PGS: select or reject? - it's not just about improving pregnancy rates

Ten years since the first reports of PGD for abnormal chromosome number, or aneuploidy, (now commonly known as PGS - preimplantation genetic screening), we are experiencing a prejudicial mistrust of a clinical approach that is generally accepted to be scientifically and clinically sound. Opponents of PGS frequently criticise its overselling, overuse and indiscriminate application as a means to increase pregnancy rates (particularly in the US). We wish to respond to some of the largely irrational claims made by opponents of PGS and suggest a broader perspective on its use, in the hope of better informing the people for whom this really matters - our patients.

Few disagree that the premise underpinning a screening test for chromosomal aneuploidy in human embryos is scientifically and clinically sound. For women of advanced maternal age, this is particularly relevant, since high levels of uniform aneuploidy have been confirmed in their embryos. What is contentious, is the exact method needed to accurately, reliably and comprehensively diagnose aneuploidy in single embryonic cells. Whether it be the precise chromosomes to identify or the number of cells to biopsy, the devil is always in the detail.

Those involved in delivering a PGS service will know that a degree of technical error and embryo mosaicism [in which cells taken from the same embryo have a different gene make up] are part of the landscape of single embryonic cell tests. However, these are both measurable limitations. We are one of the few centres to provide follow-up diagnosis of untransferred embryos, allowing us to provide quality assurance (in the form of accurate false negative and false positive rates) and clinical closure for some patients. Indeed, for some patients, the information gained from whole embryo chromosome analysis is more useful than the single cell result. With patient consent, follow up of whole embryos allows us to categorize embryos as chromosomally chaotic (negligible viability), uniformly aneuploid (resulting in miscarriage or affected children as a result of meiotic errors in sperm or eggs), or to evaluate the degree of chromosomal mosaicism. This is turning out to be a key piece of diagnostic information, since a rule of thumb indicates that a small degree of mosaicism in an otherwise chromosomally normal embryo would likely result in a 'normal' pregnancy.

A recent Cochrane review of two trials meeting the inclusion criteria concluded that insufficient data were available to determine whether PGS is an effective intervention in IVF/ICSI for improving live birth rates (Twisk et al., 2006). In many studies not included in the Cochrane review, outcome measures were improved but not significantly so - a fact which could indicate insufficient power (number of cases) or reflect the variability between patients.

IVF is a necessary prerequisite for PGS, and it is not cheap. The addition of embryo biopsy, further culture and single cell diagnosis further adds to this cost. But as the NHS implements NICE guidelines to provide more couples with free IVF, the price of IVF and PGS will inevitably be driven down by competition for contracts. The answer to the claim that PGS is unnecessary lies in its appropriate use. Tell that to patients who have had repeated miscarriage, repeated IVF failures and need some answers, or to older patients who desire rational closure to their treatment. In experienced hands, embryo biopsy is not harmful, and at worst is no more harmful than the hundreds of thousands of freezing and thawing procedures carried out on embryos worldwide each year. All studies demonstrate that pregnancy rates following biopsy and PGS are at least as good as control groups, despite embryo biopsy. Unfortunately, despite a lack of evidence of harm, this myth is still perpetuated in some clinicians' letters to patients.

Most laboratories use a simple grading scheme for selecting embryos for transfer to the uterus. In a regulatory environment moving inexorably towards single embryo transfer, we have a duty to avoid the transfer of chromosomally abnormal embryos - not necessarily to improve pregnancy rates, but to avoid miscarriage and children with abnormalities. The identification of a group of embryos with uniform chromosome abnormalities may not only provide closure to the patient in terms of using her own eggs, but also prevent future fruitless IVF cycles at additional expense to the couple;  an expense that vastly outweighs the additional expense of PGS.

We propose a broader perspective of PGS following analysis of embryos identified as aneuploid from single cell biopsies, to incorporate not only embryo selection (to improve pregnancy rates), but also screening (to reduce miscarriage) and diagnostic information to give patients a realistic roadmap towards future treatments. Despite the push to identify ever more chromosomes in PGS testing, for women over 40 a relatively small sample of chromosomes screened may be sufficient to identify abnormalities incompatible with pregnancy and live birth. In contrast, for younger women with higher quality eggs having concomitantly fewer chromosomal abnormalities, a comprehensive 24 chromosome screen may be more effective at identifying chromosomal errors governing the embryo's ability to implant. To this end, we, and others, are investigating the use of microarray technology to allow us to screen multiple regions of each chromosome simultaneously in single cells. Moreover, limited aneuploidy screening, using a DNA based approach with polymorphic markers is already incorporated into tests for specific single gene disorders.

If we evaluate the respective goals of patients and providers, it is fair to say that, for patients, a live birth is their primary focus. As providers, we must take care not to help couples achieve pregnancy only to result in miscarriage. Indeed, we have a moral obligation not to transfer embryos that are chromosomally abnormal and also have a responsibility for the health of children born following assisted reproduction. Even today's PGS, with its limitations, can prevent common miscarriages and, for that reason alone, can be of immense value to couples.

PGS is not perfect. No test is. But the misuse, inappropriate use or indiscriminate use of a technique does not automatically invalidate it. We believe that with improved microarray technology, the use of algorithms to determine eligibility and robust information to fully inform patients, aneuploidy screening will become one of the mainstream tools for embryo selection. With our current technology, intelligent application of PGS can provide immense benefit to patients. Maybe we need to manage our expectations and those of our patients more carefully, rather than expect miracles.