Now 10 years old, medical genetics company Invitae continues to forge ahead with its mission of bringing comprehensive genetic information into mainstream healthcare. It aims to accomplish this by aggregating genetic testing into a single service that can lower costs while providing higher quality testing. Sean George, a co-founder of the company and CEO, recently took time to talk with Clinical OMICs Editor-in-Chief Chris Anderson about the company's progress.
Chris Anderson, Editor-in-Chief, Clinical OMICs: You worked for a number of different life sciences companies before co-founding Invitae. What were your kernels of thinking that led you to start the company?
SEAN GEORGE, CEO, INVITAE: I think one was the simple knowledge that an individual's genome is massively impactful on their healthcare, healthcare outcomes, and quality of life. A lot of really great tools came out of the Human Genome Project but, to be honest, for much of my career it wasn't really changing anything in healthcare. In fact, the last 10 or 15 years, healthcare has gone the wrong way in how much it costs and what the outcomes are.
A couple of things that crystallized for me over time was while I love the life sciences, genetics and biotech, it is still not industrialized yet. It is still a relatively young, nascent industry that still has deep, deep academic roots. The way the companies work, the way problems are solved, still has some of that academic patina on it as opposed to the industrialized approach to solving problems that you see in other industries.
We could see the cost of sequencing curve going the way it was going. We had all seen it before, right? PCR went that way; microarrays went that way. The difference with the sequencing itself is it is the fundamental unit of information in genetics applied to healthcare. I'm a lit bit of a student of history and the revolutions fueled by Moore's Law, that is what crystallized it for us. This curve is going to continue. The implications for anything involving human genetic sequencing are profound. So we started a company that puts genetics front and center in human health and actually uses the information in an individual genome to improve outcomes and lower costs.
Also having a fair amount of experience with medical geneticists and genetic counselors I saw the crazy labyrinth of decision making, how much it costs to diagnose someone correctly and how much is missed by the current medical standard of care. The idea was let's get all the information on one platform and deliver it at the right place and the right time, as opposed to this hunt and peck of asking if genetics is involved. We decided to industrialize the whole thing and make it broadly available throughout healthcare, kind of like a utility.
You started with the idea to develop a single platform and test for all diseases and conditions that might be implicated or affected by an individual's genetic makeup. Why did you feel that approach was better than the disease-specific testing that is more the norm?
GEORGE: Now that we are ten years in and we have survived, the reason I would say it is a better approach is because on the margin, that extra information doesn't cost that much more to generate with the technologies that are available today. Now you need to solve a few things and the big one that we solved, the key to the whole thing, is doing the patient interpretation, reporting, and next steps all at scale. Provided you can solve that, the marginal costs is getting to the point where it doesn't really cost that much to generate it. Because if you can generate all that information at a given point in time on an individual, then you can cut through the energy of the heuristic of who gets what test and why. Is a woman at risk of having a complication or having a child with a disease? You know what? We can spend a lot of time asking those questions or we can just run a simple test, and we don't even need to ask the questions anymore, our machine will tell you what the answer is and we can save a lot of time and energy.
Why we think this works is because fundamentally it is a long-tail proposition. Genetics affects in a very real, direct way 20 to 30% of the population as their primary healthcare complaint. I'd be willing to bet another 30 to 40% have some clinical symptoms that are driven by genetics that we just haven't pieced together yet. The issue is that cancer isn't one disease, it is 500 molecular mechanisms gone wrong. Cardiovascular disorders, it's not heart attack, it is DCM, ACM, there are arrythmias, and then good old-fashioned atherosclerosis. Each one of them doesn't add up to a significant portion of the population, but when you add them all up it is major driver of healthcare costs. That long-tail nature of it is what also led us to see in other industries how you solve the long-tail problem. As opposed to niche, high margin, and specialized, it is comprehensive, cost-effective, universal, and global and that is why we area headed in that direction.
When you generate all the information you can on a single test from a person, does that presume that people will actually carry, or have access to, all that data and is there a way for that data to be accessed when needed?
GEORGE: No, that does not exist. That is another aspect of why genetics isn't used as a utility in healthcare and it is a problem we are solving. When I mentioned the expensive part of testing is the processing, reporting, and then delivering what are the patient's next steps, that is just for one test. Now you have generated all this information and there are other implications for that individual. The system currently isn't set up for a physician to look up your profile and say “Hey, I didn't know that. Let's get going on it.” A big piece of our front end architecture is to do exactly that. The mechanism to stay in touch with the patients and their caregivers and provide that information, over time, at the right place and the right time—that is the front end of our system. It is one of the enablers in this future view we have of genetics running into medical utility in healthcare. That has to happen first, and it is not there yet.
What kind of consenting are you doing for patients who are getting an Invitae test, to be able to make the right decision when it comes to delivering results?
GEORGE: That is a key point and also brings up an underlying principle. First, there is a good reason that you don't just dump a whole lot of information on a person and that has to do with consent. A woman is coming in and consenting for a carrier screen, not to find out her risk of Alzheimer's disease. That is a separate conversation, a separate consent, and that is the way we deal with it.
Fundamentally, anticipating where this will go in 10 to 20 years, what we also have as a founding principal is that the patient owns and controls their genetic data, and the patient is the one who decides who gets to see what, to what level you can go down, who the data can get sent to. That is a very important aspect of what we do.
When a person is tested on your platform, are they provided with some kind of access via portal where they can get and share their information, say with their cardiologist, to alert them they may have information there that is helpful?
GEORGE: To be clear, at this time we are running everything from relatively small panels to whole genomes, so not everybody has the whole genome, or much more information than the question that was being asked at the time.
We have a patient portal. Our team can follow up with a patient if they want more information, or can get them in touch with researchers or pharma partners if there is something relevant there. A patient can request all of their data and we will send it to them. We can notify a patient when there has been an update or an upgrade to the genetics they have orderd and we can encourage a patient, in the carrier screen example, we do have an option to say to the woman who is getting carrier screened “while you are getting tested do you also want to understand your cardiovascular risk, your cancer risk?” and then they can select and we can run those at the same time. Our patient portal is not yet to the point where there is an ongoing relationship with the patient over time. We are getting there, as every year we release more of that functionality. Right now, if they want more information from their testing, we follow up with the individual if there is someplace interesting we can hand to them, but that is about it.
Based on that, your vision from 10 years ago, is it fair to say you are is still developing that holistic approach you want to take with the patient and their data.
GEORGE: Things always take longer than you would think. Especially in healthcare, from the regulatory side to the reimbursement side, if I knew then what I know now, would I do it? I'm not sure. I will say that me and my colleagues knew just enough to get the thing going, but not enough to be dissuaded from doing it. There was enough knowledge around the table to know this was going to take a while, and I think also you've seen this over the past 15 years to what is working. What has been really important for us, is we know that the odds of really changing healthcare by banging from the outside are slim. We have seen this time and again. We always knew what we would have to do is go deep, deep, deep into the center of healthcare and become the number one brand in genetics for clinicians, medical geneticists, OB-GYNs, oncologists and only then would we then be able to start changing the way the information was used. Which is exactly what we are doing. However, that takes a while and take a lot of capital and it also take a lot of patients. Our general view is that you can't change it without becoming a leader within it.
