Next-Gen Sequencing: Expanded Genetic Testing in the Real World

Marshall L. Summar, MD; Brendan C. Lanpher, MD; Sean E. Hofherr, MD


September 15, 2014

Editorial Collaboration

Medscape &

In This Article

Do Different Technologies Mean Different Results?

Dr. Summar: Let's shift to some of the practical aspects. Running a molecular laboratory, you stay up on the current techniques. Is it a bit of a misnomer to refer to "whole-exome sequencing"? What exactly are we getting when we order these tests?

Dr. Hofherr: Unlike the first generation of sequencing -- which is Sanger sequencing, when we were typically looking at one gene for one disorder -- next-generation sequencing is a collection of different technologies and different platforms. Each platform has benefits and disadvantages.

Each laboratory has a choice of which platform they will purchase and use. On top of the platform that generates the data and differs from lab to lab, available reagents and kits also differ. Whole-exome sequencing library prep kits are available from several different vendors, and they are completely different products. Therefore, you are going to have variation, depending on which preparation kit you use.

Furthermore, the amount of coverage that you get on a single base across the exome in general also varies, depending on how the laboratory sets it up. The more samples that you multiplex and put together in one run, the less coverage you will get per base. The less coverage you get per base, the worse the test's sensitivity.

Dr. Summar: Let's explore that. When you say "coverage," do you mean that on the same chip or whatever modality we are talking about, the same point of DNA is interrogated multiple times? Is there a certain number that you consider reliable or ideal?

Dr. Hofherr: The way next-generation sequencing works is that it sequences many different regions at the same time. A given base is sequenced many times -- anywhere from once to 10,000 times. The more times it is read, the better the sensitivity.

Ideally, with a clinical whole exome, we are looking at coverages of between 80 and 200 times per base on average across the genome. That does not mean that every single base is covered 80-200 times. Some regions of the genome are difficult to sequence, and some regions are unable to be sequenced. You end up with a large variation. Some regions are going to be left off because they are unable to be sequenced, and some regions are only going to have a couple of reads on that given sequence.

Dr. Summar: There is a misconception that "whole-exome sequencing" means that every known gene is sequenced -- all the exomes. Is that a misconception?

Dr. Hofherr: Yes. There is overlap from one technique to another and from one kit to another among all of the different capture and library preparation techniques. No single kit has coverage of the entire exome. Typically, you get approximately 80% of the gene-coding regions (exomes) of the genome, and there are clinically relevant regions in that other 20% that are not being sequenced.

Dr. Summar: So among the multiple coverages and the large number of fragments being looked at, there are huge information processing issues?

Dr. Hofherr: There are. That is also where the variability from lab to lab comes in. Next-generation sequencing is very heavy in informatics and bioinformatics, so the way you look at the data is going to differ from laboratory to laboratory. Because we are looking at so many regions and so much data, we have to use filters so that we are not looking at all of the results at the same time. As you start using filters, sometimes you find that you are filtering out clinically relevant regions. It's impossible to have no filters, because it would be too much information to go through. Every lab sets up their bioinformatics differently.

Some third-party solutions are available, and more are coming to market, and they are getting better. Many of the large reference labs are using their own bioinformatics solutions. There have been multiple reports about sending samples from the same patient to several different clinical laboratories and receiving different results from different laboratories. It's a consequence of all of the differences we have talked about.

Dr. Summar: What you are describing sounds more like a screening test than a diagnostic test.

Dr. Hofherr: Whether it's a screening test or a diagnostic test, we have to be aware that with the current technology and the way that every single laboratory is performing this test, there are major limitations. It doesn't mean that it's a bad test, or that when you get a result, it's not diagnostic.

You have to realize that a negative result is difficult to interpret because when you have a negative result at the whole exome, you have to wonder whether it is negative because of the technology. Is it negative because of the state of knowledge at this particular time point? Or is it negative because the patient does not have a mutation in any of the genes? All of those factors have to be taken into account.

Whether we are going to be resequencing these patients at a later time is still up in the air. We have gone through the same process with such tests as chromosomal microarray. Some patients were tested on an earlier chromosomal microarray platform and reported out as negative, and when these patients came back to the clinic years later, clinicians wondered whether they should be retested with the newer version, in case regions were missed on the earlier test. The same thing might happen with next-generation sequencing.

Dr. Lanpher: This isn't a screening test per se. It's a very powerful and useful test for us in clinic. You have to be very careful in how you counsel families to understand the benefits and limitations of the test.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.