Intervention Insight’s Trapelo is an innovative platform designed to resolve the complexities of precision medicine in real time. With an initial focus on cancer, where the treatment journey can involve complex and arduous processes, Trapelo aligns the priorities of oncologists, practices, labs, and payers within the practice workflow, eliminates the need for prior-authorization, and speeds up access to the most appropriate treatments for cancer patients when time matters most.
Big3Bio’s Marie Daghlian spoke with Intervention Insights CEO Clynt Taylor at PMWC19 in Silicon Valley to find out how Trapelo came to be and how it fits in the evolving precision medicine landscape.
B3B: Tell me a little bit about your company and the inspiration for Trapelo.
CT: I’ve been in healthcare technology most of my career and I’ve been working primarily in oncology since 2008. I was involved with some of the earlier oncology decision support products, one of which is still used today. I’m also familiar with the lab industry, having helped a lab launch their NGS test. So I’ve helped oncologists use decision support, I’ve helped payers try to resolve the complexity of oncology costs, and I’ve helped labs. I was contacted by Intervention Insights two years ago to come and lead their company. When I got there, I saw that the company had a remarkable precision medicine database of genes, therapies, and the evidence that supports an association between a gene and a therapy. They had been building it since about 2012 and it is curated on a regular basis. They were raking in that evidence and making it available. They had started a pilot that was looking at informing test results coming from a lab, but I didn’t think the solution was comprehensive enough to really solve what I felt were the real problems that exist in precision medicine.
Because of my experience, I felt that we could take this database and sell applications that serve payers, or oncologists, or labs and resolve the complexities at the point of care and that affect everyone, especially the patient. If you look at the way oncologists use precision medicine today, they usually have an idea that they want to use it, they think about ordering a test, they might think about the lab they want to order the test from, then they see if the payer will pay for that test from that lab, and maybe decide to do the test anyway because maybe the lab will cover some of the cost, maybe get assurance that the patient is not going to get stuck with a large bill. So I’ve already described a lot of complexity. Then the oncologists are going to download a form from that lab, and someone on their team is going to fill out that form and fax it to the lab.
That’s just some of the problems they have. Now when the results come back, it’s a 30-page report, a very comprehensive, beautiful report, and it even has some evidence in there about various treatment options. But it hasn’t necessarily taken into account the context—the setting for that patient. So the oncologist has a lot of pages to sift through to find the answers they are looking for. I did a lot of research going into this job and I found out that oncologists aren’t equally equipped to read those reports. So that’s just what happens on the oncologist end of the process.
Labs have a problem. Labs want doctors to know about their tests as soon as they are available, and then they want doctors to remember to use that test when a patient who could use that test comes through, and they want to somehow find a way to get payers to reimburse for that test. Those are the challenges they are facing. Finally, payers are saying, “look I don’t want to pay for research. I want to pay for clinically sound, evidence-based treatments for my patients.” When you take all these conflicting problems and you throw them in the mix, it takes about four weeks from when a doctor decides to do this to getting the test done, getting the patient back in, and getting the payer to agree to pay. It’s a long and painful delay.
So we decided to take our platform and create the first configurable technology platform for precision medicine that can be subscribed to by providers, payers, and labs to resolve the complexities in real time. We call it Trapelo. We started by solving the problem for the doctor because that is the quickest way we can start to improve the experience for the patient. We came up with one of the first lab-agnostic order entry solutions focused on precision medicine. A doctor can go into Trapelo, answer a few questions and see instantly what genes should be tested at a minimum to yield an actionable result—a result that is supported by high levels of clinical evidence. Now you can test the other 290 genes and perhaps get actionable results, but not ones driven by evidence. This is important information because it helps build a case for the payer on why they should reimburse for this test.
Trapelo also automates the order process, avoids mistakes in order entry, and saves time. And when the results come back, they are normalized. What that means is we’ve now captured information that is very important in the patient setting—the type of cancer, whether it was metastatic, whether it was treated with prior therapies or testing, whether the patient is interested in clinical trials–and that informs the report. And when we get the report back from the lab, we get a digital version of the results that we marry with the initial information into a single consistent view – appropriate therapies, clinical trials open for recruitment, and so on. The oncologist can click on an option and we show them the evidence that supports that option.
The other thing that we do is we say to the payer, “you can streamline authorization if the treatment being selected is supported by all this evidence and the doctor used a sound treatment decision process to get there.” We decided to go to providers first because we want them to know this is a solution geared to give them instant efficiencies even before talking to a payer, and makes it easy for them to use precision medicine in decision-making, with normalized results that make it more comfortable to use. Our goal is to say that every patient gets prescreened for the use of precision medicine. We don’t want any patient to come into a practice, and because the doctor wasn’t comfortable with precision medicine, miss the opportunity get a targeted therapy that might be more effective for them. And we sell that value to payers.
B3B: What did you have to do to validate your platform? I’m sure no hospital just decided to take a chance with your platform without some evidence.
CT: Because the company started several years ago, they have been developing this curation process and the technology that supports it for a while. We’ve been vetted by some of the large insurance companies. We also have an active clinical/medical advisory board of leading oncologists around the country who have actively engaged with us—some as long as five years—and they bring great credibility to it as well.
They have had a very deep look to the way the data gets curated. We use data that has been published—clinical data that is constantly updated with the most current published clinical evidence. We have six PhD scientists who lead the curation process along with some technology used to access the data from all the published literature that we track.
