Executive Conversations: Unlocking the potential of integrated diagnostics with Jonathan Usuka of REALM

Digital transformation in medicine is paving the way for innovation across healthcare disciplines, and integrated diagnostics is one area seeing significant breakthroughs. Dan Sheeran, General Manager of Healthcare and Life Sciences at Amazon Web Services (AWS), sat down with Dr. Jonathan Usuka, Chief Strategy and Informatics Officer at REALM, to understand the drivers behind healthcare’s move towards integrated diagnostics and how it will improve patient outcomes. They also discuss roadblocks to adoption, and the role of technology and cloud computing in helping these initiatives come to life.

Dan Sheeran: Jonathan, we know REALM is a pioneer in integrated diagnostics, so we appreciate you taking the time to share what you’re doing with us. Can you please tell us a bit about REALM—why was it founded and what is its mission?

Jonathan Usuka: Of course, thanks for speaking with me and I’m excited to be here today. REALM is a healthcare company on the forefront of integrated diagnostics, and our goal is to provide a more personalized approach to patient diagnosis and treatment using multi-modal data. Integrated diagnostics bring together genomics, radiology, pathology, and clinical data using differentiated technologies, and then leverages advanced analytics to draw out valuable insights. REALM is bringing this data and leaders together to derive actionable insights that result in better prediction, diagnoses, and treatment of disease.

In my role as chief strategy and informatics officer at REALM, I, and my team are working continuously to leverage technology to advance REALM’s mission of finding better medical solutions for healthcare’s most challenging diseases.

DS: Which factors are pushing the healthcare industry towards integrated diagnostics and why is this shift important?

JU: There are three factors that are moving the industry toward the use of integrated diagnostics. The first is the technological advances like cloud computing, which make it possible to collect, store, process, and analyze massive amounts of data. With cloud computing, we can do things now that we couldn’t do just five years ago.

The second is clinicians’ need to better understand a patient’s diagnosis because the better we understand a condition, the better we can treat it.

And this leads to the third factor—creating superior pharmaceutical and medical interventions. With the knowledge we glean from integrating genomics data, molecular data, and imaging data, the pharmaceutical industry can develop next-generation therapeutics.

It’s important to remember that these data sets existed before, but in silos. What’s changed is that today, there is an industry-wide need to break these silos, so that researchers and clinicians can analyze these multimodal datasets in tandem to get a fundamentally better picture of the disease.

DS: What is the promise and potential of integrated diagnostics in improving patient outcomes?

JU: Integrated diagnostics will allow us to really personalize medicine so that clinicians can prescribe medications that are not only specific to the patient’s condition, but specific to the patient themselves. The potential is that we can do this for many types of diseases.

Oncology is a good example, where clinicians are tying treatment back to biomarkers, like specific genetic mutations in tumors, along with the location of cancer derived from radiology and biopsy reports. On the research side, an entire generation of oncology therapeutics being developed today (like immuno-oncology or cell and gene therapies) depend heavily on integrated radiology, clinical, and genomic diagnosis.

We’re seeing this extend to areas such as neuroscience—in Alzheimer’s and Parkinson’s disease, for example. Where oncology has been over the last few years is very much where the central nervous system or neurological disorders are going in terms of our understanding of those diseases.

The other big therapy area with promise is rare diseases. Ten years ago, we didn’t have a good catalog of genomic profiles for rare diseases because there wasn’t enough data collected, and we could not aggregate across clinical sites to understand their related symptoms and mutations. Today, we can trace back the progress we’re making in our understanding of these diseases to digitization and advanced analytics. This is built on a foundation of integrated genomics, pathology, and radiology data, which is helping researchers find commonalities and patterns among diverse patient populations.

DS: That’s great. So, are pharmaceutical companies the main users of REALM’s data?

JU: There are two main users of REALM’s data.

First is the clinical segment—clinicians, healthcare providers, medical services. Integrated diagnostics can help clinicians with better patient diagnosis and assigning the right treatment to alleviate that condition. Physicians are very busy and have a hard time keeping up with the latest science, let alone to become an expert in different diagnostic modalities, like radiology and genetics. What REALM is doing is providing the context associated with the diagnostic reports so that they can dig deeper to make an informed decision quickly, like changing a medicine or optimizing a dose.

