How often do "bad" embryos lead to "good" babies?

Fifty women decided to transfer embryos that testing had condemned. Here's what happened.

5 minute read

An older couple is having trouble conceiving. They go to a fertility clinic. Their doctor advises them to test their embryos for chromosomal abnormalities, so they do — at great expense. When the test results come back, however, the news is grim: every last embryo is deemed abnormal.

Ironically, if they had not done testing, they could have transferred any of those embryos. But their clinic's policy is to not transfer embryos known to have chromosomal problems, so, many thousands of dollars in, the couple is left without a single usable embryo to pin their hopes on.

Over the last decade or so, the use of "preimplantation genetic testing for aneuploidy," or PGT-A, has dramatically increased. In the US, according to the Society for Assisted Reproductive Technology (SART), some 32 percent of all IVF cycles used PGT-A in 2017, up from 13 percent in 2014. In Canada, according to the Canadian Assisted Reproductive Technology Register (CARTR), 26 percent of patients used PGT-A in 2020, up from just one percent in 2013.

The hypothesis behind testing is that chromosomal abnormalities are a leading cause of miscarriage and failure to implant. So if only you can figure out which embryos have these abnormalities and then exclude them, success rates will go up.

Sadly, despite all the hype and promise and expense, there is still no compelling evidence that this is true. The UK's Human Fertilisation and Embryology Authority (HFEA) still gives PGT-A a red light, because, it says, there is no good evidence that it improves the chance of having a baby. A recent paper in the New England Journal of Medicine found that for people with three or more high-quality embryos, testing did not improve outcomes.

Some doctors have been skeptical of PGT-A for some time. In 2014, the Center for Human Reproduction, a fertility clinic in New York, started cautiously transferring embryos that had been deemed through testing to be abnormal. In 2015, they reported five pregnancies. The same year, an Italian group reported six births. Since then, around the world, hundreds of babies have been born after the transfer of embryos that had previously been condemned as faulty.

Over the past few years, 69 patients have shipped 444 abnormal embryos to the Center for Human Reproduction — where, despite their designation as abnormal, they could be used — from clinics elsewhere in the US or in Europe — where they could not. Fifty women, working with the New York clinic, opted to go through transfers. The clinic followed those attempts prospectively. Last month, they published a paper in Human Reproduction about the outcomes.

The findings raise questions yet again about whether testing is leading us to avoid or even destroy perfectly good embryos, and if so, if the technology is in some cases doing more harm than good.

After five to seven days of development, an embryo has about 250 cells. When it undergoes preimplantation genetic testing, up to ten of those cells are collected using a tiny pipette. Importantly, they are taken from a part of the embryo that will go on to form the placenta, not the fetus. The cells are sent away for testing, usually at a specialized lab, while the rest of the embryo is frozen for the two or so weeks it takes for the results to come back.

The testing is looking for chromosomal abnormalities. Healthy cells will contain 46 chromosomes, in pairs, but sometimes a cell can have three copies of a chromosome, or only one.

Different labs make different calls on what the anomalies mean. Often, if fewer than 20 percent of the cells taken from an embryo have abnormalities, the embryo is designated good, or "euploid." If more than 80 percent are abnormal, it's considered bad, or "aneuploid." Anything between those two thresholds will be deemed "mosaic."

A central shortcoming of PGT-A is that it is sampling the future placenta ("trophectoderm"), not the future baby ("inner cell mass"). Even if all the cells sampled from the trophectoderm are aneuploid, it does not mean that the inner cell mass is. It does not even mean that all the cells in the trophectoderm are — a sample is just a sample.

An ongoing concern is that embryos are being wrongly labelled.

What's more, we know that aneuploid embryos often self-correct later in development. We also know that at term, placentas belonging to euploid fetuses are often mosaic.

Still, many fertility clinics refuse to transfer aneuploid or even mosaic embryos. So after doing PGT-A, many patients find themselves without any embryos to transfer.

Fifty of the 69 patients who shipped their rejected embryos to the Center for Human Reproduction, and who had no good embryos to transfer, decided to transfer mosaic or aneuploid embryos. They were advised against transferring embryos with certain kinds of anomalies — ones that could lead to known disabilities, such as Down syndrome or Klinefelter syndrome.

In all, there were 57 transfers, involving 141 embryos, the researchers report. Nineteen of the 50 patients — 38 percent — got pregnant. Eleven had miscarriages. Eight — 16 percent — had live births.

These were poor prognosis patients, say the researchers — they were old at retrieval, they had an average of three failed IVFs each behind them, and they had had low pregnancy rates and high miscarriage rates in the past. All had been advised to do egg donation. "These IVF cycle outcomes, therefore, must be considered surprisingly good," the researchers write.

The average age at retrieval for women in the study was 41 years. According to SART, in 2019, 13.4 percent of women aged 41 to 42 using their own eggs achieved a live birth after IVF (one retrieval and all transfers).

At birth, the children in this study appeared to be healthy. (One had a heart problem, since corrected, that was unconnected to the aneuploidy.)

The researchers also tested nine of the 11 "products of conception" that were lost to miscarriage. Four, they discovered, were euploid, despite having had an abnormal result from PGT-A. One was aneuploid, but not in the way that PGT-A had indicated. Just four of nine were consistent with the PGT-A diagnosis.

"This study demonstrates that, even in poor prognosis patients, transfer of chromosomal-abnormal embryos resulting in a normal pregnancy is possible," they write. "Even older women, who are often told that all their embryos are chromosomal-abnormal, still appear to have a reasonable chance of normal pregnancy.

"That conclusion alone would rescue many such women from having to prematurely advance into third-party egg donation cycles and rescue large numbers of human embryos from non-use or even disposal."

David Barad et al. "IVF outcomes of embryos with abnormal PGT-A biopsy previously refused transfer: a prospective cohort study." Human Reproduction. 2022.

Junhao Yan et al. "Live birth with or without preimplantation genetic testing for aneuploidy." New England Journal of Medicine. 2021.

Stephen Hall. "There is new hope for women told they hit an IVF dead end." The Cut. 2017.

"Mosaic embryos." HeyReprotech. 03 Dec 2019.

HFEA on PGT-A