Thomas R. Mika president & CEO CollabRx
Apps that help oncologists identify which tests should be considered for tumor molecular analysis and how to use the results of those tests to evaluate therapy options are now available.
The power to access and analyze enormous datasets has the potential to improve our ability to anticipate and treat illnesses. In a recent Forbes article, Dwayne Spradlin, CEO of the nonprofit Health Data Consortium, articulates this view, with reporter Natalie Burg adding that these datasets can help recognize individuals who are at risk for serious health problems.
There is no doubt that a big data revolution is undeway in healthcare. As noted in a McKinsey & Company report published last year, over the last decade pharmaceutical companies have been aggregating years of research and development data into medical databases, while payors and providers have digitized their patient records. “Meanwhile, the U.S. federal government and other public stakeholders have been opening their vast stores of healthcare knowledge, including data from clinical trials and information on patients covered under public insurance programs,” write Basel Kayyali, David Knott, and Steve Van Kuiken. They add that in parallel, recent technical advances have made it easier to collect and analyze information from multiple sources.
Big data has the power to assist physicians in predicting a range of conditions from sepsis to metabolic syndrome, which in turn can lead to chronic heart disease, stroke, and diabetes. However, the ultimate goal appears to be the use of big data in oncology. It can be a challenge for physicians to determine the best treatment approach for a nuanced disease such as cancer. Consequently, it is valuable to ponder the extent to which big data can assist in this endeavor.
Some of the biggest names in cancer—Memorial Sloan Kettering in New York City, M.D. Anderson in Houston, Johns Hopkins in Baltimore, and the American Society of Clinical Oncology—are part of big data initiatives, reports Nick Mulcahy in a November 2014 Medscape article. He goes on to note that the technology and business partners in various high-profile big data projects include some giants in computing, such as IBM, