Metabolomic testing as a method of predicting IVF success

It is estimated that one in five couples of reproductive age experience some form of fertility problem, and up to half of those may require treatment [1,2]. Currently, the most advanced and successful treatment option available to these couples is IVF. However, to maximize pregnancy success rates, it has become routine practice to transfer more than one embryo in an IVF cycle (in most countries) but still, most women will not conceive owing to the relatively low efficiency of the procedure [3]. On the other hand, a significant proportion of those who do achieve the desired pregnancy may face dire consequences owing to complications of a multiple-embryo implantation [4]. Both negative and ‘overly successful’ outcomes are associated with significant health, emotional and economical consequences [4]. Therefore, it is necessary to continue to improve success rates, while decreasing risks of twin or higher order multiple pregnancies and, to achieve this goal, the current gamete and embryo testing method(s) need to be further developed.

Traditional embryo assessment

Since the successful introduction of IVF by Steptoe and Edwards 30 years ago [5], embryo assessment has relied on visual evaluation of morphological parameters using light microscopy. During the elapsed three decades, there have not been any significant changes in how embryologists evaluate embryos; thus, predicting embryo viability and IVF success has remained limited. The ability (or the lack of it) to correctly estimate embryo viability is critical, as most IVF patients now have a larger number of embryos available for transfer and/or cryopreserve. This is owing to significant improvements in both clinical protocols and medications that can now result in multiple ovarian follicle development [6], as well as the ability to assure fertilization with intracytoplasmic sperm injection [7]. Current ‘testing’ approaches include observation of gametes and embryos at different stages, starting with the oocyte, followed by the two-pronuclear-stage embryos, cleavage-stage embryos through to the blastocyst stage, and these observations provide detailed information on morphology and development. Combining these observations provides an improved estimate on later embryo developmental potential and it may modestly help to predict IVF success [8]. Despite its limited predictive value, morphological assessment of gametes/embryos has a clear advantage: it is simple and noninvasive. The only additional tool that is occasionally suggested to use to distinguish a viable embryo from a nonviable one is aneuploidy testing [9], also referred to as pre-implantation genetic screening using fluorescent in situ hybridization (or other techniques, such as comparative genomic hybridization or microarray) [10,11]. These approaches are very valuable, especially from a scientific point of view; however, routine and clinical application has been impaired by their complexity and ‘invasiveness’ owing to the need for manipulation that involves removing part of the embryo.

Recently, a number of new noninvasive embryo-viability tests have been suggested. These tests aim to measure specific changes from the culture medium to determine what the embryo used, consumed or secreted (oxygen, amino acid, protein and metabolite) and to correlate these parameters with embryo viability. Thus, oxygen consumption by the oocyte or embryo can be measured and related to viability [12–14]. Pyruvate and glucose uptake are also possible to measure and seem to be predictive of embryo viability in different studies [15,16]. Amino acid turnover, measured by high-performance liquid chromatography, provided correlation with embryo development and viability, specifically of those of glycine, leucine and asparagine by different studies [17]. Proton nuclear magnetic resonance established a correlation between glutamate change in embryo culture medium and predicting success of IVF [18]. Yet another approach, SELDI-TOF mass spectrometry, can be used efficiently to establish protein profiles providing characteristic outcomes to individual embryos that can be related to their viability [19]. Although all of these briefly described approaches demonstrate the ability to identify certain markers (noninvasively) from the culture medium and relate them to embryo viability and IVF success, none are simple and practical enough to be used in routine embryology laboratory settings. However, there may be one novel approach that is simple enough for routine use, which is based on the metabolomic profiling of culture medium using the near-infrared spectroscopy.

