage rates, which are excruciating regardless of one’s age.

Finally, PGS proponents argue that the advent of “mosaicism” is a sign of progress: the technologies are improving, we’re fine-tuning our ability to discern between good embryos and bad embryos, and this is exactly the wrong time to dismiss the test. Mosaic embryos have limited potential compared with euploid embryos, so recognizing when an embryo is mosaic further helps doctors select the best embryos to transfer.

PGS Doesn’t Make Sense for Younger Patients

Some critics argue that PGS is not only unhelpful to the older, poor prognosis patients, but it’s also unhelpful to the younger, better prognosis patients. These patients theoretically have more eggs, more embryos, a higher mix of chromosomally-normal embryos and thus their doctor’s chances of selecting a good embryo are high without PGS. Most of these PGS cynics champion a study out of Cornell that clearly shows patients who avoided using PGS had similar outcomes to those who elected to use it.

PGS bulls have many issues with the above. First, they disagree with the set-up of the Cornell study (the PGS patients were much harder to treat, the sample size was small) and its conclusion. While young patients have a higher proportion of good embryos to choose from than older patients, 30 – 50% of their embryos are still likely to be unhealthy and lead to a failed transfer or miscarriage. Thus PGS still has an important role to play in this patient population.

PGS cuts down the risk of miscarriage in younger patients, not to the degree it does in older patients, but still by a meaningful percentage (from 15% down to 10%). For many patients, that is substantial given the emotional and financial setbacks involved with a miscarriage.

PGS Rates Are Overstated

PGS skeptics believe advertised PGS success rates mislead patients. Patients often hear “PGS-normal embryos have a 60 – 70% success rate.” But that is on a per-transfer basis. Meaning that if you begin a cycle, retrieve eggs, produce embryos, then do PGS testing, and at least one embryo comes back normal, 60 – 70% of the time it will lead to a live birth. So the equation assumes you have good embryos to transfer which requires passing over several big hurdles. Many patients don’t have good embryos, so they don’t make it to the point of transfer. On a “per treatment” basis, PGS success rates are much lower than 60 - 70%.

When one looks at all patients who started a cycle, and compares the ultimate results between those who opted to do PGS and those who did not, PGS results still look favorable but doesn’t match up with the 60 – 70% success rate statistics that are stuck in most patients’ heads. Below you can see PGS, and non-PGS, cycle starts that culminated in a live birth.

The way to read this chart can be confusing. “First transfer” refers to whether a live birth occurred after the first embryo transfer. If that transfer failed, and the patient elected for another transfer (which does not always happen), the “subsequent transfer” rate reflects if that next transfer resulted in a live birth.

PGS Isn’t Worth the Costs

A cycle of IVF costs around $23,000, and since most patients cycle multiple times, the average patient who achieves a successful outcome spends $66,000. PGS itself accounts for 20 – 25% of the cost of an IVF cycle, so many anti-PGS arguments are made in the context of its economic value.

The crux of these arguments is that while PGS does improve success rates per transfer, it doesn’t necessarily improve overall cycle success rates, and doesn’t justify the price tag. You’ll notice in the chart above that patients under the age of 35 had similar live birth rates regardless of whether they did PGS.

Cynics argue that the cost of PGS testing is the same or higher as a frozen embryo transfer (which is about $3,000), so by foregoing PGS patients can put that money towards a second transfer if the first transfer fails.

PGS proponents have multiple issues with this argument. First, they point out that for most patient populations, use of PGS cuts down the rate of miscarriage, and one cannot ascribe a value on that. For many patients, a miscarriage is devastating and many would happily pay to avoid it.

Next, PGS proponents point out that use of PGS helps avoid fruitless transfers (which is why PGS success rates are so good on a per transfer basis). This has two benefits: first patients save the money they would have spent on a failed transfer (again, a transfer costs $3,000 and PGS costs $5,000). Second, patients don’t waste time undergoing, and recovering from, the failed transfer. If that patient needs to cycle again to get better embryos, the younger they do it (even by a matter of months), the better.

Of course, every patient is keenly aware of time when undergoing fertility treatments, and the quicker time to pregnancy with PGS is another benefit.

Finally, many PGS cynics assume after a failed transfer, a patient will come back for another. PGS proponents point to data that shows IVF patients often don’t return after failure, and so getting a good outcome as early as possible is crucial, especially for patients who don’t have the emotional or financial capacity to return after a failed transfer.