Cumulative LBR in PGT-A falls below that of routine IVF in new SART analysis

Published 30 January 2023

Following neutral results in several high profile RCTs, PGT-A has once again failed to demonstrate its benefit in terms of LBR, now in a new cohort analysis from the SART database. Can the results be explained by mosaicism in some embryos, misdiagnosis, or biopsy?

A retrospective cohort study from the SART Clinical Outcome Reporting System has found that PGT-A among all patients with blastocysts available for transfer or screening was associated with a lower cumulative LBR than routine IVF.(1) This negative association of PGT-A with cumulative LBR was especially pronounced in patients under the age of 35 - though was not found in those aged 40 or over.

The study analysed 133,494 autologous first cycles – with or without PGT-A - reported to SART throughout 2014 and 2015 and their linked transfers up to the end of 2016. When cumulative LBRs were compared in those with blastocysts either for transfer or PGT-A, those in the PGT-A group had a lower cumulative LBR than those in the ET alone group – and at all ages up to 40 years. For example, at age 35 cumulative LBR per cycle was 70.1% for untested embryos and 57.4% for those tested by PGT-A. However, a subgroup analysis of only those patients who had PGT-A with subsequent FET, and thus excluding those without transferrable embryos, showed a very high cumulative LBR, ranging from 71.2% at age 35 to 50.2% at age >42.

It was also noted that rates of multiple gestations, preterm birth, early pregnancy loss, and low birth weight were all greater in the non-PGT-A group. As found in other studies, there was a significantly lower rate of twin deliveries in those who had had PGT-A (in all age groups except beyond 42 years) than in those without PGT-A - although the authors do point out that at the time of data collection (2014-2017) ‘elective single embryo transfer had not yet become common practice’. The still current ASRM opinion on PGT-A notes that one ‘clinical scenario in which PGT-A may be of significant benefit is to increase utilization of eSET’.(2) The study also found a 7% early pregnancy loss rate in those having PGT-A and 12.5% in those without.

However, despite the apparent benefits in multiple pregnancy and miscarriage, these latest results, now from a large cohort analysis, concentrate on LBR and thus raise yet again the ongoing source of controversy in the question of PGT-A benefit in IVF populations. A 2021 multicentre RCT from China, in which 1212 subjects were randomised to PGT-A or conventional IVF and transfer, found no superior outcome from PGT-A, with cumulative live births occurring in 77.2% in the PGT-A group and 81.8% in the conventional-IVF group.(3) Similarly neutral results had been found in an updated Cochrane review from 2020 which concluded that no study had convincingly shown a clear benefit of PGT-A, including cumulative LBR.(4) What all these results, including several high profile trials from the past few years (including the STAR trial and ESHRE’s own ESTEEM study), appear to have in common is an apparent inability of PGT-A to improve LBRs in a routine population.

With an explanation for the ‘non-inferior’ LBR found in the PGT-A group in their RCT, the Chinese authors had two possible proposals: first, the decision not to transfer mosaic embryos (which may otherwise have developed into viable pregnancies), and second, the possibility of false positive or negative results for all embryos (‘which may compromise the effectiveness of PGT-A’).

Similarly, commenting on the inferior PGT-A results in this latest cohort study, the SART investigators also raise the controversial subject of ‘embryo loss’, possibly as a result embryo injury during biopsy, or, as the Chinese RCT suggested, embryo mosaicism or incorrect diagnosis. The non-transfer of mosaic embryos, they wrote, ‘may play a role in partially explaining our findings’ of a reduced cumulative LBR, especially in younger patients.

What about the biopsy procedure itself? A new Human Reproduction Update review - of the obstetric, neonatal and long-term health outcomes following embryo biopsy - described the effect of trophectoderm biopsy on implantation and live birth as ‘unclear’, although an RCT in 2013 had appeared to demonstrate the safety of the procedure.(5) The review authors explain that, because trophectoderm cells ultimately give rise to the placenta, ‘it is plausible that its disruption [through biopsy] might result in abnormalities in placentation’.

Although the review finds no clear evidence of obstetric or neonatal harm from biopsy, it still recommends a ‘cautious approach’ in developing a PGT strategy based on biopsy. And adds that a non-invasive approach, based on analysis of embryo cell-free DNA from culture media, might be pursued to overcome the ‘potential limitations of embryo biopsy’.

However, this review ends by once again describing PGT as ‘one of the most intensely contested procedures of reproductive medicine’. And the latest results from the SART database - albeit from data first collected almost a decade ago - can only extend the controversy. ‘We conclude that PGT-A may show utility in patients of advanced maternal age, and that PGT-A is associated with lower miscarriage rates,’ the authors ruefully end their report. In a characteristically provocative editorial in Human Reproduction following the NEJM’s publication of the Yan et al study, Gleicher and colleagues proposed that PGT-A had now reached a dead end, with the trial results ‘a final nail in the coffin of the PGS/PGT-A hypothesis’.(6)

Yet as this and other editorials noted, PGT-A continues to account for multiple cycles, especially in the USA. A report in 2022 noted that PGT in 2018 accounted for around half all ART cycles performed in the USA, with numbers apparently growing year on year despite the lukewarm advice of the ASRM and paucity of hard beneficial evidence.(2,7)

Meanwhile, PGT-A is included in the draft text of ESHRE’s forthcoming good practice recommendations on add-ons; the recommendations have concluded their stakeholder review and are now being updated.(8) The draft text covered efficacy evidence from recent RCTs (including Yan et al from 2021) and meta-analyses but concluded, ‘based on the current evidence showing lack of improvement of live birth rates, or a decrease in miscarriage, routine use of PGT-A is not recommended’. Of course, this is from the draft text and, following stakeholder comment, this recommendation may be amended. Similarly, the add-on advice for PGT-A from the UK’s HFEA remains fixed on red, with ‘no evidence from RCTs’ that PGT-A will improve the chance of having a baby.(9)

1. Kucherov A, Fazzari M, Lieman H, et al. PGT-A is associated with reduced cumulative live birth rate in first reported IVF stimulation cycles age ≤ 40: an analysis of 133,494 autologous cycles reported to SART CORS. J Assist Reprod Genet 2023; 40: 137-149.
2. The use of preimplantation genetic testing for aneuploidy (PGT- A): a committee opinion. Fertil Steril 2018; 109: 429-436.
3. Yan J, Qin Y, Zhao H, et al. Live birth with or without preimplantation genetic testing for aneuploidy. N Engl J Med 2021; 385: 2047-2058.
4. Cornelisse S, Zagers M, Kostova E, et al. Preimplantation genetic testing for aneuploidies (abnormal number of chromosomes) in in vitro fertilisation. Cochrane Database Syst Rev 2020; 9: CD005291.
5. Alteri A, Cermisoni GC, Pozzoni M, et al. Obstetric, neonatal, and child health outcomes following embryo biopsy for preimplantation genetic testing. Hum Reprod update 2023;
6. Gleicher N, Barad D, Patrizio P, Orvieto R. We have reached a dead end for preimplantation genetic testing for aneuploidy. Hum Reprod 2022; 37: 2730-2734.
7. Hipp HS, Crawford S, Boulet S. Trends and outcomes for preimplantation genetic testing in the United States, 2014-2018. JAMA 2022; 327: 1288-1290.
8. See

Get notified of new articles with our ESHRE newsletter.

Sign up and never miss an update