AllSeq is proud to be a media partner for SelectBio’s upcoming “Next Generation Sequencing & Single Cell Analysis” conference that will be right here in San Diego on September 28th-30th. One of the benefits (in addition to the 15% discount we were able to negotiate for our readers) was the opportunity to interview a few of the keynote speakers. This gives us the chance to ask the experts how they’re actually using next generation sequencing and how it’s impacting clinical care. For our first installment we talked with Dr. Jennifer Friedman, Associate Clinical Professor at Rady Children’s Hospital in San Diego.
AllSeq – What is your specialty and how are you using next generation sequencing?
Dr. Friedman – I’m a neurologist at Rady Hospital and my areas of specialty are movement disorders and neurogenetics. I see a lot of patients with rare undiagnosed and difficult to diagnose disorders. The area I use NGS in is rare, undiagnosed disease. I think it is a tremendous tool in that arena.
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. Friedman – I think there are certain ‘high yield’ areas and NIPT and cancer genomics have the potential to be the highest yield. Another ‘high yield’ area that wasn’t on your list is pharmacogenomics. In very narrow markets, in terms of the ability of the data to accurately predict the response there are some very specific examples in pharmacogenomics where that correlation is high. In my practice, I don’t use it, except for very, very rare instances in anticonvulsant choice, but I think the data is fairly good about risk factors for side effects for certain anticonvulsants as well as blood thinning agents.
I think for future areas there is a lot of interest in using NGS to predict health risks and to extrapolate from that to preventative screening or treatment. I imagine there may be many companies that pop up outside of healthcare centers that purport to do just that. That area is underdeveloped and I don’t think that the data that we have at this time really supports that, and so I think that area is further off.
AllSeq – How far along the adoption curve do you think we are for clinical next generation sequencing?
Dr. Friedman – I can really only speculate about other people’s barriers, but my sense across the country is that people aren’t aware of it or really understand it, and even if they are, they don’t know how to order it or get it covered for their patients. For my practice I’ve gotten past most of those things and I understand about the limitations and applications of the technology. I’m fairly lucky in being able to get it covered for most of my patients which I think is a struggle for many people. But I also face barriers in that different clinical laboratories have different processes for interpreting and reporting back on variants. It’s a very complex landscape and I don’t fully understand what the differences are between labs and what one lab might be missing that the other lab might be reporting.
The other challenges are that unless there is a slam dunk, absolute diagnosis that comes out of the exome sequence there is a lot of effort that goes into trying to validate VUS (variants of uncertain significance), to either clinically or research validate, get input from other clinicians or researchers in helping with the interpretation. Really that becomes a time consuming and collaborative process for which there is no time and no reimbursement. There really is no mechanism in the current clinic healthcare space to handle this. Even for clinical labs, it’s not clear if they will reinterpret sequences over time. There’s no way for laboratories to bill for that. It’s a very valuable service that could be available but nobody is really in that space because there’s no way to bill for it – insurance companies won’t pay for it. The intellectual aspect of interpretation is really not valued or reimbursed in our current healthcare space, Whether that interpretation is by the clinician or by the lab, it’s just not valued.
AllSeq – Do you have a solution in mind? What would you do if you could wave your magic wand and make any changes you wanted to?
Dr. Friedman – Change healthcare. Again, these numbers are really very broad guesstimates, but there are a number of publications showing that the diagnostic rate of exome sequencing in rare diseases is around 25%. That leaves 75% without a diagnosis, but there may be variants that could be significant that could be investigated further in a research setting. That is not an activity that has any place in our current medical system. Physicians don’t have time and it’s very difficult to bill clinical research collaborations outside of a research center.
Solutions? Clinicians need to be better educated. But beyond that, what always drives healthcare is practiced is how the money flows. And until some of those things are reimbursed or valued in the system, they’re not going to be part of the medical system. They’re going to remain outside in research institutions or potentially in companies that start and serve a very wealthy clientele that can pay out of pocket for those things.
AllSeq – What are your thoughts on targeted sequencing vs exome vs whole genome?
Dr. Friedman – It’s different on a clinical and a research basis. I’ll start with clinical because that’s really what I do mostly. Clinically I’ve never done whole genome – it’s not easily available and very, very expensive. Clinically I do either exome or targeted sequencing and this is a struggle that I have that I don’t have the answer to. It used to be when exome first became available I would run through a list of what I thought were likely genes first and if I struck out (with targeted) then I would go to exome.
There has been a trend toward using exome earlier and earlier in the diagnostic process. Some would advocate ‘exoming’ first, which I don’t – I think early, but not first. I think there need to be thoughtful, standard diagnostic assessments prior to exome sequencing to decide whether that really is the next test of choice. Over time, as I’ve started using exome earlier and earlier, I am disturbed by things that the exome might miss – poor coverage regions, various differences in coverage lab to lab – these are impossible really for the clinician to sort out. So I am struggling with that and I’m struggling with it for several patients.
