Microarray CGH Embryo Testing: The Jury is Still Out!

My thanks to Levent Keskintepe PhD, Executive Laboratory Director of the Sher Institutes for his input on CGH-genetics ..in this post.

- Geoff Sher
------------------------------------------------------------
Ordinarily, an embryo transferred to the uterus during in-vitro fertilization will have no more than a 25% chance of implanting and resulting in a pregnancy - even in younger women. Of those who conceive, another 15-20% of such pregnancies will be lost through miscarriage. In both cases, the reason for this high attrition is that embryos often have the wrong number of chromosomes…a condition referred to as aneuploidy. To make matters worse, the risk of aneuploidy increases as the woman gets older – so that by her mid forties, the chance of conception is well under 5% per embryo transferred, and the chance of miscarriage rises to more than 40%.

Currently, at the vast majority of IVF clinics, the only method used for assessing embryo quality has been microscopic grading. Unfortunately, such assessment fails to differentiate between normal and aneuploid embryos. In an attempt to improve the ability to diagnose aneuploidy – the rate limiting factor in human reproduction – physicians turned to a technique called fluorescence in-situ hybridization (FISH). At the time, this was the only available method by which they could examine the embryo’s chromosomes. With FISH, specific chromosomes are stained with fluorescent DNA to reveal their structure. But alas, FISH is far from ideal since it cannot reliably access more than 12 of the 23 pairs of chromosomes in human cells.

In 2007, we at SIRM became the first to report on the use of a genetic process that can access all of the chromosomes in the egg or embryo. The method, known as Comparative Genomic Hybridization (CGH), was previously used in cancer research. It reliably indicates which eggs and embryos have the right number of chromosomes, and thus are most likely to develop into a baby.

CGH testing involves taking a sample of DNA from a cell and evaluating chromosome structure for extra or missing pieces, much more accurately than can be achieved by traditional cytogenetic testing. In fact, it even detects subtle losses in parts of a chromosome as well as chromosome duplications that will ordinarily be missed by traditional cytogenetic tests. These smaller alterations, often called "submicroscopic" because they cannot be seen through the microscope, can still cause major birth defects, mental retardation, and genetic syndromes.

Initially, CGH testing methods involved targeting chromosomes in a specific stage of their division known as the metaphase. While this so called “metaphase CGH” (mCGH) provided the first efficient approach to scanning entire genomes, resolution was limited such that only larger-sized chromosomal events could be analyzed. Moreover mCGH requires a high level of experience to perform accurately.

Recently, a new approach known as array CGH (aCGH) or “microarray” was developed. This method makes use of high resolution DNA microarrays enabling the detection of much smaller DNA-copy-number changes. But this technology is still very expensive and its interpretation is often difficult. The real challenge that remains is the transition of aCGH technology from the research realm to the clinical setting.

Since embryo transfers are done 2 to 5 days after egg retrieval, the ability to place “fresh” embryos in the uterus (in the same cycle of stimulation) of necessity requires that any genetic testing be completed rapidly. Since, in the case of CGH, embryo biopsies for testing are done on the third day post-egg retrieval, this leaves only two (and sometimes three) days to receive testing results and thereupon transfer the chromosomally “normal” embryo(s). Trying to beat the clock on this presents a formidable challenge. Consider the following: The embryos must first be biopsied. Thereupon, the DNA specimens need to be transported rapidly and safely to a central genetics laboratory that can conduct CGH. This part takes at least 24 hours, leaving only another 24 hours to report on the result. Once at the laboratory, sample preparation will takes at least another few hours. This is followed by the critical amplification process that requires yet another 2-6 hours to complete. Then follows hybridization, plating, reading and interpretation followed by reliable reporting. This requires that all the laboratory testing be completed over a 24 hour time period, which means that the genetics lab needs to be operational around the clock.

To say that this is “cutting it very fine” is a gross understatement. And it does not even begin to address the biggest problem of all, which is the inability to ensure proper quality control when “the race is on”. Proper quality control would require first testing the DNA to make sure it is of human origin (that it has not been contaminated), testing properly against controls and the ability to repeat the CGH testing when results are “equivocal”. The latter presents a major problem because unlike mCGH, aCGH in its current form does not do well in testing the minute amount of DNA derived from a single cell. When conducted on such small amount of DNA, there is often so much background noise on the reading that the results are hard to interpret. This opens up the possibility of “false positive” as well as “false negative” results.

Recently, the New York Post (October 4th 2009) commented on the clinicall utility of a variation of aCGH where, prior to embryo testing, cells are taken from the inner cheek (bucal mucosa) of one of the genetic parents. These cells are then CGH tested in advance and used as a control template against which the CGH profile of the embryo is matched. It is claimed that this reduces the risk of “false negative” and “false positive” results and speeds up the process considerably, making it possible to perform the transfer of fresh CGH tested embryos. While this could be true, it should be kept in mind that at the time of writing this article, no babies have as yet been born from the transfer of fresh embryos selected using this method. Additionally, it does not rule out any anomalies that the genetic parent (whose cells are being used as controls) might have.

There is no doubt that ultimately aCGH will be so refined as to allow for it to replace the more cumbersome mCGH process altogether. However, at this time more than 98% of babies born as a result of CGH embryo selection have come from mCGH testing, rather than from aCGH. Moreover, the fact that companies selling this technology have not, to our knowledge, done in-house testing to confirm the reliability of aCGH in the clinical setting should be a source of deep concern.

Presently, mCGH testing of both eggs and embryos for “competency” has a proven track record. We have demonstrated through more than 100 live births that a single “CGH-normal embryo, if transferred to a receptive uterus, will have better than a 60% chance of propagating a healthy baby and that the cryostorage (vitrification) of such chromosomally normal embryos does not significantly compromise their viability. This begs the question…. “If it ain’t broke, why try and fix it?”