Name: Joris Vermeesch
Title: Professor, molecular genetics, Katholieke Universiteit Leuven, Belgium
Background: Vermeesch heads the constitutional cytogenetics unit of the Center of Human Genetics in Katholieke Universiteit Leuven. Previously he was head of the genomics unit in the basic research division of Aventis CropScience. He holds a PhD in biochemistry from the University of Nebraska. Vermeesch is also a faculty member of the European Genetics Foundation and is president-elect of the Belgium Society of Human Genetics.
The Center of Human Genetics at Katholieke Universiteit Leuven in Belgium claims to run 50,000 tests annually. Its constitutional cytogenetics lab, which runs more than 10,000 tests a year, performs pre-implantation, prenatal, and neonatal genetic analysis, and genome-wide screening for chromosomal anomalies, according to the center.
UK-based Oxford Gene Technology announced last week that the center recently began assessing its CytoSure arrays for use in prenatal screening. The center already uses OGT chips in its neonatal screening service, and is looking into implementing them for prenatal cases. According to OGT, over the next year the center will use the arrays alongside its existing platforms to determine the feasibility of using them in the clinic. OGT's chips are manufactured by Agilent Technologies.
Joris Vermeesch, an associate professor at the University of Leuven and head of the cyto lab, will oversee the evaluation of OGT's platform. According to a statement, his lab will use the chips to investigate the extent to which high-resolution array-based comparative genomic hybridization can improve diagnosis in cases that show normal karyotype results but abnormal developmental phenotypes.
BioArray News spoke last week with Vermeesch about the evaluation. Below is an edited transcript of that interview.
Why did you decide to consider offering array-based prenatal diagnostics?
We have been applying arrays for post-natal diagnosis, but everyone is asking should we or should we not use it in prenatal diagnosis. There are a number of dilemmas and questions in prenatal diagnostics. One issue is that we clearly don't understand all CNVs that can lead to pathogenesis. Not all pathogenic CNVs lead to abnormalities. So should we assess samples for these imbalances if they do not cause abnormalities? The third issue is that we detect imbalances that may be late-onset disorders. We have the same issues with neonatal samples. The difference is that one does the analysis after there are the same questions, but the baby [has been] born. There is no issue of terminating the pregnancy.
Most people think that array technology might be most relevant in those cases where you detect phenotype traditionally. If you have an abnormal ultrasound and then detect the disorder using arrays, it might help in counseling patients. We have certain criteria that we use to evaluate the potential diagnostic utility of arrays. That is the reason we have decided to start evaluating the OGT platform. They can potentially help us in those cases.
What technologies have you used to date?
For prenatal samples, we did not use arrays for making diagnoses. We did use arrays to look at miscarriages and spontaneous abortions. We published on that work in Genetics in Medicine. We have also applied arrays where conventional karyotyping detected a de novo imbalance. The arrays were then used to confirm findings or to see that there were no chromosomal imbalances near balanced translocations. Most of it was done using our own in-house bacterial artificial chromosome arrays.
Why did you decide to use OGT's platform?
There are a number of reasons. First of all, from our experience in using BAC arrays in neonatal diagnostics, we decided we needed a higher-resolution array. OGT came into the picture because they want to target the diagnostic market. I started collaborating with them in making a custom neonatal array so that is where the relationship began.
From that experience, we started to design prenatal arrays. The other microarray platforms we have evaluated are not as good at detecting copy number variants. The Agilent-manufactured OGT oligo arrays have a good dynamic range. That allows you to distinguish deletions versus alterations and also detect some kind of mosaicism in samples. That is something that other platforms did not provide. For a diagnostic platform, that ability adds some value.
[ pagebreak ]
How will the evaluation be carried out?
There are two levels of evaluation. First, we will over the next 12 months run karyotyping and arrays in parallel. Secondly, if we find imbalances, we will double-check them by fluorescence in situ hybridization, qPCR, or other methods to validate what are the consequences of having those imbalances in a clinical setting. Doctors and patients have to deal with findings. They have to know what the social impact is and so this will all be guided in this social protocol.
And then will you officially start offering a service on the platform?
It depends on the outcome, but [if] things go well, and the arrays' clinical utility is demonstrated … we would implement it. Obviously we will continue using the existing technologies.
What does your array look like?
The current Agilent 105K array helped guide the design of the neonatal array we are using. For the prenatal chip, it is simply the same platform. One of the outcomes might be that we eventually develop a more useful array for prenatal diagnostics. The resolution of the current array is probably too high, but we will start here. The current design has full coverage of the genome, meaning that every 30 kilobases there is an oligo. It targets the syndromes known to be involved in genetic disorders. It also has eliminated to some extent the benign CNVs, so we don't get lost in all copy number variants that are associated with these disorders.
You mentioned you have used a BAC-based platform in the past. Do you think oligo-based platforms will replace BAC arrays in cytogentics?
In my opinion, when you compare BACs with high-resolution arrays, the clinical difference increases only marginally, though it depends on the question you are asking. Of course if you are looking for new syndromes, though, you need a higher-resolution platform. I am not personally against using BAC arrays in cytogenetics, but we are shifting towards oligo arrays.
There is also the issue of using genotyping arrays in cytogenetics versus CGH. Why have you gone with CGH?
For companies, the whole area is a moving target. I am using both SNP and CGH platforms in the lab. To some extent, each platform has its advantages. Clearly SNP arrays give more information, but, until recently, the CNV detection was lower. I haven't used the Affymetrix 2.7M platform yet and I haven't used Illumina's offering. From a research view it is very valuable that these tools exist. You have the SNPs and there are interesting things that you can do with them. But if you are a cytogenetics lab then the added value is not very high.
Are the regulations concerning the use of arrays in cytogenetics testing different in Belgium compared to the UK or the US?
The issue of using arrays clinically exists all over the world. In the US, many companies are already providing prenatal array-based services, so you could argue that in Europe we are slightly more careful of using them. Of course there is anxiety about using a new technology, but ultimately our protocols will be decided by the university. Beyond that there is no formal level of discussion with Belgian authorities.
What's been your setup for data analysis?
I have been involved in data-analysis software firm called Cartagenia that is helping us with the interpretation of array data. We will use their tools as well as OGT's software for data analysis in our evaluation.