NEW YORK (GenomeWeb) – Earlier this year, Nashville-based health technology firm NextGxDx rebranded as Concert Genetics.
The name change reflected the company's growing focus on aligning the disparate interests, standards, and systems of stakeholders in the healthcare system, and getting them humming toward advancing genetic testing and precision medicine. After engaging with doctors, health plans, labs, and other groups, the company this week released a report (see PDF below), entitled "Connecting the Genetic Health Information Network," which identifies eight players in the precision medicine space — patients, clinicians, hospitals, health plans, labs, policy makers, pharma companies, and researchers — and outlines the steps they'll need to take to create a more efficient system for communicating and sharing data, so patients can reap the benefits from advancing genomic knowledge.
Currently, the so-called Genetic Health Information Network exists in an early, inefficient form, lacking the connections necessary to realize the value of precision medicine, according to the report. But Concert Genetics has identified some changes the groups can make in the short and long term to improve things. In an interview, Concert Genetics Founder and Chief Innovation Officer Mark Harris and Gillian Hooker, VP of clinical development, discussed why making these changes in order to build a more efficient Genetic Health Information Network is critical to the advancement of precision medicine.
Below is an edited transcript of the interview.
The report asserts that the lack of an efficient Genetic Health Information Network is what's preventing precision medicine from being of greater value to patient care and lower costs. What is a Genetic Health Information Network, and why is it suboptimal for delivering precision care today?
Harris: The Genetic Health Information Network is made up of those eight stakeholders and the communication between those stakeholders. The network today is suboptimal because the technology, tools, and the standards upon which the communications run were built for systems before genetics arose. Genetic diagnostics are the first step toward a world where precision medicine is realized in the clinic. Without efficient, transparent, and informed genetic testing decisions being made, you can't incorporate genetic medicine effectively into the clinic.
The report identifies eight groups that are already involved in this network, but operating inefficiently. Can you talk a bit about how you see these groups working today and some of the challenges holding them back from contributing to and getting more out of the network?
Hooker: We started by talking to clinicians who are spending an inordinate amount of time trying to assist patients through the different options for testing, [considering] the different requirements to ensure that testing can get covered, or [figuring out] how much testing might cost their patients. Then, they have to understand what to do with the results, so it's useful for the patients.
On the hospital side, one of the biggest problems is that most hospitals couldn't tell you of all the patients they've seen in the last five years, [or] how many have had genetic testing. That's because most genetic tests exist right now outside of the electronic medical record, or as miscellaneous codes in the EMR, or results exist within PDFs.
Health insurance companies still really don't know what they're being asked to pay for. There are a few hundred codes for about 70,000 tests. Increasingly the codes are being used in different combinations to reflect different tests, and insurers have had a hard time parsing through that. Labs also suffer from this complicated billing system. Articles have called out 'coding wars,' where labs and health insurance plans are investing more and more money into parsing coding to improve reimbursement.
Policy makers are clearly struggling just to know what's on the market and what the market looks like right now. The pharmaceutical companies and the research sector are looking at the potential data they can access. If the results of genetic tests were more accessible, there are a lot of questions that could be answered in a real-world setting. But, because we don't have the systems or the standards, they're not answerable at this point. With better transparency and standards, all these stakeholders would be able to do their jobs better.
The report lays out improvements that can take place in the next two years and in the long term. Can you provide some practical examples for how the eight groups can become more efficient members of the network in the near term?
Harris: The first step to delivering precision medicine is the integration of genetics in clinical care in an efficient manner. The targeted therapeutics that will become personalized are generally directed at specific gene mutations. Within that there are areas of inefficiency today. One of the largest is the lack of standards and interoperability across the systems of these different stakeholders.
Transparency in coding is one example. If I'm a lab, I want to get paid appropriately for my test. If I'm a health plan, I want to make sure I'm paying for the right test for the patient. With the lack of granularity with the CPT coding system in genetics, it makes it very difficult for those two groups to transact and speak effectively in the same language.
There are also rising questions about the differences in quality across different vendors of genetic tests. That's an issue for health plans and providers, who want to order the best test for their patients. There's no quality framework today, [and] there is also a need for a standard terminology so we can compare tests across vendors.
This is a unique area in medicine because you have both overutilization of some tests, and underutilization. Not all the people who can benefit from a test are getting it, and at the same time, there is evidence that around 30 percent of tests are misordered. There's opportunity for physician education, or utilization management tools that ensure that the right tests are being ordered for the right patients.
Finally, there needs to be access to datasets for proof-of-concept research studies across stakeholder groups, so we can start to understand what the clinical utility is for these tests. This is very difficult to determine today because the data lie in silos.
