Clarified Precision Medicine is on a mission to ensure every cancer patient receives the right therapy at the right time. By combining the power of AI and Deep Learning with the expertise of a precision medicine panel, they analyze molecular tumor profiles to identify the most effective treatment options—at scale.
Through our Reshape Health Grants, we’re supporting their efforts to build a comprehensive therapy recommendation dashboard designed to integrate seamlessly into the workflow of Pharma professionals. At the same time, we’re working together on a robust platform that delivers insights to health systems, helping standardize tumor data for more consistent and effective care.
We sat down with Daniel Rotroff, CTO of Clarified Precision Medicine, to learn more about the company’s origins, vision, and trajectory—and to explore how the Reshape Health Grant is accelerating their mission to improve cancer care for millions around the world.
Q: Thank you Daniel for your time with us. Please tell us more about yourself.
A: Sure. My name is Daniel Rotroff and I’m the Chief Technology Officer for Clarified Precision Medicine.
Q: What is your Clarified Precision Medicine’s mission and vision? What motivated its creation?
A: Clarified Precision Medicine aims to make sure that all cancer patients receive the right therapy for them at the right time during their course of treatment. So right now, when a patient is diagnosed with cancer, oftentimes they will have their tumor sequenced to identify mutations within the tumor that may make them vulnerable to the tumor vulnerable to certain chemotherapies or targeted therapies. And so, what happens, the results that come back are oftentimes very difficult and complex to interpret and they require a lot of expertise. So, because of that, the tests oftentimes either aren’t ordered for the patients due to the complexity of interpreting them or it requires, an expert panel to review these, and it’s not very scalable. So, it has that solution is difficult has difficulty meeting the demand out there.
And so what Clarified Precision Medicine does is it basically create a scalable solution so that we can offer that expertise and that therapeutic guidance for patients at a level that can meet the demand for all patients with cancer. And so that was really what necessitated the development of Clarified Precision Medicine in the first place was due to some of our clinical colleagues and founders essentially being asked to consult, to have their expertise help identify what the best therapy for patients would be.
Q: Can you walk me through your product?
A: So, right now, when a patient is diagnosed, they send a sample of their tumor or a lot of times the blood will be sent off for a liquid biopsy and it’ll go to a sequencing lab. We work with any of the major labs in the country, and their tumor will get sequenced and a report comes back with all the different mutations that they found. What we do is we take all those mutations, we ingest them into our knowledge base, and we identify therapies that are known to act or potential resistance have resistance to those individual mutations, we then have our expert panel, precision medicine expert panel, review those and identify which therapies are the best for that patient.
And we do this by leveraging AI to sort of scale that solution, offering the recommendations to the reviewers so that it’s quick and easy and accessible to them. And then the reviewers can quickly identify either they confirm that or they can make changes as needed, but it makes it so that we can actually scale this offering while still maintaining the clinical expertise that’s needed to have that sort of hands-on guidance to get back to the physician.
Q: What makes you different from other projects on the market?
A: So, we kind of address two challenges on either ends of the extreme. So either you have an AI-only solution, which struggles to sort of keep up with the latest, greatest, therapeutic guidances out there because AI typically is trained on a lot of data in order to make recommendations. But a lot of times, the latest greatest thing is coming from a clinical trial. There’s not a lot of data yet. And so we need to make sure that we’re acting on the newest things and not necessarily waiting for a large body of evidence to accumulate.
And so that’s why a lot of the AI-only solutions don’t quite meet the needs for patients today. On the other side, the manual panels that are assembled, the molecular tumor boards at large academic medical institutes are very difficult to scale to meet the needs of patients. A lot of times, they can only get to maybe four or five patients a month, and it leaves a backlog of many, many patients that, otherwise don’t have access to that kind of expertise. So, we sort of address that by blending the best of both of those worlds. We use the AI to help us scale, but then we make sure that there’s an expert at the other end to confirm that what we’re doing is based on a best practice.
Q: What is the core problem you are solving, and how did you identify it?
A: I think, you know, a lot of our founders and our clinical colleagues really identified that, you know, a lot of physicians were not ordering the sequencing necessary to identify these therapies mainly due to the complexity of interpreting these results. A lot of times these results come back to 40 to maybe sometimes even upwards of a hundred pages long, and they have multiple matches therapeutic matches that may be on there.
And it’s not clear which one would be the best to start with. Right? If you have six different therapies to choose from, which one should you start with? And so that was really the core problem that we were trying to solve because the evidence is pretty clear. Patients that get on targeted therapies that best match their tumor, they have better outcomes, they have lower adverse events.
I know, overall, just, you know, it’s a better therapy for them. So what we wanna do is make sure that as many patients get on those appropriate therapies as possible. And so by doing this, we can make sure that we increase testing utilization. We aid in the interpretation of those results and help give the support to the physician so they know that an expert has their back. It helps get the patients on the drugs, helps make sure that it’ll get paid for, all with appropriate evidence meeting medical guidelines.
Q: Who are your primary users/customers (patients, doctors, hospitals, researchers)?
