We caught up with Dr. Nazneen Aziz, Keynote speaker at the upcoming SelectBio conference on NGS and Single Cell Analysis, to discuss how NGS is being used at Phoenix Children’s Hospital where she’s the Chief Research Officer and Senior Vice President.
AllSeq – You’ve had an interesting career – starting in academia, going to biotech/pharma and then on to CAP and now the Phoenix Children’s Hospital. What attracted you to your current position?
Dr. Aziz – It’s very interesting that I started as a molecular geneticist. I got trained at MIT in molecular biology and that was the early days where not a lot was known and things were not in kits so you really had to struggle on your own to do things. I feel that that group of people who did it the hard way learned a lot about this field. And there was great excitement. People would say that genetics and genomics are going to be very huge. We never even used the word genomics, I believe, at that time. Then I went into the biotech industry when there was a lot of excitement with a lot of companies getting into genomics. But because of the economy and the realization that target identification is not that easy, a lot of companies who were in genomics suffered and went bankrupt, and I watched that happen, too. It was interesting for me to even question was this the right field. Although I loved the subject material and what I was doing, I was a little unsure about where the field was going. But then I had a complete turnaround to now where I am in a position where genomics is actually being applied in the clinical care setting. It’s very satisfying for me to be in this position where I went through and saw how the whole field evolved and now we are really at a place where patients are being treated based on their genetics and their molecular profile.
AllSeq – How are you applying NGS in the clinic?
Dr. Aziz – We are using NGS in both childhood cancer, which has been very much ignored, and inherited diseases. We are developing a test for pediatric cancer that is very unique – it will be a very comprehensive test that looks at the entire exome from our pediatric patients, both of the cancer and the normal DNA as well as the RNA. We’re integrating all three types of information to develop a very comprehensive test. Once the test is fully validated we’ll run a clinical trial to see if childhood cancer can be approached differently compared with what is currently used. Right now treatment is more radiation and chemo, which does help around 80% of children with heme cancers in that they do survive. But there are a lot of long-term consequences with these treatments that oncologists are not quite focusing on. But if we do this trial to show that children’s cancer can be dealt with using targeted therapies, by looking at the genomic profiles, then we can get to the point where we can avoid many of the unnecessary co-morbidities of these traditional types of care that have been used in pediatric cancer. We’re also very much focused on inherited diseases of many different clinical areas – neurological, developmental delay, autism, congenital heart defects, and so on.
AllSeq – How far along the clinical adoption curve do you think we are and what’s preventing it from being more widely adopted in the clinic?
Dr. Aziz – We are still in the early stages of clinical adoption as it takes a few things to be in place. One is reimbursement and many private insurance companies are reimbursing for this genomic analysis and testing of patients. And the other thing is physicians’ knowledge and that’s where I think we’re still not quite there. It will take a few years before the medical curriculum changes and there’s a new crop of medical students who are very well aware of what genomics is and how it can be utilized. The third barrier, even with the professionals who are using this testing, is that the infrastructure to use it more efficiently has not yet been developed. The clinical grade databases that can be used by all professionals – knowing that the data in the database is reliable, with no misinformation or errors. Another infrastructure problem is we haven’t quite figured out how to get a genomic analysis data into the EMR (electronic medical record) as a searchable field. So those are a few remaining blocks for full-fledged clinical adoption. We are at very early stages but it’s very exciting because people can see the opportunities of what this can do for us in the future.
AllSeq – What are your thoughts on targeted vs exome vs whole genome sequencing?
Dr. Aziz – For inherited diseases I think people are pretty much still focusing on exome, but it will slowly get to whole genome also, and that’s not too far away. In cancer the depth of coverage is very important, particularly in solid tumors when you have a biopsy – it’s not fully 100% cancer; it is mixed with normal tissue. So the sensitivity of the test can differ depending on whether you have 10% tumor or 20% tumor in your sample. And so the depth of coverage is very critical there, for giving the sensitivity that we need. For that gene panels do much better – we shoot for 1000-fold coverage up to 3000-fold coverage. Exomes, which generally are only in the 150-fold to 300-fold coverage range, don’t do as well as gene panels. But they are superior in that they’re not hypothesis driven. You’re looking at all genes and that allows you opportunities to discover other genes that might be involved in pediatric cancer. With gene panels, as we understand new genes are involved you have to add that to your panel and that requires revalidation – so it sets you back. But with exome you’re really looking at all genes. Whole genome would great because it’s even more hypothesis free as you’re looking at the entire genome. But we don’t get the depth of coverage. We only get about 30X coverage, which is not ideal for cancers where we need the sensitivity of the test.
