Skip to main content
Premium Trial:

Request an Annual Quote

NKI's Nederlof on Array CGH's Impact on Breast Cancer Research


Name: Petra Nederlof

Title: Head of Molecular Pathology, The Netherlands Cancer Institute

Professional Background: 1997-present, head of molecular pathology, Department of Pathology, The Netherlands Cancer Institute, Amsterdam; 1995-1997, senior scientist, Department of Structural Biology, Max Planck Institute for Biochemistry, Martinsried, Germany.

Education: 1991 — PhD, medical sciences, University of Leiden; 1986 — chemical biology, University of Amsterdam.

Array comparative genomic hybridization has drawn significant attention for its applicability in the field of cytogenetics, as organizations like Baylor College of Medicine, CombiMatrix Molecular Diagnostics, and Signature Genomic Labs have begun using the technology for tests that identify constitutional changes in patients.

But array CGH also shows promise for breast cancer research, according to a recent article in Breast Cancer Research that discussed some of the major issues that researchers in the field currently face when using the approach [van Beers EH, Nederlof PM. Array-CGH and breast cancer. Breast Cancer Res. 2006 Jun 30;8(3):210 (Epub ahead of print)].

BioArray News spoke with Petra Nederlof, an author on the paper and a molecular geneticist at the Netherlands Cancer Institute (NKI), last week.

Why did you write this overview, and have you been using array CGH at NKI?

I was asked to write a review specifically for breast cancer. There have been several reviews on the technology itself, but not related to breast cancer. A lot of information is becoming available now, and the journal thought it was time to get an overview of what was going on in the breast cancer field.

I have been working with this technology for several years now. I started with the other platform, chromosome-based CGH, in 1997. We switched to the array CGH a couple of years ago. So we have done over 200 array CGH [experiments] at this point [in our lab], and mainly on breast cancer.

Do you think that's reflective of greater [array] CGH adoption?

The classical CGH with chromosomes started in the early 1990s when it was introduced by [Anne and Olli] Kallioniemi. Since then there has been a lot of data collected from breast cancer [using that method]. Now people are quickly turning towards array CGH because it has so many advantages over the older technology. It's rapidly being implemented all over the world, I would say.

Is it being viewed as a niche tool or a mainstream tool?

It's a wonderful tool. It's really a breakthrough in many types of research, as well as diagnostics. Apart from breast cancer, it is being used with all types of syndromes, compared to the cytogenetic methods that have been used for years to look at the same syndromes. You have this high-resolution tool to detect small deletions, which is very difficult by classical cytogenetic methods, so in that field it has really been booming.

For cancer research it also has been important because it's a nice tool that can be used on archival material. For many techniques you need to bring tumor cells into culture, which is very difficult for several tumor types, and also in the process of culturing cells you introduce changes in the genome, and that's not what you want. So it's having a major impact on studying solid tumors in general.

I think the effect of array CGH on the syndromes is even larger than in the breast cancer field, especially on the diagnostics part. For diagnostics, it is now widely implemented for the syndromes, like small-deletion syndromes, and is really being used as a diagnostic tool.

But for breast cancer we are trying to implement it as a diagnostic tool, but we are not there yet.

Is that the result of the nature of each challenge? Cancer seems like a tricky area.

With the deletion syndromes it's pretty simple. You have a normal genome where a small portion of the genome is lost. So if you have an assay that can detect a small deletion it's a simple assay because the remainder of the genome is normal.

In breast cancer, or cancer in general, there are many aberrations present in the genome. It's a very unstable genome and as a result you see all kinds of aberrations. If you want to study the early events of breast cancer, and you look at the tumor, there are so many events going on, because of the development of the tumor, that it is difficult deduce what happened at the early stages of tumor development.

So what we've tried to do is develop a classifier that will look at a tumor profile by CGH and determine whether a tumor is from a patient who has susceptibility for breast cancer. We are trying to develop that as a diagnostic tool at the NKI.

There are other labs in the world working on that, but there are not many people doing diagnostics for array CGH and breast cancer yet.

Another field that is being developed — but we are not there yet — is for prognostic issues. So prognosis of the tumor of a patient is being done with expression arrays [like Agendia's MammoPrint]. Similar things can in principle be done using array CGH.

What is the advantage of using array CGH for prognostics?

The advantage of using array CGH for this is that you can use paraffin-embedded archival material. Not always fresh frozen material is present, and that is something you need for expression arrays at present.

I expect in the future array CGH will become more important.

You mention in the paper a need for some sort of data repository for cancer researchers that are using array CGH.

Well the point is if you look at many publications, the number of patients studied in each paper is relatively small. If you want to do a comparison and really get information out of your data, you need to compare usually your patient group with other patient groups or control groups. It is difficult sometimes to obtain that in your own lab, so it would be wonderful if there would be a general database where you have this data available.

There are some databases available where they have tried to set up something like that, but if you look at the number of samples in these databases, it's still a relatively small number. It's really a pity, because I think it would be of great value for the development of the technique and to get closer to having diagnostic applications.

That's my interest. I am in a diagnostic laboratory, so I always want to have a technique that is useful for the patient, not just an academic tool. We can go further than that.

Where does the responsibility to create a database lie?

There is no international array CGH organization yet.

I have noticed that it is difficult to get people to collaborate on this thing. Many people are afraid to give their data because they want to get as much out of it as they can. People are very afraid to give it away.

So I think that it will remain very difficult.

What kind of arrays do you use?

We make them in our own lab. We have a printing facility that makes both expression arrays, oligo arrays, and BAC arrays.

People had a lot of questions about DNA arrays being used in diagnostics. Are those same issues — noisy data, for example — also present for BACs?

Well, we tried to use different kinds of oligo arrays for tumor DNA. In our hands — I am talking about archival paraffin-embedded DNA — you have so much noise on the oligo arrays, because the signal intensity is so much lower than the BAC arrays, that you lose the resolution because you have to do so many tricks to get the noise down. We feel at this point that oligo arrays are not suited for archival paraffin-embedded tissue. So for many people it won't be useful.

But if you have optimal DNA from cell lines or from frozen tissue you may have a higher resolution using those platforms.

We have used commercial oligo arrays specifically developed for CGH. We tested our paraffin samples on them and we even had someone from a[n undisclosed] company in our lab doing the experiments with us. We — both us and the person from the company — were not satisfied with the results. We did have good results with the DNA on our BAC platform.

Which platform you choose should depend on the application, what you want to study, and what is your material. But if you have paraffin-embedded tumor material, I think that for a long time coming BAC arrays will be the best way to go.

The Scan

Genes Linked to White-Tailed Jackrabbits' Winter Coat Color Change

Climate change, the researchers noted in Science, may lead to camouflage mismatch and increase predation of white-tailed jackrabbits.

Adenine Base Editor Targets SCID Mutation in New Study

Researchers from the University of California, Los Angeles, report in Cell that adenine base editing was able to produce functional T lymphocytes in a model of severe combined immune deficiency.

Researchers Find Gene Affecting Alkaline Sensitivity in Plants

Researchers from the Chinese Academy of Science have found a locus affecting alkaline-salinity sensitivity, which could aid in efforts to improve crop productivity, as they report in Science.

International Team Proposes Checklist for Returning Genomic Research Results

Researchers in the European Journal of Human Genetics present a checklist to guide the return of genomic research results to study participants.