Insights into the utility of preimplantation genetic testing from data collected by the HFEA

The recent report [1] in this journal of an audit of data collected in the United Kingdom (UK) by the Human Fertilisation & Embryology Authority (HFEA) (www.hfea.gov.uk) concluded that there was strong evidence for the benefits of preimplantation genetic screening for aneuploidy (PGT-A). The authors comment on the need for data for individual clinics and the purpose of this communication is to address that need and to offer further insight into this topic.

The individual clinics audit (see Supplementary electronic material) included cycles carried out in the UK between January 2016 and the end of December 2018, the most recent HFEA three-year aggregate data available for individual fertility clinics which includes live birth events. A search of the public A–Z clinic list (https://www.hfea.gov.uk/choose-a-clinic/clinic-search/) was made to identify all clinics with PGT-A and PGT-M/SR activity (monogenetic disorders and structural rearrangements, which are not distinguished by the HFEA; PGT-A is recorded as a separate activity) using the patients’ own eggs in this period. Summary PGT and IVF plus ICSI data were stratified by maternal age (less than 35 years vs. 35 years and older). Due to the way the HFEA collects cycle activity with genetic testing, the number of PGT cycles abandoned before egg collection is not known. The HFEA reports frozen cycles as those with an attempted thaw. For the freeze-all approach, this means that the number of stimulated cycles is not available and the reasons for abandonment prior to an attempted thaw cycle are unknown. The cancellation of an attempted freeze–thaw cycle is mostly due to thaw failure. The reasons for treatment are unknown and the specific testing methodology is not provided. Other caveats might be that PGT is offered to women of better prognosis (e.g. centres may have criteria regarding a minimum number of eggs to qualify for PGT-A), and the extent to which embryos are pooled from more than 1 stimulation cycles prior to a transfer cycle and clinical pregnancies are lost to follow-up is unknown. The substantial minority of fresh cycles reported indicates that clinics were still in transition to a freeze-all strategy for PGT during this period. For the subset of clinics included in the analysis, the outcomes from fresh IVF/ICSI embryo transfers and the cumulative live birth rate from all the fresh and frozen embryo transfers within two years of one 2016 egg collection are also available. Additional information is provided on the proportion of single (elective and non-elective) and multiple embryo transfers for PGT-A and IVF/ICSI for these clinics.

For the analyses presented here, the primary outcome measures are the number of live born infants (the sum of singleton and twin deliveries) from the number of embryos transferred, and the putative clinical miscarriage rate represented by the deficit between the number of clinical pregnancies (an intrauterine gestation sac that contains a foetal pulsation) and the number of live born deliveries. These analyses do not address the question as to whether a woman is more likely to take home a baby starting a full cycle but offer insight into the capability of testing for aneuploidy to differentiate viable and non-viable embryos.

• i.The PGT-A analysis pertains to 7 centres likely to have used blastocyst biopsy with next-generation genetic sequencing (NGS) for > 50 vitrified-warmed cycles and compared with IVF/ICSI outcomes for those centres. Taken at face-value, PGT-A has potential utility, at least for women > 35 years, to differentiate a viable and non-viable embryo (Fig. 1) and to mitigate the risk of clinical pregnancy loss (Fig. 2). The heterogeneity statistic indicates consistency between these clinics. For younger women aged less than 35 years, the evidence is less convincing (pooled RR 1.12 [1.012–1.241] P = 0.035; heterogeneity P = 0.392) (see Supplementary data material).

Fig. 1.

PGT-A vs. IVF/ICSI freeze–thaw embryo transfers: individual centre risk ratios for live birth. Centre 0006 reported only fresh PGT-A cycles during this period

Fig. 2.

PGT-A vs. IVF/ICSI freeze–thaw embryo transfers: individual centre risk ratios for putative clinical miscarriage

Centre 0006 reported only fresh PGT-A cycles during this period. From 101 oocyte retrievals (10 < 35 years, 91 35 + years), 49 (49%) were cancelled, 29 due to all embryos testing positive and 20 for other reasons not specified. Comparing PGT-A vs IVF + ICSI fresh cycles for women 35 years or older, there were 19% (17/91) and 22% (992/4489) babies per oocyte retrieval (RR 0.845 [0.549–1.302] P = 0.427).

For younger women aged less than 35 years, there was no evidence that testing mitigated clinical pregnancy loss (pooled RR 1.093 [0.776–1.539] P = 0.617; heterogeneity P = 0.111) (see Supplementary data material).

• ii.The PGT-M/SR analysis pertains to a comparison of 2 centres: 0102 is known to have tested only for chromosome rearrangements using array-comparative genomic hybridization (aCGH), and only monogenic conditions using preimplantation genetic haplotyping (PGH) [2] (Prof C Mackie Ogilvie, pers. com.); 0044 is known to have combined testing for unrelated chromosomal abnormalities using aCGH and Karyomapping [3].

Taken at face-value, PGT couples have a better prognosis than IVF/ICSI couples (they often do not require assisted conception as such), and testing for unrelated aneuploidy is unlikely to improve a transferred embryo’s chance of success. The significant heterogeneity (Fig. 3) reflects the superior live birth rates obtained by centre 0044 (PGT and IVF/ICSI), which are also materially superior to the national averages (see Supplementary data material). Debate is ongoing concerning the possible detrimental effect of biopsy on embryo viability and PGT-A [4]; since biopsy is already necessary for PGT-M/SR, a non-selection study (post hoc analysis of unrelated chromosomal abnormalities once treatment is complete) as an experimental study could be informative for testing (Fig. 4).

Fig. 3.

PGT-M/SR vs. IVF/ICSI freeze–thaw embryo transfers: individual centre risk ratios for live birth

Fig. 4.

PGT-M/SR vs. IVF/ICSI freeze–thaw embryo transfers: individual centre risk ratios for putative clinical miscarriage

It has been shown previously that a superior live birth rate per embryo transferred should not be expected to translate into a superior cumulative live birth rate for a full cycle when all embryos with an abnormal test result are excluded from transfer [5]; however, an inferior cumulative live birth rate could be avoided by using testing to determine only the order for transfer and accepting a marginally smaller reduction in the risk of an adverse pregnancy [6].

I agree that it would be helpful for the HFEA to review which data are collected and how they are reported. In particular for the freeze-all strategy, we need to see the outcome from all transfers within two years of 1 egg collection. National and international data collections are useful; however, they will not obviate the need for well-designed clinical studies with full reporting of the data to allow unbiased assessment of the costs and benefits of preimplantation genetic testing.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

The author is responsible for the content and writing of the article.

Availability of data and material

Supplementary file provided.

Code availability

Not applicable.

Declarations

Ethics approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Conflict of interest

The author declares no competing interests.