GENEWATCH
WILL INSURANCE COVER GENOME SEQUENCING?
By CRG staff - interview with Joanne Armstrong
Joanne Armstrong, MD, MPH, is a senior medical director for Aetna, where she heads the Department of Women's Health and is the clinical and strategic lead for genetics.
GeneWatch: There is a common assumption among many people that personalized genomics is going to revolutionize medicine-but ultimately, do you think it will be up to health insurers to decide if and when this happens?
Joanne Armstrong: I actually think that this is an evolution, not a revolution. I think we're still in the early stages, but there are plenty-hundreds-of these technologies that are in clinical care. There has been an acceleration over the last five years or so, and I expect that as technology platforms improve and the science continues to be established around the validity, we'll see more and more of these technologies in clinical care.
So I don't think health insurance is what's standing in the way of the "revolution." I think that an evolution is occurring because the science is establishing the value of it, and I think that will continue.
How do health insurers figure out what genomic technologies they will cover?
I would say that with genomic technology, the principles of what gets covered are the same as the principles for non-genomic technologies. In other words, it's not exceptional; it's the same process.
So as an overview, the services that are covered are those related to the prevention, diagnosis or treatment of an illness. The information that you get from the covered service has to affect the course of treatment; the care or treatment should be likely to improve the outcome, and that improvement should be attainable outside investigational settings-meaning it's not just a research project, but in broad clinical practice you can see improvements; and finally, the service has to be consistent with the plan design, meaning that the customers who are buying the insurance have to have included this in their plan.
Those are the broad principles of coverage for genetic technologies, and it's the same for everything else. When you get down to the next level there are more specific standards for what gets covered. The technology you're talking about must have evidence published in the scientific, peer-reviewed literature that permits conclusions about the performance effect of those technologies on health outcomes. I think this is where a lot of genetic technologies are still sitting, at various stages in this technology evaluation pipeline. The test needs to meet three standards: it needs to have analytic validity; it needs to be clinically valid; and it needs to have clinical utility.
For some of them it's pretty straightforward: There's a mutation that is known to be associated with a certain disease, and there is a test that has been well studied and validated which identifies the gene associated with the disease; and when this gene is identified, you can act on it.
For something like whole genome sequencing, for example, I think that some of it is still in the analytic validity stages. The technology platforms are developing very quickly, and we don't really have technical guidelines yet around those platforms-what quality controls are needed, for example. There is still a need to define standards to analyze the data that comes out of it. What are the standards to assess the quality of the sequence that you've read? What are the standards for measuring false positives and false negatives? Are these standards the same from one technology platform to another?
Then you ask the next level: What's the standard against which this genome is being compared? What is it being benchmarked against? What gets reported, what does not? How do you know when you have enough data that it goes from a variant of uncertain significance to one of significance, to one that should be reported, to one that should be acted on? All of this stuff is really at the early stages of analytic validity and clinical validity.
And then there's this final step, which is clinical utility. What do you actually do with the results? Does the information translate into a measured improved health outcome?
So the challenges span both sequencing and analysis-do you see those two pieces, whole genome sequencing and the analysis of the sequence, being treated similarly from an insurance standpoint?
If you're asking from the point of view of whether they are covered, definitely they are related. The sequencing is just the method by which you get all the billion units. The critical thing is: What do you do with it? How does it translate into a measured improved health outcome for a population or for groups or individuals? That requires that you understand what that alphabet means, what's significant and what's not. The exercise of building that understanding is in progress, but it's a big job. It means building libraries of data, maintaining them, understanding when something moves from a variant of uncertain significance to one of significance to one that gets reported. Then, when do you report it? So there are lots of challenges. It certainly is moving quickly, and it's exciting, but it definitely is in the early stages in terms of getting this to the bedside.
Although the price is coming down, it still costs thousands of dollars to get your genome sequenced. Is that issue of cost one of the things that we're waiting on?
The cost is not the issue that's hanging up coverage policy around it. There are massive technology issues and clinical utility issues, but it's not a cost issue.
Even if a genomic technology is shown to have clinical utility, what about the translation of that raw information into something that doctors and patients can understand and use?
I think there is a lot of concern that has been expressed today about the translation of genetic tests and results into clinical care. There is a legitimate question about who is going to be translating this information for the patient, and given the rate of information that's coming out on these mutations and how it interacts with other clinical or genetic information, a very high level of sophistication is required to understand it. That's just for the physician to understand it, and then you've got the next hurdle of having patients understand it. This will be a challenge for all of us, for all clinicians, for probably decades to come.
Is genetic counseling very often covered in health insurance plans?
We cover it broadly for people with, or at risk for, a range of known genetic conditions. I think, though, it's a little simplistic to believe that a genetic counselor today is going to be able to interpret the results of a whole genome or exome sequence. With that much information-today we have challenges meeting the storage and computational needs just to keep this information archived. So the need for training is across all levels of the medical workforce. Even if the technology were available today, it's not like we have the trained workforce to understand what this means.
