Q&A: Big Data for oncology

Oncology Research Information Exchange Network could make huge steps to bring together oncology research, oncology patients and drug developers

Register for free to listen to this article
Listen with Speechify
With a tagline that says, “We Empower Stakeholders by Learning from Patients,” informatics solutions company M2Gen seeks to accelerate discovery and development in the area of personalized medicine. Meanwhile, the James Cancer Hospital and Solove Research Institute is the patient-care component of The Ohio State University Comprehensive Cancer Center (OSUCCC), calling itself a “transformational facility that fosters the collaboration and integration of cancer research and clinical cancer care.”
They are collaborating in a way that is set to give new meaning to the term “Big Data,” they say, and to find out more about that, DDNews spoke recently with Dr. William Dalton, the CEO of M2Gen at Moffitt Cancer Center, and Dr. Michael A. Caligiuri, the CEO of the James Cancer Hospital and Solove Research Institute and director of OSUCCC.
DDNews: Please tell our readers about ORIEN and the consortium’s potential impact in the oncology community.
Michael Caligiuri: The Oncology Research Information Exchange Network (ORIEN) is a unique research partnership among North America’s top cancer centers that recognize collaboration and access to data are the keys to cancer discovery. The program was founded by The Ohio State University Comprehensive Cancer Center Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC–James) and Moffitt Cancer Center.
We believe the ORIEN is poised to bring together oncology research, oncology patients and drug developers, based on the consortium’s technical ability and mission to accelerate cancer discovery and delivery through collaborative learning partnerships. But how do we do this and what does it mean from a practical perspective? I often like to illustrate our work with Amazon.com’s personalized model for online shopping.
Consider what happens when you purchase a book from Amazon—for instance, Of Mice and Men. As soon as you purchase that book—or view it—Amazon immediately recommends books that are likely relevant to your personal tastes. Amazon is able to do this because the company has the capability to instantaneously cross-reference hundreds of thousands or even millions of purchases to determine the most-purchased books that are common among 99 percent of customers that have purchased Of Mice and Men. Now—imagine that you could do this for cancer patients and you’ll start to understand the true value and disruption that ORIEN brings to the field of oncology.
Cancer research institutes that are part of the ORIEN understand, at a molecular level, cancer is heterogeneous. We already have the ability to track the molecular profiles of more than 120,000 patients (and counting) over their lifetimes and will be able to recommend clinical trials with unprecedented efficiency.
This efficiently has powerful implications not only for getting patients with highly specific cancer types into potentially life-saving clinical trials, but also should accelerate the drug development process for pharma companies. It truly is a transformative time for oncology research.
DDNews: What is involved in the technology that’s being used by ORIEN and how is it different from other Big Data approaches?
William Dalton: ORIEN is an alliance of cancer centers who have agreed to use data science to generate patient-derived data, and share the data for mutually agreed upon research and create a continuous learning environment. This data is derived from Total Cancer Care, which is a shared research and IRB-approved protocol that secures consent from patients to collect clinical, molecular and self-reported data about their cancers. Importantly, the protocol allows us to follow the patients throughout their lifetimes so that we can learn from them. This ability to follow and recontact patients with their permission and consent is central to what makes ORIEN unique.
We are gathering a broad range of patient data, such as medical history, diagnosis and pathology data, treatment type, treatment response, disease progression and many other factors over time so that we can learn from each patient, identify need and ultimately, predict his or her needs. This is achieved by taking a single patient’s data, which is a very patient-centric and personalized approach, and comparing that patient’s data to other patients’ data already in the data warehouse. We follow the individual and try to determine how he or she compares to similar populations so we can begin to identify patterns and ultimately predict needs.
This is hugely important because we are studying hundreds of thousands of patients and we are getting a lot of data from each patient—essentially from the point of diagnosis and for the duration of life. The needs of individual patients will vary greatly. We felt it important to generate as much data as possible from patients throughout their lifetime journey and to study it in the context of cohorts or populations of similar patients. By studying these populations we will then be able to predict individual needs and ultimately develop evidence-based approaches to meet patient need.
DDNews: What specific functionality does M2Gen provide ORIEN?
Dalton: This is best answered by highlighting the nature of Total Cancer Care and drawing the distinction between ORIEN and M2Gen. As noted earlier, ORIEN is a contractual alliance, not an organization or company. This contractual alliance is important because it acknowledges that all member institutions have entered a collaboration to generate patient data in a similar manner, using the Total Cancer Care (TCC) protocol. They have also agreed to share the data generated at their respective institutions, collaborate on projects and pool resources to achieve what a single research center is not able to accomplish alone.
