Elenee Argentinis: Life sciences companies engaging in oncology research face multiple challenges. Getting data that is specific to the population of interest is difficult. Oncology data, oncology clinical encounters are inherently more complicated than other therapeutic areas.
Mike Sanky: So there are some oncology specific data sources available, which detail the oncology care, but they don’t detail care given by specialties outside of oncology; such as urologists for prostate cancer, pulmonologists for lung cancer, dermatologists for melanoma.
In areas like oncology, that information is often recorded in provider notes or in pathology reports, and it requires a lot of technology investment in the form of natural language processing and machine learning in order to extract that information.
Elenee Argentinis: Optum has robust counts of patient lives across many types of cancers, hundreds of thousands of breast and prostate cancer patients, tens of thousands of patients with multiple myeloma and leukemia. But in addition to those patient counts, we understand the specificity that’s required to conduct oncology research.
We’ve embarked on a comprehensive enrichment plan around oncology, to improve the completeness and representativeness of our data. We’re adding external mortality data to improve the completeness of death data; we’re enriching our data with natural language processing to extract stage, and information about tumor, node and metastasis; and finally, we’re mining for genetic mutations so that we can make a more complete, richer, more relevant oncology dataset.
Mike Sanky: Having better data for oncology is really helping pharmaceutical manufacturers in two primary areas. One is designing better clinical trials, and the second is in evidence generation.
Elenee Argentinis: So when you’re doing outcomes research, that end-to-end journey, pre-cancer diagnosis, post-cancer diagnosis and post-surgery are all things that we can bring to researchers.
Mike Sanky: As we think about the future roadmap of our oncology dataset, we’re going to continue to invest heavily in natural language processing, we’re going to incorporate genomic information, married with phenotypic information in the EHR, which will become more and more important as oncology continues to become an area of precision medicine.
Elenee Argentinis: The enhancements that we’re making to our oncology data are not only relevant to researchers, but also to commercial teams, because precision therapeutics need precision data to identify the right patients for your brand.
Mike Sanky: With smarter trial design, manufactures can launch products more quickly, and get these life-extending and enhancing treatments in the hands of patients.
Elenee Argentinis: It’s exciting that Optum data can be used to power the discoveries of life sciences companies, and accelerate the development of therapeutics that could change the lives of oncology patients.