A: There’s multiple stakeholders that we serve. So, the patients, of course, benefit by making sure that they get on the best therapy for them. So that’s a clear benefit to the patients. All the evidence is very clear that these target therapies do better when patients get tested and put on those therapies as early in their treatment as possible.
And then we also help the physicians. We help them not only give them some additional expertise that they may not have, but also we help even when they do have the expertise, we help them scale so that they don’t have to spend nearly the amount of time they have to sort of sift through all these pages and results are sitting in molecular tumor boards that can look at our, basically, one-page report that gives very clear guidance, and it helps them so that they can meet the needs of all their patients, more effectively.
And then we also help the hospitals because patients that get on target therapies generally have fewer adverse events. They’re less likely to come back and occupy a bed due to a side effect of therapy or toxicity. So, all these things end up, really it’s a it’s a win-win-win across the board.
Q: How do you integrate AI into your solution?
A: AI is really helpful for us to help do the first step. We kinda like to say that, you know, we use AI to do about 80% of the lift. So we use our AI and our own curated knowledge base to identify based on this patient’s molecular tumor profile, you know, what are all the different therapeutic matches that they could have. And then based on the current knowledge that’s out there and our own history, that we’ve captured in our database, which therapies are most likely to be recommended highly. So we use AI to essentially prioritize our treatments. That does a lot of the lift, and then it generates a report. And then the clinical expert, the precision medicine expert, only has to come in and just verify if that’s correct or they can make changes if they feel like it’s necessary.
If they do make a change, we capture that. It feeds back into our system so that it’ll be a little smarter next time. But we do that to basically help with that lift of bringing all the knowledge together, assembling it, and putting it so it’s right there available to the reviewer so that they can efficiently go through these reviews, making this scalable to where we could, you know, we could review thousands of cases a week, whereas oftentimes, in molecular tumor, we’ll mainly be able to do four or five in a month, when they meet.
Q: What would motivate a doctor to change their recommendation?
A: A lot of the time, it could be related to the other treatments the patients received or basically having just expert knowledge in a cancer. A lot of, especially, doctors in the community setting, they see all comers. They’re seeing patients with all different types of cancers. And so having someone who’s an expert in the course of treatment in this specific cancer, when you’re choosing between two or three, what seem like equal options, having an expert that knows how the pharmacology of that drug works in that setting. And what we found is that physicians, when they get a report, they actually change their recommendation about a third of the time. So that’s a significant amount, you know, time where the physician actually altered their course of treatment based on the guidance they received from Clarified.
Q: Thinking of Clarified Precision Medicine’s journey, what were your needs before winning the grant?
A: So I think one of the things that we’ve needed really is we’ve accumulated quite a lot of data, and this data can help us gain insights into the clinical practices that we’re working with. And they can also help us understand some of the gaps that exist either within a certain cancer type or across cancer types. And that’s where we wanted to work with Arionkoder to develop tools to help us, you know, really take a deeper insight into the data that we’ve collected, identify things like how often patients are switching therapies based on the results of their sequencing result. How often does a Clarified Precision Medicine test influence that? How often are the therapeutic suggestions that we give adding new classes of drugs to what was listed on the next-generation sequencing report?
We often find that not only does our knowledge base identify therapeutic options that weren’t originally put on the next-gen sequencing report, but sometimes we remove options as well that aren’t seen as ideal. The next-gen sequencing report may list four or five different options, and we can help narrow that down to the most appropriate options. And so gaining those insights on how we can help influence practice to make sure these patients are getting on the right treatment, is something we really wanna work with Arionkoder on developing those tools, so we can not only just create these reports more efficiently for patients, but really leverage the data that we’re generating to, to highlight, where the gaps are and how we can help close them.
Q: What are your expectations after we finish the Grant process?
A: Well, I hope we’ll work together to, you know, build on this platform that we’ve created to really help, you know, drive insights into clinical practice and really close gaps in precision oncology.
Right now, there’s a big need. There’s still significant gaps despite the therapies that are available, the science and research supporting it. There’s still, you know, gaps in implementation, and I think the best things we can do together help to highlight those gaps and see if we could develop solutions to them, to make sure that, you know, we can help meet the needs of patients.
Q: How would you measure success?
A: Well, we hope to deploy this platform that we’ve developed, and I think it’ll lead to a series of white papers and academic, academic peer-reviewed papers that right now will highlight some of these gaps that right now are blind spots. We really don’t know how often certain guidance is being acted on, how often it’s consistent with what’s considered best practices.
So, you know, I think that’ll be the next step is really to deploy the tool that we’ve developed to, you know, shed light on, on what’s going on sort of behind the scenes across across the the space of precision oncology so that we can, you know, better address those needs. I think, the more data we can generate in those insights, I think, are gonna be clear. So I would say, you know, the hard deliverables will be basically what we’re able to deliver to, to the public, to the community, to the physician community, to the testing labs. I think those things will, those insights that we can help, because, ultimately, we’ll wanna work with those groups as well to sort of help close all these gaps. So I think, you know, we will our success is gonna be measured by how we can reach more patients and make sure that they get on the right therapies.