Some of this is because we don't really have the evidence of what this information means right now. That's why the biggest challenge, in my view, is not coverage-it's evidence standards. What does this information mean, and what does it mean for the patient?
By CRG staff - interview with Joanne Armstrong
GeneWatch: There is a common assumption among many people that personalized genomics is going to revolutionize medicine-but ultimately, do you think it will be up to health insurers to decide if and when this happens?Joanne Armstrong: I actually think that this is an evolution, not a revolution. I think we're still in the early stages, but there are plenty-hundreds-of these technologies that are in clinical care. There has been an acceleration over the last five years or so, and I expect that as technology platforms improve and the science continues to be established around the validity, we'll see more and more of these technologies in clinical care.
So I don't think health insurance is what's standing in the way of the "revolution." I think that an evolution is occurring because the science is establishing the value of it, and I think that will continue.
How do health insurers figure out what genomic technologies they will cover?
I would say that with genomic technology, the principles of what gets covered are the same as the principles for non-genomic technologies. In other words, it's not exceptional; it's the same process.
So as an overview, the services that are covered are those related to the prevention, diagnosis or treatment of an illness. The information that you get from the covered service has to affect the course of treatment; the care or treatment should be likely to improve the outcome, and that improvement should be attainable outside investigational settings-meaning it's not just a research project, but in broad clinical practice you can see improvements; and finally, the service has to be consistent with the plan design, meaning that the customers who are buying the insurance have to have included this in their plan.
Those are the broad principles of coverage for genetic technologies, and it's the same for everything else. When you get down to the next level there are more specific standards for what gets covered. The technology you're talking about must have evidence published in the scientific, peer-reviewed literature that permits conclusions about the performance effect of those technologies on health outcomes. I think this is where a lot of genetic technologies are still sitting, at various stages in this technology evaluation pipeline. The test needs to meet three standards: it needs to have analytic validity; it needs to be clinically valid; and it needs to have clinical utility.
For some of them it's pretty straightforward: There's a mutation that is known to be associated with a certain disease, and there is a test that has been well studied and validated which identifies the gene associated with the disease; and when this gene is identified, you can act on it.
For something like whole genome sequencing, for example, I think that some of it is still in the analytic validity stages. The technology platforms are developing very quickly, and we don't really have technical guidelines yet around those platforms-what quality controls are needed, for example. There is still a need to define standards to analyze the data that comes out of it. What are the standards to assess the quality of the sequence that you've read? What are the standards for measuring false positives and false negatives? Are these standards the same from one technology platform to another?
Then you ask the next level: What's the standard against which this genome is being compared? What is it being benchmarked against? What gets reported, what does not? How do you know when you have enough data that it goes from a variant of uncertain significance to one of significance, to one that should be reported, to one that should be acted on? All of this stuff is really at the early stages of analytic validity and clinical validity.
And then there's this final step, which is clinical utility. What do you actually do with the results? Does the information translate into a measured improved health outcome?
So the challenges span both sequencing and analysis-do you see those two pieces, whole genome sequencing and the analysis of the sequence, being treated similarly from an insurance standpoint?
If you're asking from the point of view of whether they are covered, definitely they are related. The sequencing is just the method by which you get all the billion units. The critical thing is: What do you do with it? How does it translate into a measured improved health outcome for a population or for groups or individuals? That requires that you understand what that alphabet means, what's significant and what's not. The exercise of building that understanding is in progress, but it's a big job. It means building libraries of data, maintaining them, understanding when something moves from a variant of uncertain significance to one of significance to one that gets reported. Then, when do you report it? So there are lots of challenges. It certainly is moving quickly, and it's exciting, but it definitely is in the early stages in terms of getting this to the bedside.
Although the price is coming down, it still costs thousands of dollars to get your genome sequenced. Is that issue of cost one of the things that we're waiting on?
The cost is not the issue that's hanging up coverage policy around it. There are massive technology issues and clinical utility issues, but it's not a cost issue.
Even if a genomic technology is shown to have clinical utility, what about the translation of that raw information into something that doctors and patients can understand and use?
I think there is a lot of concern that has been expressed today about the translation of genetic tests and results into clinical care. There is a legitimate question about who is going to be translating this information for the patient, and given the rate of information that's coming out on these mutations and how it interacts with other clinical or genetic information, a very high level of sophistication is required to understand it. That's just for the physician to understand it, and then you've got the next hurdle of having patients understand it. This will be a challenge for all of us, for all clinicians, for probably decades to come.
Is genetic counseling very often covered in health insurance plans?
We cover it broadly for people with, or at risk for, a range of known genetic conditions. I think, though, it's a little simplistic to believe that a genetic counselor today is going to be able to interpret the results of a whole genome or exome sequence. With that much information-today we have challenges meeting the storage and computational needs just to keep this information archived. So the need for training is across all levels of the medical workforce. Even if the technology were available today, it's not like we have the trained workforce to understand what this means.
Some of this is because we don't really have the evidence of what this information means right now. That's why the biggest challenge, in my view, is not coverage-it's evidence standards. What does this information mean, and what does it mean for the patient?
.png)
No hay comentarios:
Publicar un comentario