Editor’s Note: The following was originally posted on EdgeBio's Views from the EDGE blog, and is reposted here with permission. - Janine Holley
March 19, 2012 : Dean Gaalaas (CEO, EdgeBio) : I attended the Human Genome Meeting last week in Sydney, Australia, due to our involvement with the X Prize. It was my first time at the meeting and I came away very impressed at the level of sequencing being presented during the talks. Many of the talks led off by John McPherson’s talk on Cancer Genomics, and were highlighted by Michael Snyder’s talk on using Omics profiling for assessing disease risk and heath states. Through all of the talks a central theme emerged in my eyes, that the $1,000 genome would lead to widespread adoption of NGS as a clinical diagnostic tool. Much has been written and said about the promise of genomics technology and its impact on personalized medicine. However, that is not the focus of this spot. I would prefer to assess the merits of the premise that the cost alone will lead to rapid adoption sequencing as a clinical assay.
To me this seems like an overly simplistic view of what is needed to bring NGS to the clinic. I’ve attended many talks over the past few years and most believe that sequencing is essential to learn more about complex diseases such as cancer and neurological disorders. This draws an interesting parallel to sequencing. As a species, we humans seem drawn to the simplistic view of cause and effect. In the case of human health we tend to think that a change in one gene will cause a disease state and that by fixing this one gene, through a variety of methods, we can cure that disease state. However, we know this not to be true for a majority of disorders.
When it comes to the adoption of NGS, I believe we are falling prey to this overly simplistic model. The current thought is, if we can just make sequencing more affordable like the $1,000 genome, that the adoption curve will start to go asymptotic. To me, it seems that getting NGS into the clinic as a matter of Standard of Care, will be much more complex. Yes, cost plays an enormous role and is a fair enough starting point but there are many more factors that need to be considered. The factor that is most intriguing to me, and the one I have heard least about, is how do we get insurers to reimburse?
In Australia, many of my international colleagues posited that if sequencing were to only cost $1,000 USD, that many would turn to it as an out of pocket expense as this is a matter of routine for most outside of the US. Fair enough, but what will sequencing your genome once tell you about your disease state? Presumably, you are paying that money out of pocket because you suffer, at the time, from a particular ailment. Michael Snyder’s talk focused on looking at his genome over several time points, referencing his genome at a baseline (relatively healthy state), versus disease state (rhinovirus infection), post-infection, and so on. To me this seems like a very reasonable assessment on how to use sequencing as a diagnostic; to monitor changes in the genome over the lifespan of a patient. This means having your genome sequenced multiple times, which makes it doubtful that patients would be willing to pay out-of-pocket as many times as necessary.
Certainly in the US, few if any are willing to pay out-of-pocket for medical expenses. So, will the $1,000 genome cause insurers to line up to reimburse? My thought is no. They wouldn’t rush to reimburse even if it were $100 per genome. The health industry in the US tends to move at what could best be described as a gingerly pace. Standard of Care is routinely cited, and Standard of Care tends not to change in lockstep with discovery or technological advances. No, it tends to change based on legal risk and cost. Tort reform in the health industry is a pet cause of mine but not the focus of this post, so let’s just look at cost.
Insurers are very black and white when it comes to reimbursement, if reimbursing for a test will save them money over what is currently available, they can be convinced to change. Consider, if you will, what is needed to convince insurers that this will indeed save them money. A reasonable scenario may look like this : insurers actuarial tables (and they have reams of data to support them) say that a typical head (term for insured) will stay with the insurance company for an average of 6.8 years and generate $20,000 of revenue for the insurance company over that lifespan. What needs to be shown is that by leveraging NGS as a matter of Standard of Care, the amount of revenue generated per head is greater than $20,000. A very simplistic model to be sure, but should serve to illustrate the point. Clearly a study could be designed to do this, correct? Certainly, but how much power would a study like this need? Over how long of a time period?
I came away from the Human Genome Meeting more excited than ever at some of the groundbreaking research being done across the globe with NGS technology. Now, if we could just match it with groundbreaking research on the cost impact of leveraging NGS in the clinic we will be much closer to the promise that the $1,000 genome seemingly offers.