Second, is the research segment, where pharmaceutical companies can use it to develop better, more personalized, next-gen therapeutics. Integrated diagnostics can not only give a more granular view of the disease condition, but can also answer more complex questions that drive R&D.

For example, pharma companies are increasingly using REALM to design better oncology trials. They’re using REALM’s data to refine the inclusion/exclusion criteria for patient recruitment based on biomarkers. And, since a majority of outcomes associated with trials are radiology based (like, decrease in size of a tumor), they’re using REALM’s data to build data assets that look at all modalities to determine the right target patients for an experimental therapeutic.

DS: In your view, what are some roadblocks that integrated diagnostics face?

JU: The fundamental challenge is getting all three types of data (clinical, genomic, and imaging) for every patient. We can’t derive accurate insights with one patient’s genomics and a different patient’s medical images and a third patient’s medical record. We need the three types of data together.

Much of real-world data collection today is siloed—providing just imaging or just genomics. This is, in part, because the data aggregator industry evolved around pharmaceutical claims. So, we’re working with specific consortia worldwide, like the Michael J. Fox Foundation for Parkinson’s disease and the PRECEDE Consortium for pancreatic cancer, to generate that data for each patient.

DS: How has technology and cloud computing advanced the adoption of integrated diagnostics?

JU: Cloud technology has been a complete game-changer in how science is done.

The modalities we are analyzing come with huge amounts of rich data, and that data is sensitive. Cloud technology powered by AWS has given us the ability to collect this data by bringing it from third parties and health systems to a central, secure cloud-based storage location. We can also run machine learning-powered advanced analytics on it to generate reports economically and at scale. Thanks to the cloud, we have the computational intensity and storage capacity we need to analyze and compare files for data from different modalities, in a way that wasn’t possible years ago.

The second area where cloud technology has been really transformative is in combining data from multiple sources (like testing companies and third-party sources) for analysis at population scale. You can’t be generating the data and growing a proprietary data asset through investors’ money—the data production needs to be monetized by sharing and application in different settings. The cloud has made federated and secure data-sharing possible, while ensuring compliance with privacy regulations.

Lastly, it has helped us become global. We’re operating on multiple continents. Lab partners and clinics globally continue to engage us to use our testing, or our pharmaceutical services, through a sustainable business model powered by the cloud.

DS: What advice would you offer to organizations wanting to pursue integrated diagnostics? What are you most excited about?

JU: My advice is to pair integrated diagnostics with contextualization. Provide context to medical practitioners and researchers so they can better understand why findings are important.

I am really excited about the advancement we will see in just the next couple of years. Not that long ago, genomic data used to be shipped by hard drives. Five years ago, it was common for me to be involved in a clinical development program where sequencing patients would involve the shipping of hard discs in a box sent in the mail.

Now, we can get results of gene sequencing faster, store/share it in the cloud, and begin analysis immediately. It’s impossible not to be excited about the next five years. We can have this conversation again then, and I think it will be a very different one.

DS: Absolutely. Integrated diagnostics and multi-modal data analysis is a key priority area for AWS, and we are equally excited about what the future beholds.

I appreciate you taking the time to discuss your breakthrough in transforming integrated diagnostics, and how AWS could power this innovation. It is an exciting time in personalized health, and we look forward to seeing what discoveries lie in these previously unexplored regions.

_______________________________________________________________________

See how AWS is supporting other life science researchers in their quest to expand biological understanding and improve human health.

Dr. Jonathan Usuka has served the Chief Strategy and Informatics Officer for REALM IDx since July 2021. Prior to joining REALM, he served as Senior Expert in Pharmaceutical and Medical Products at McKinsey & Company, led R&D Informatics at Celgene Corporation, served as Senior Director, Life Science Marketing at Accelrys, been the Director of Global R&D Information and Associate Director of Genome Informatics at Roche Pharmaceuticals, and Chief Executive Officer of Ubi. He earned his B.A. from Princeton University, an M.B.A.  from The Wharton School, University of Pennsylvania, and a Ph.D. in Chemistry from Stanford University.