M2Gen’s job is to implement and operationalize TCC across all partnering centers and to coordinate all efforts. Centers benefit by having access to a greater volume of data to support research initiatives. M2Gen also works with pharma companies in a number of ways including identifying and matching patients to target-based clinical trials and we facilitate partnerships between cancer center members and industry. Pharma companies benefit by receiving insights into patient needs that will ultimately inform what types of therapies to develop and by matching patients to the “most suitable” clinical trials based on their clinical and genomic characteristics. It is a multi-stakeholder effort in which patients, cancer centers and industry come together to serve and meet patients’ needs.
A practical benefit of all this integration is unprecedented speed. For example, a recent paper authored by Messina, et al, described a new molecular signature that predicted prognosis in melanoma and was then studied in approximately 14,500 tumor specimens from patients with many different types of cancer. The original discovery was made in melanoma using data and tissue contributed by patients with melanoma who had consented to be followed throughout their lifetimes (TCC protocol), donating data and tissue for research. Following the discovery in melanoma, the investigators were able to show that the signature was also prognostic in many other types of cancer. Because the investigators had access to data from more than 14,000 patients with different types of cancer it took just six weeks to obtain by using data from the TCC data warehouse. Had the investigators not had access to that data, it would have taken years to generate the data prospectively.
A significant distinction of the Total Cancer Care approach is the ability to “assign” patients to precise patient cohorts—a group of patients that look similar in terms of clinical and molecular characteristics. We can make comparisons between patient populations and basically learn the “needs” of patients. For example, we can determine which population may be at risk of developing progressive disease and may eventually need a new treatment, likely only available in a clinical trial. Informaticists call this “event prediction.” We are in the process of building an informatics system that will study patterns allowing us to predict events based on aggregate assessment of information. Our goal is to use this system to predict and anticipate patient needs, including predicting those who might need a clinical trial, and then be able to provide options of what types of trials may be most suitable for a particular patient.
DDNews:  You have mentioned that about 120,000 patients have consented to your protocol. What types of data are being collected?
Caliguiri: Through Total Cancer Care, ORIEN collects patient tissue and clinical data that includes molecular, clinical, epidemiological data and corresponding genomic data. ORIEN members have access to one of the world’s largest clinically annotated cancer tissue and data repositories.
DDNews:  How does the process work in terms of data collection and analysis?
Dalton: Data collection begins when we ask patients for permission to follow them using our IRB-approved TCC form. This allows us to store clinical data, tissue samples for molecular analysis, and gives us permission to re-contact the patient throughout his or her lifetime. If we have another question or if we discover information that might benefit the patient, such as an appropriate clinical trial, we can reach out to the patient. Data collection starts at consent.
We use a number of sources, chiefly electronic medical records, cancer registries and patient self-reported data. We then coordinate to conduct quality assurance testing to ensure the data is accurate, then integrate and harmonize all of this data to build detailed patient profiles. Once the profile is complete, we can begin comparing the patient to broader populations in our data warehouse. It’s a constantly growing and learning system. And this is where Dr. Caligiuri’s Amazon.com example comes in: The more we can characterize and associate individual patients with broader patient populations, the more our ability to predict future patient events and patient needs, such as when they might need to participate in a clinical trial. This is analogous to what Amazon.com does for customer recommendations.
If you can anticipate the need, you are able to meet the need much more effectively. If you wait for the patient to have the need, it may be too late. We want to anticipate the need and start thinking of options before the patient needs them, so by the time a medical need arises, the patient and oncologist are already well informed. This is the Total Cancer Care approach.
DDNews: How does this data or technology offer an advantage to pharma companies seeking to discover new therapeutics?
Caliguiri: A unique element of partnering with ORIEN is M2Gen, a subsidiary of Moffitt Cancer Center created in 2006, which serves as ORIEN’s operational and commercial provider for support, bringing expertise in data management and informatics. Through M2Gen, industry researchers and pharma are able to match targeted drugs to participating patients within ORIEN cancer center members based on their molecular profiles, promoting greater clinical trial precision and flexibility. ORIEN’s approach to clinical trial matching presents a significant opportunity for pharmaceutical companies to modernize trial recruitment and facilitate adaptive clinical trial design.

Subscribe to Newsletter
Subscribe to our eNewsletters

Stay connected with all of the latest from Drug Discovery News.

DDN Magazine May 2024

Latest Issue  

• Volume 20 • Issue 3 • May 2024

May 2024

May 2024 Issue