PGT-A, the theory versus the evidence: a systematic review on efficacy, safety, and cost-effectiveness

Preimplantation genetic testing for aneuploidy (PGT-A) aims to increase live birth rate (LBR) after IVF/ICSI. While observational studies and most systematic reviews have indicated benefits with increased LBRs and reduced miscarriage rates, results from most randomised controlled trials (RCTs) have failed to show any increase in LBR when calculated per randomised patient. According to international, professional societies, PGT-A is currently not recommended for routine clinical use, still the use of PGT-A is increasing.

The purpose of this systematic review was to evaluate if PGT-A results in increased LBR in general or in specific patient groups. Efficacy is summarised in meta-analyses, and certainty of evidence was assessed. It also includes systematic reviews on possible complications for children and mothers and a health economic analysis.

A systematic search was conducted from inception to November 2024 in MEDLINE, Embase, CINAHL, Cochrane Library, PsycINFO, PubMed, and Scopus. Only RCTs where the PGT-A technique analysed all chromosomes were selected for efficacy. Concerning safety, observational studies were also included. Risk of bias (RoB) was assessed using Cochrane tools. Results from meta-analyses were presented as odds ratio (OR) and risk difference (RD) with 95% CI. Certainty of evidence was assessed using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system.

The search for efficacy and complications resulted in 9673 references. We included seven RCTs on efficacy, 55 studies on complications, and two studies on health economics. The primary outcome, live birth per randomised woman, assessed both for the first planned embryo transfer (ET) and cumulatively per oocyte retrieval, was comparable between IVF with PGT-A and IVF without PGT-A (OR 1.16; 95% CI 0.94-1.45 and OR 0.87; 95% CI 0.64-1.19) (low certainty of evidence), also for women over 35 years. The secondary outcome, live birth per ET, was higher after PGT-A (low certainty of evidence). It was not possible to evaluate if miscarriage rate is affected by PGT-A since the certainty of evidence was very low. Only two studies evaluated time to pregnancy. None of them showed a shorter time to live birth by use of PGT-A. There was no indication of an increase in complications for the child or the pregnant woman after PGT when comparing IVF with or without embryo biopsy using PGT-A, PGT-M, or PGT-SR. In a Swedish setting, adding PGT-A to IVF almost doubles the cost of an IVF treatment without any increase in live births.

The strength of this report is that the main outcome is live birth per randomized woman. Many observational studies, earlier systematic reviews/meta-analyses, as well as two RCTs using all-chromosome PGT-A, were excluded from this systematic review; they only report live birth for those women who received an ET, which gives an incorrect and too positive view. No increase in complications for the child or the pregnant woman after PGT was observed; thus, embryo biopsy seems to be safe. Since our systematic review does not show any clear patient benefit, and PGT-A almost doubles the cost of an IVF treatment, PGT-A is currently not considered ethically justifiable.

PROSPERO-CRD42024529876.