I’ll just give you a couple of examples. Patients for whom one lab might report out a variant and another lab might not. And that becomes an interpretation difference that goes on behind the scenes. The physician really doesn’t have a window into that process. As a clinician when you run a lab test you want the answer to be the answer and knowing that there may be significant variability lab to lab is disturbing, especially when as a clinician you really don’t have the tools to assess the quality of those different labs or the reasoning that has gone into the differing interpretations. There are other patients where I have had a variant picked up in a research lab (with targeted sequencing) that’s not been picked up by exome. I understand there may be regions of poor coverage in certain genes or few reads across particular genes. It can be quite difficult and time consuming to really get those answers.
There is a little bit of a worry that there’s some complacency if someone has had whole exome that they’ve had everything checked, especially for clinicians that don’t really understand the technology. And so now I’m struggling with a few patients that have had exome that are negative where I am contemplating going back in and doing targeting sequencing. I’m not sure if that’s a waste or not, but I’m worried about areas of poor coverage.
I have another patient in whom one variant was found in a gene that was a reasonable gene for the condition, but since it was a recessive disorder we expected to see two variant mutations. We were able to go back and do targeted sequencing and find that there actually were two where the exome had only shown one. So there are cases like that where if one understands the technology the limitations can be dealt with. But these are highly technical aspects that I’m concerned that with widespread adoption these limitations are not really well understood among those who may be ordering the tests – they may be assuming things have been ruled out when in fact they have not.
So getting back to your question – targeted or exome in the clinical space – I think it’s a mixture of both and it’s a mixture of knowledgeable application of the technology that’s based upon really understanding the pros and cons of both aspects. It clearly touches on panels vs exomes – panels will often have a combination of Sanger and targeted sequencing whereas the exomes don’t, so it’s a very tricky problem that I don’t know the answer to. If I look at a patient and feel that the phenotype is classic I will go to targeted sequencing because I feel I have more confidence and the price is lower for the targeting sequencing. It’s those cases that are less well defined, less classic or have had a number of targeted sequences done first, then I will use the exome. Another difference is turn around time. Targeted sequencing might take two weeks while the exome might take five months, depending on the lab.
AllSeq – Driving down the cost of sequencing has been the biggest push for the NGS market up until now. Do you feel this is still the area that needs the most improvement, or are other factors becoming more important?
Dr. Friedman – In the near future, for sure, it’s other factors. I think we’re very close to that tipping point on cost. A lot of the actual cost depends on patients’ insurance and various contracts that are negotiated. For some people, the amount that an insurance company will pay for a particular test varies tremendously based on negotiated contacts, out-of-pocket, and what particular insurance is involved. So there is still a perception that the sequencing is very expensive because for some individuals it is exceedingly expensive – I’m talking $16k. In reality I think the price has come down quite a bit and I think for my patients I am able to get it for a relatively low price compared to other tests that I order – much lower than for imaging tests or procedures they might need to go through. So from my perspective price is not the issue in my particular practice in the vast majority of cases. I think lack of clinician education, awareness, and understanding are the big barriers. On the other end, I don’t think there’s enough education on the limitations and no capacity in the healthcare system to really support the intellectual time that could or should go into interpreting and applying the results.
AllSeq – Any final thoughts?
Dr. Friedman – One of the biggest challenges for me is that obviously we want to benefit patients. In the rare disease space, when patients come to me they would ideally like a cure, or a treatment to improve as a second best, or even just a diagnosis can be helpful – just a name. I think we have gotten to a reasonable point in being able to provide a name and a diagnosis in roughly 25% of patients but it can be very disappointing. I think there’s a lot of hype around the exome providing the diagnosis. It can be very disappointing for patients when they realize that getting the name for a disease may not put them any further forward than they were – it may not really improve their actual situation with regard to treatment or cure. So there’s a huge leap between making the diagnosis and actually improving the condition of the patient. There is a ‘clinical to research’ gap that needs to be bridged. There’s really not a mechanism for that in current clinical medicine. I think that the number of patients in this situation is going to explode as we identify more and more novel genes. I have quite a number in my practice who are on their own trying to form patient groups and research interest groups and to raise money. And they all, each one of them, can be a consuming research endeavor to try to treat their child. That’s sort of a sad place to be. They search for the diagnosis and when they get it they realize that there’s nothing that modern medicine really has for them. Where I see the large opportunity or need is in that translational space from gene to therapy – drug development and high throughput screening mechanisms for ‘n of one’ trials for these rare patients.
See Dr. Friedman’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.