A small but telling study of PGT-A testing for aneuploidy embryos

Especially considering that this paper comes (with some Australian participation) out of China, where we are used to very large study participant numbers, this has to be considered a small study. But the patient population chosen for this retrospective observational study (with, therefore, obvious limitations) is interesting, offering in many ways a surprisingly well-controlled outcome experience.1

The study, involving treatment cycles between 2016 and 2023, includes 140 women with various causes of infertility who previously experienced two or more pregnancy losses (including at least one following an IVF cycle). They now underwent blastocyst warming (i.e., thawing of previously cryopreserved blastocyst-stage embryos). Among those patients, 72 had undergone PGT-A in this thaw cycle on a total of 274 blastocysts, and 74 women went directly to embryo transfer without PGT-A.

The fact that the choice for or against utilization of PGT-A was not randomized and, indeed, goes in the manuscript mostly unexplained is, indeed, the most important weakness of this paper (there are also other weaknesses). Using generalized estimating equations with some cofounding adjustments, the primary study outcome (and principal reason why this study is of interest) was the cumulative live birth rate.

Ages, gravity, and number of prior pregnancy losses did not differ between the two study groups. As one would expect, the PGT-A group, however, showed more prior aneuploidies in pregnancies (68.1% vs. 39.2%; p< 0.001). No other statistically significant differences were, however, apparent, though the PGT-A group, again as expected, had marginally more blasts cryopreserved (median 4 vs. 3; P=0.06) The study findings were as follows: 14/72 (19.4%) had no euploid embryos for transfer, in itself an important finding, considering the young age of the PGT-A group of 34.9±5.0 years. All tested blastoids, indeed, demonstrated only a 48.6% euploidy rate. The cumulative live birth rate was significantly lower in the PGT-A group (25/72, 34.7% vs. 39/74, 52.7%; P=0.048). Secondary outcomes, including individual cycle live birth rates and miscarriage rates, however, did not differ.

Here reported difference in cumulative cycle outcomes, of course, reached significance only barely, but considering the relatively small study size, this is not surprising. As already noted, this study has substantial design shortcomings beyond those already mentioned and including the long study enrollment period of seven years during which undoubtedly substantial practice changes were introduced at the IVF clinic where the study was performed. But the study is, nevertheless, informing because it reports on cumulative outcomes, and that is what, ultimately, matters in every IVF cycle – fresh or frozen-thawed.

Once an embryo cohort has been established, this cohort has, in principle, an unchangeable cumulative outcome (unless, of course, the laboratory or physicians are incompetent). Though still widely advocated by many colleagues and much of the genetic testing industry, the believe that PGT-A may influence this cumulative outcome, as in these pages repeatedly discussed before, is simply ridiculous! It is now well established – and even confirmed by a recent ASRM/SART Practice Committee Opinion2 – that PGT-A does not improve IVF cycle outcomes and does not, as shown again here, even in repeat aborters, reduce miscarriage rates to significant degrees.

All PGT-A, in the best of all circumstances, therefore, can achieve is a marginally shorter time to pregnancy. But, as discussed, the study also demonstrates that the outcome benefit is only rarely and only in the best-prognosis patients achieved. Much more frequently, as again shown in this paper, PGT-A actually adversely affects the cumulative pregnancy chances of patients. We, therefore, can only again and again repeat the next logical question: Why then perform PGT-A at all at an additional out-of-pocket expense of at least $5,000?

REFERENCES

1. Cai et al., Reprod Biol Endocrinol 2025;23:31

2. ASRM/SART Practice Committees. Fertil Steril 2024;

122(3):421-434