Welcome to this latest blog, which will discuss embryo biopsy and pre-implantation genetic diagnosis (PGD).
While many of you may feel that theses processes do not apply to you, the science is progressing to the point that more and more people are becoming candidates. We can’t completely tell yet, but it is possible that in the future almost all IVF cycles will involve PGD. Plus more people may become IVF candidates because of PGD. We’ll see. In the meantime, I would like to bring you up to speed with some newer information about the treatment.
This is the first of a 2 part series. The first will discuss the reasons why PGD took a while to become potentially helpful. This one is a bit technical, so if you’re really not biased towards technical things, you can wait for the next blog, which has a little more practical information. Here we go.
PGD has been around for 2 decades. In fact, one of my partners, Dr. Grifo, was the first doctor in the US to successfully perform the procedure. 20 years is a long time in the world of medicine, but interestingly, the progression of PGD and its usefulness had stalled. The early techniques were important and useful, however limitations it their reliability kept PGD from being used to it’s potential.
The early problems were four-fold. One, the embryos needed to be biopsied on day 3 because we could not at first grow embryos 5 days in the lab. A day 3 embryo has progressed to 8 cells ideally, however, on day 3 many embryos are 5,6,or 7 cells. Of course we do see embryos that are 2-4 cells on day 3, but we usually do not consider them viable enough for a biopsy.
The biopsy is performed by opening the shell of the embryo, as is done in hatching, and then plucking off one cell to be tested. One cell from a 6 cell embryo means that a big percentage (1/6th, or 17%) of the embryo is removed. Embryos may not like to have a big chunk of them pulled away and may not grow as well as an embryo that is undisturbed. In addition, some of the early cells are destined to be part of the embryo itself and others the placenta. At that stage we can’t tell which cells are which so we just take one at random. If we remove one of the few embryonic cells, the fetus may not develop, we think.
These percentage issues led to the second problem, which is that usually only one cell could be removed for testing. If we could routinely take 2 cells we would have get twice as much DNA and therefore be much more accurate with our diagnosis. There are cases where a second cell has been removed; if for instance, if the first cell was damaged in the extraction process. Some IVF clinics have routinely removed two cells to enhance the accuracy of the testing process, however it was shown that removing 2 cells is too harmful to the embryo.
And this leads to the third problem, the biochemistry of the analysis. It’s the DNA of the cell that is tested. I will not go into the gory details of DNA analysis here but I will touch on a couple of things.
Early on, a good but flawed method of analyzing the DNA was used. This involved making chemicals that latched on to the single chromosome, and theses chemicals were of different colors. Because each chromosome is very different, you could have one chemical that only stuck to an area on the 21stchromosome. This chemical for instance, would be green. Other chemicals of different colors would stick to the other chromosomes, such that chromosome 15 could be yellow, chromosome 18 could be red, etc. The cell would be treated with these ”probes,” and under the microscope one could look at the cell and count up the colors. Two of the same was the goal. 3 greens, as an example, would indicate the embryo had 3 chromosome 21s, therefore meaning the embryo had trisomy 21, or Down’s syndrome. This technique also worked well, but not well enough to be near perfect, and in medicine, near perfect is the minimally acceptable result.
Here is a picture (Munne).
You can see that the colors are sometimes faint. One problem occurred when 2 color spots were very close to each other making accurate reading difficult.
One shortcoming of this technology was that only 14 probes were available (fewer earlier on), and we have 23 pairs of chromosomes. Therefore, if an embryo had an extra chromosome 20 and there were no probes for chromosome 20, we would have to say it’s probably normal and do the transfer and hope for the best. This could lead to pregnancy failure or miscarriage. This probe technique is sometimes still used, however newer tests are better for most things.
Other tests, which are sometimes still used, do not use colors and probes. They involve instead making millions of copies of a small specific area in the cell’s DNA. This test is more used when looking for subtle genetic defects, as in sickle cell disease. Having millions of copies lets us confirm that our results are correct. We can actually see the piece of DNA we are looking for.
Here is a picture. Don’t try to figure it out, it’s just an example. Each dark line is millions of copies of DNA (Girardet et al).
In an attempt to make al of those copies, things can go wrong, resulting in the wrong diagnosis or no diagnosis at all. This system is actually a good system, but when using a single cell, the amount of starting material is so small, problems and errors can occur.
The fourth problem was mosaicism. We were all taught that after the egg divides, all of the new cells have the same DNA and are identical. What we have learned from embryo DNA testing is that some cells are different than others. In about 30% of the cases, the embryo is made up of 2 cell types. This is called mosiacism. This is problem when trying to get a picture of the whole embryo based on the DNA extracted from only one cell. If we take off cell that we test as normal, but the rest of the embryo is abnormal, we will transfer that abnormal embryo leading to no pregnancy or miscarriage. The opposite issue of discarding and “abnormal” embryo that is really mostly normal also can happen.
Next time we will talk about the newer developments that may make PGD more acceptable.
Thanks for reading and don't forget the disclaimer 5.17.06.