Podcast EP 7 | Is AI Redefining Oncology Research Enabling Structured Data? Ft. Dr. Amy O’Sullivan, SVP and Chief Scientific Officer at Ontada

I received a PhD in health economics from the Johns Hopkins University School of Public Health and have proceeded to spend, as I said, the last 25 years in health economics and outcomes research both on the consulting side, so developing products and services and doing real-world research for pharmaceutical companies, again to help support the value of their products, as well as on the pharma side. So, I spent about eight years leading health economics and outcomes research functions within pharmaceutical companies.

Host: That is very inspiring and interesting. Also, give us a gist about Ontada’s services and goals. Who exactly are the potential prospects of the platform?

Amy: Yeah great, thank you. So Ontada is magnificent, we are focused on leveraging real world data in oncology to improve patient outcomes. We have a real-world research team, a real-world data and analytics team, and also a team called Ontada Learn, which focuses on physician engagement and education.

In terms of our goals, our goals are to leverage the tools that we have available to us, which includes the electronic medical record platform called iKnowMed that is used by over 2,000 physicians, community-based oncologists, and we leverage those data to generate evidence to answer questions in the healthcare arena specific to pharmaceutical products, and hoping that that evidence that we generate helps to inform decision-making within healthcare and helps patients gain access to oncology treatment faster.

Host: That is really fascinating. Next up, how does your regular day look at Ontada as the SVP and Chief Scientific Officer of the organization?

Amy: Yeah, thank you so much. So, you know, a regular day would comprise of me meeting with my teams. So, we have a really experienced and talented team on the research data and Ontada Learn sides.

They’re epidemiologists, biostatisticians, public health folks, health economists, data scientists, and others meeting with the team to help move forward the projects and studies that we have ongoing for pharmaceutical companies. I also meet with the other members of the leadership team to discuss the strategy and how we can leverage the McKesson ecosystem to help continue to build the products and services that we can offer to life science companies.

Host: That seems really exciting. So, what role does real-world data play in identifying healthcare gaps and seeking the appropriate treatments and how to choose reliable sources for healthcare data?

Amy: Yeah, those are great questions. So real-world data plays a huge role in identifying healthcare gaps and identifying valuable treatments. You know, historically, we have always used data from clinical trials to inform what drugs are efficacious and safe.

And there’s a recognition, I think, now that those data, while they’re always the gold standard to establishing efficacy and safety of drugs, they’re not enough. We need evidence from real-world settings where these medications and treatments are used in real patients in their lives, where there might be other things going on, and they might be used by broader patient populations that are in the trials. So we can use real-world data to identify where are the unmet needs in real-world populations, where should we dedicate research and development to innovate and develop new medicines, and where is the most value in terms of the treatments that we have available today, where do patients get the most value.

In terms of some of other questions around how to choose reliable sources of data, that’s always a big question and challenge for those of us who do research. We’re always looking for, number one, high-quality data with high levels of completeness. So, you want to find data sources in which there’s not a lot of missing data because that could create bias in your research.

You want to make sure that there’s depth and breadth of data that is collected. For example, in our electronic medical record, it’s used to assist physicians to provide care to patients. And so, there’s quite a lot of rich data there around how they’re diagnosing patients, how they’re making treatment decisions, genetic testing and results, and outcomes for those patients, side effects they’re experiencing, etc.

So those are some of the things that we consider when we look at datasets for research.

Host: Definitely valuable data if you consider the data sources is very crucial in healthcare. Moving on, what are the key components of oncology research and how is Ontada revolutionizing research methods for oncology?

Amy: Yeah, thank you. I mean, I think oncology research is constantly evolving and innovating as the treatments for oncology and different tumor types are evolving. I think right now if it’s all about precision medicine, we know that there’s more and more treatments that are being developed for specific patients with specific biomarkers and genetic profiles.

And so, some of the studies that we’ve done historically in larger tumor types, larger size tumor types are now becoming smaller and smaller because, for example, breast cancer is now defined by specific biomarkers that the patients have. Their treatments can be directed accordingly, and we know that they get better outcomes if they are given a treatment specific to their genetic profile. In terms of how Ontada is thinking about moving forward with methods and revolutionizing methodology and oncology data, I mean, I think there’s a number of things here.

Of course, like everybody else in the industry, we are doing our best to leverage advanced analytical methods that have been coming fast and furious, like large language models, AI techniques, NLP, to really extract all of the value from the EHR data source that we have access to. So, for example, there’s a lot of rich data in the physician notes, which are not present in structured fields, but we can use advanced analytics to extract those data using NLP models, for example, and get access to richer data that will provide more robust insights for us.

Host: Ontada is truly setting benchmarks in oncology and real-world data for the same time. Let’s shift the focus to tech advancements in healthcare. Do you think that AI implementation can help in assessing unstructured oncology data?

Does Ontada use AI capabilities for efficiency?

Amy: Yes, absolutely. I mean, AI has really transformed research in a very rapid fashion in terms of how AI can help leverage unstructured fields. As I said earlier, a lot of the richest data are contained in progress notes, physician notes, and other unstructured fields, like PDFs within the medical record.

Actually, about 80% of all the data are contained in those unstructured fields, so AI is really needed to rapidly extract those data in a consistent way, and so we absolutely have started using AI capabilities in the last several years. We have a partnership with Microsoft where they’re helping us sort of cull through millions and millions of records so that we can use AI technologies to get at those data faster.

Host: Definitely, AI can be very advantageous in oncology. What opportunities does Digital Education by Ontada offer, and how does it connect individuals to oncologists quickly?

Amy: Yeah, that’s a great question. So, we at McKesson work very closely with physicians in the U.S. Oncology Network, which is the largest community-based network of practicing oncologists in the country, and we therefore can help them with education by providing digital education through the EHR and other means like podcasts and, like you said, digitally, and really tailor it to how those physicians like to receive their education.

We know that different physicians like to receive education in different ways. We are able to connect with those physicians because, again, of our relationship with the network and because Ontada owns and maintains the EHR that those physicians leverage every day. So, we take great pride in the education we’re able to provide to these physicians who see not only one specific tumor type every day, but as they’re practicing in the community, they can see numerous patients with numerous types of cancer every day, including rare cancers that they may not have had a lot of experience with in the past, so that education becomes very important.

Host: Fantastic. So, how have healthtech practices evolved over the years? In your opinion, will technology transform the healthcare industry in the upcoming decades?

Amy: Yeah, I mean, I think health technology is the number one reason that we’ve had advances in research that we’re conducting, in data sources, in the speed at which we can access those data, and the speed at which we can get to insights from those data. So, absolutely, it’s pivotal in sort of the sea change that we’ve seen, and I think it will continue to be. So, absolutely, with the advent of AI and those technologies, we’ve seen how quickly it’s changed and evolved even over months of time, not even years of time, but over months of time.

We know that those technologies continue to improve, so I’m personally really looking forward to being able to conduct research that at one time took years in a matter of months or weeks due to these advanced technologies.

Host: Technology will surely have an edge over fulfilling healthcare needs in the future. Thank you so much, Amy, for sharing your experiences and knowledge with us. Your contribution to health economics and Ontada’s growth is certainly amazing.

Amy: Oh, thank you so much. It was great to be here. Thank you for having me.

Host: Thank you so much. Thank you, everyone, for joining us in this episode of ExtraMile by YourTechDiet. I’m your host, Sudakshina, signing off for today.

See you in the next session with our next guest and leader, sharing their wisdom with us. Until then, stay tuned.