B3B: Has the FDA ever asked to see what you are doing?
CT: We have talked to them. They basically see us as a library—taking the data that is already there—an advanced library.
B3B: Kind of what IBM’s Watson was going to do.
CT: Yes, but we added a team of PhDs and we use technology to augment their efforts. We don’t expect the technology to do all the work. We believe that you can’t automate the whole thing. That’s why we have a team, and a medical advisory panel.
We are focused in oncology right now. It will be amazing when the technology gets to a place where the machine can understand a document as well as a human can – and we’re close. But since there’s no real standardization in the way research findings make their way into articles, get published and become part of practice. You have to read it, comprehend it, and understand the nuances and context to really know what it is talking about to apply it appropriately. We still need people, PhDs, to do that. The AI only finds the articles, brings them to us, sorts them by disease, and cues them up in levels of relevance for the curators. We might have a team of 20 people doing all that if we didn’t have the AI doing it, but when it comes to understanding the context and sometimes, sub-findings outside the main study—those surprising side results become part of our knowledge, too—part of our database. It gets weighted based on the kind of research that was done, what the findings showed.
B3B: So you have some kind of grading system.
CT: Yes, we have a data-driven approach that looks at types of research studies – population size, blinded or not — it all makes a difference and gets entered into our knowledgebase. The system then scores it in terms of relative scale of evidence. What we do is not limited to the value of that scoring, although what’s important is that it does let you see quickly. Our team is always updating the data and the oncologists can see the sources of the evidence. We also make finding clinical trials easier. While there may be 20,000 trials in clinicaltrials.gov, we curate a subset that are informed by biomarkers, by genetic markers.
B3B: Kind of like a Cliffnotes for the doctor.
CT: Yes, along with an alert system for the doctor who makes the final ordering decision and treatment selection. We bring them information they can use in real-time and put it into context. Payers want to know that a decision support tool of some kind was used. Everyone benefits from this model and that is why we think it is pretty transformative. It is unique, innovative, and requires collaboration. It’s a win-win situation where everyone benefits. People often ask me “who pays for this.” We have a model where everybody pays something: providers pay to improve their efficiencies; payers pay to have the rules—so everybody contributes in a way where everybody benefits.
B3B: When was Trapelo launched?
CT: We launched it in the spring of 2018. So in the first year, we signed contracts with 800 oncologists. Subscribers get a trial period. We are very collaborative. We work with them to make it easy for them to use it. We are working with payers and oncology groups around the country looking at how to implement this in their settings—we expect it to be a great year. We also create partnerships with other organizations. For example, we have a partnership with InformDNA. They do familial, or germ line, testing. We don’t do that. We have a complementary relationship, and we have others like that where we can streamline their process.
My dream is that within a short amount of time, every oncologist starts by putting some information into Trapelo and learning what things should be tested for that person, given a profile and some potential journeys—options for therapies, options for clinical trials. How do we make sure patients are screened appropriately and can find out whether they might be a candidate for the new targeted therapies.
Payers rely on policies for how they figure out what they will pay for and what they will not pay for. When I’ve talked to payers they are challenged in a couple of ways – can a prior authorization solve the overuse of testing—it can help and we have technologies like ours that can resolve these issues. We have some data that shows the number of times we are under-testing and missing the opportunity to use precision medicine. We need a prescreening for precision medicine.
B3B: How is the company financed?
CT: Company backers include Chrysalis and Beringea, two firms that have been involved with and committed to Intervention Insights since 2012. With the launch of Trapelo, they both stepped up and recommitted for the long term. We are very happy with them.
We don’t have plans to go public but we will likely raise another round of capital in 2020. Our strategy is fluid and depends on the rate of adoption of Trapelo. As we are growing, we are getting a lot of interest from strategic players. I think even pharma companies are going to be interested in what we are doing because when a pharma company wants to recruit for clinical trials, it’s always a challenge—they need an accurate patient profile—diagnosis and testing. Today few places marry those pieces of information in real time. It is usually done by the doctor, who is the bridge. He makes a diagnosis, orders a screening test, then closes that gap between the test results and the initial diagnosis. Trapelo does the same thing for the doctor, but faster.
B3B: Who are your competitors?
CT: When you look at the landscape there are a lot of good companies aiming at meeting the needs of labs, oncologists or payers. So what we did is we kind of navigated into this space in the middle of all of them to solve a bigger problem. While we do some of what these companies do and it might at first glance seem like they’re our direct competitors, I don’t know anybody else who is really doing what we do the way we do it – obviating the need for traditional prior authorization.
B3B: It’s very important. I keep hearing how huge a problem doctor burnout is, and it is mostly due to all the paperwork they have to do.
CT: I’ve had the chance to follow oncologists in clinics. They are dealing with life and death every day. They have to absorb a massive amount of information. They’ve got the burden put on them from payers. They’ve got constant in-their-ear diagnostic companies telling them about the cool tools they can use. And they have just a few minutes to take in all that information and figure out how to help a particular patient. So I want to be a part of helping them figure it out—offering tools that they can feel confident about—we have a whole team of people passionately working to build and maintain it, and we’ve partnered with some incredible companies who are contributing to this, so that at the end of the day, I can give doctors a set of tools that works within their EMR, takes away administrative burden, and gives them quicker access to better information, so they can make better decisions for their patients. Doctors win. Labs win. Payers win and patients win. That’s what we are striving for.