AllSeq – NGS is starting to be adopted in the clinic, with NIPT, cancer, and rare childhood diseases being the first examples. Where do you think we’ll start seeing NGS adopted next?
Dr. Aziz – I would love to see it being used in non-invasive methods of cancer screening or cancer testing. It’s the same concept as with NIPT – the tumor DNA is floating in our bloodstream. The cell free testing of tumor DNA and reconstruction of the tumor genome using a non-invasive method rather than a tissue biopsy for solid tumors would be a great next step, especially for pediatric cancer. This would mean you don’t have to take multiple biopsies from children. And as for NIPT, it’s still looking at gross chromosomal abnormalities like trisomy 21, 18 and 13. But that field might advance to give you information at the single gene level – single gene mutations rather than at the gross chromosomal level. We’re not there yet with NIPT and I would love to see that area expand, too.
AllSeq – Driving down the cost of sequencing has been the biggest push so far. Is that still the case, or do you think other factors are becoming more important?
Dr. Aziz – I think it will always help to push down the cost even further than where it is, but I think we are at a very good point now. It’s affordable – raw sequencing of a genome is around $1000. There are other issues that are still preventing full adoption. For example, physicians have to be able to understand and integrate the genomic analysis report into their clinical care and so that’s where I think we’re still not there. Also, the bioinformatics pipelines are not foolproof and different vendors’ software will give different variant calls even with the same patient’s DNA. Longer reads will help as will a higher signal-to-noise ratio. A lot has to go into the data analysis downstream and that’s were you need the bioinformatician and that becomes difficult in community hospital settings where they don’t have the infrastructure. At Phoenix Children’s were are pushing genomic analysis in a large way by creating a joint venture with Dr. Patrick Soon-Shiong called the “Chan Soon-Shiong Children’s Precision Medicine Institute” where we’ll do a lot of molecular profiling of children by looking at their genomics, their transcriptomics, and also proteomics. So we have the investment in infrastructure that many hospitals and institutes may not have. If we could make the whole data analysis pipeline very much automated then it allows the adoption and uptake of this technology into the clinical setting more easily. So I think the instrument companies like Illumina, PacBio and Thermo (Ion Torrent) – these companies could really work on those criteria of how to increase the signal-to-noise ratio, how to clean up the chemistries, how to make the read length longer – these all help in the downstream data analysis.
AllSeq – Finally, do you think clinical sequencing will be dominated by ‘point of care’ devices or by more centralized services?
Dr. Aziz – There’s a lot of debate about how it will actually go but I think for the community hospital setting they’ll send it out for sequencing and data analysis.They’ll get the report back and they’ll base it on that. But most academic centers, academic institutions and hospitals like Phoenix Children’s, will want their own facility because this is becoming too important to not have that knowledge internally. It’s not just the report you rely on but a lot of research could be done on new targets and new gene discoveries that are disease causative. They would want their own labs so they could continue the research, so I think you’ll see a mix. Now point-of-care, it may happen for infectious diseases or if you’re looking at a point mutation where NGS could be used. We heard a lot from [Oxford Nanopore] on this with the MinION machine, but I’ve not seen it used in clinical care. But one could envision that those kinds of smaller devices could be applied for a quick point-of-care answer for genotyping or for infectious disease. I have not really seen that happen yet, but it’s possible in the future.
See Dr. Aziz’s keynote presentation at SelectBio’s Next Generation Sequencing, Single Cell Analysis & Mass Spectrometry Conference in San Diego, CA – September 28-30. Use code ‘ALLSEQ’ for a 15% discount on registration.
Also see our interview with fellow keynote speaker Dr. Jennifer Friedman