Hooker: One of the hallmarks of precision medicine is recognizing aspects of health or genetics that make people unique. The fact that they are inherently unique makes them more rare. What that means is that we can't really organize big randomized-controlled trials for every genetic test that's out there. We're going to have to use real-world data. The only way we're going to get sufficient quantities of data to make determinations about clinical utility, is to have these systems integrated.
Harris: To summarize all of that, what we're saying is that in the next few years, we need to start setting up these processes, and determining these standards so we can creating a base from which we can start to build up this more efficient system.
What should the network look like in 10 years, and do you think the ideal vision of the network you've laid out in the report is achievable in a decade?
Harris: To further enable precision medicine we're going to have to start bringing in raw genomic data into the EMR system. We're going to have to enable decision support systems that can identify appropriate treatment protocols based on genetic test results. A few areas of focus over that time period will be patient control over their genetic data and who can access it, which is critical to having the public trust in order to have this information stored within the medical record.
This area of medicine is too complicated for the human mind to parse through. There are too many tests, treatment protocols, and it's changing too quickly for us to keep up with. We need to leverage technology to help us make the most informed decisions and the system itself has to learn. We need to be plowing more data into decision support systems over time so that the system itself will get smarter, and the individual providers will be kept up to date and will be able to make decisions quickly. Part of that will be EMR integration, and we're starting to see a number of groups trying to pull genetic data into the EMR and tie that into therapeutics.
This learning system must be able to evaluate new tests so … we can continue to support tests with evidence behind them and phase out tests without that evidence. The onus for that must be on the whole system, not just the individual labs, because the labs don't have all the information they need to determine clinical utility in many cases.
Hooker: We all have a stake in making this work.
The report assumes that personalization in healthcare is the way forward, and that all the groups in the network will also increasingly adopt personalized medicine. But within some of the groups you've identified there's been resistance to adopting precision medicine. So, wouldn't the relative speed at which these groups warm to precision medicine also hinder efficiency within the network?
Harris: We assume innovation is going to continue at a breakneck pace in genetics and precision medicine as a whole. All the factors that we measure indicate those trends aren't going to be slowing down any time soon. Science is driving that innovation and more genes are being linked to disease state. With that test volumes go up, computational power is increasing, which enables us to establish more gene/disease linkages, which lead to more tests and data. It's a flywheel that's moving.
The targeted therapeutics are lagging behind, which is logical. It simply takes more time to develop those drugs than it takes to identify the underlying cause of the diseases. We see around 33 FDA-approved drugs with companion diagnostics. That number is going to increase, since the majority of drugs under development are targeted at specific mutations. The actions that can be taken based on these mutations will go up.
But the key question for precision medicine in the long run is if you do the test, can the associated therapies help patients, and do they deliver value to the healthcare system? Can we identify and treat these diseases earlier, save lives, and drive down costs? That has yet to be proven and that's why you're seeing some of the pushback in certain areas. But our point is, let's put the system in place that's going to increase our odds of success there.
Industry observers have said that one of the reasons the network hasn't changed faster to enable precision medicine is due to vested interests. The report touches on this a bit in relation to labs, for example, that aren't sharing data on genetic variants. How would you propose managing those interests in order to realize a more efficient network?
Harris: We've found a significant amount of commonality among stakeholders, but also significant distrust and frustration. The relationship between labs and health plans is a perfect example. But the disconnect we believe is in the system, not with either party. The CPT codes aren't specific enough. Labs are trying different codes to identify what's going to get paid and health plans can view that behavior as manipulation of the codes. Each side is acting logically, but the system is flawed. And that's just one example of how the lack of standards can cause distrust among stakeholders. But if you have transparency and apply the right standards to the problem … it's not an insurmountable problem.
I'll caveat, though, that in this market there are groups that will profit from inefficiency, and [they'll] want to maintain their position and their interests. But in other markets, such as consumer goods, where over time markets are becoming more transparent and efficient, fighting to maintain the status quo for your own benefit isn't a sustainable, long-term business strategy.
Do you envision that one overarching organization needs to lead this effort, get all the groups together, and formally go through the exercise that you have outlined in the report?
Harris: I think we have to work together in a multi-stakeholder fashion, but we're also not implying that one organization or entity can be the sole driver or provide the sole technology backbone for this. We need to move forward together toward a better solution, but it's going to take a lot of different parties working together to get there.
What are some next steps? What do you hope will happen after this report?
Harris: With this report we want to start the conversation. As an organization we want to facilitate the discussion and keep it going, so we can work toward realizing precision medicine.