Defining the technique: metabolomics

Metabolomics has emerged as a multidisciplinary science, requiring cooperation between chemists, biologists and informaticians. A variety of global and targeted methods can be applied and the data integrated to try to provide as complete a picture of metabolic status as possible. The metabolome is dynamic, based upon the activation and interaction of the various metabolic pathways within the cell and is considered to be a reflection of phenotype. It provides an immediate ‘snapshot’ of all current biological functions reflecting up-to-the-minute events. There are several different detection methods that can be used to obtain the metabolomics data; however, Raman and near-infrared spectroscopy-based tests may qualify to be the best for routine use because of the following advantages: direct sample measurement, no preparation is required; little chemical bias; instrumentation costs are considerably low, instrumentation size is much smaller, the instruments are easier to run and maintain, and rapid, simultaneous analysis of multiple biomarkers take place at one time. Recently, there have been a number of studies performed and published describing Raman and near-infrared spectroscopy as a platform to gain metabolomic data in the field of embryology and correlate it with embryo viability and success of IVF treatment [20–23].

Metabolomic testing in IVF

Metabolomic testing of culture medium can help to select viable embryo(s) and maximize the chance of pregnancy (improve efficiency of IVF treatment). It can correctly assess embryos with implantation potential to avoid multiple pregnancy (reduce or eliminate the multiple risks of IVF treatment). It is able to estimate the overall viability of the cohort of embryos and, thus, help in recommending treatment options (fresh embryo transfer only; fresh embryo transfer and freezing of viable extranumary embryos and no transfer or not to freeze nonviable embryos). It can also provide gamete assessment (both sperm and oocytes) to assist in selection of viable, or the best, gametes for use in IVF [24,25].

To date, there have been several studies performed and many of them are published or in press [25–29]. Common to these studies is that embryo culture medium was collected from individual embryo cultures using small volumes that do not interfere with the culture conditions and do not require any changes of the typical laboratory routine. These samples, from various IVF clinics around the world, were sent to a central location where blinded testing was performed and metabolomic data were retrospectively correlated with embryo development data and pregnancy/implantation information. Since, in most of these studies, a single-embryo transfer was performed, it was also possible to gain direct correlation between metabolomic profile and the viability of the embryo, as well the predictive value of IVF success based on the metabolomic testing. Additionally, predictive values of the metabolomic profile were possible to correlate with predictive values of (the traditional) morphological evaluation. All studies have demonstrated that metabolomic profiles obtained from embryos at different developmental stages (day 2, 3 and 5 embryo development and sampling) have strongly correlated with embryo viability, as measured by the implantation/pregnancy outcome of the IVF cycle; thus, also being able to predict the success of treatment [26–29]. Additionally, it was also demonstrated that metabolomic profiles obtained from the embryo culture medium were a stronger predictor of embryo viability and IVF success compared with the traditional morphology evaluation of embryos. Metabolomic profiling of culture medium, in another study, was performed on oocytes (before insemination), and following single culture, the metabolomic data were correlated with embryo development (morphological grading of day 3 and 5 embryos) and also with viability/implantation potential of the derived embryos (by Nagy and colleagues) proving that as early as the metaphase-II stage of the retrieved oocyte, metabolomic testing provides valuable predictive information on success [25].

One of the practical benefits of rapid, noninvasive metabolomic profiling in IVF is the improvement of embryo and gamete selection procedures, allowing a reduction in the number of embryos to be transferred without compromising pregnancy rates. The longer term advantage is the potential to limit the incidence of multiple births and associated healthcare risks and expense. Additionally, metabolomic profiling may also provide valuable ‘diagnostic’ information regarding viability of oocytes/embryos at different stages, which can ultimately assist reproductive endocrinologists to provide the best treatment strategy for each individual patient [30].

Conclusion

The primary objective of IVF treatment is to achieve a viable (singleton) pregnancy for infertile couples by creating viable embryos and selecting the ‘most fit’ for transfer. The technology of metabolomics allows rapid, noninvasive assessment of embryonic reproductive potential prior to transfer. Additionally, application of metabolomic profiling to assess gamete viability also appears promising. Viability testing of gametes and embryos used in cryopreservation cycles is also a logical application of metabolomics in IVF. Further studies will help to clarify the potential practical benefit of metabolomic testing to assist in predicting the IVF treatment success.

Financial & competing interests disclosure

Zsolt Peter Nagy is a member of the Scientific Advisory Board of Molecular Biometrics. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.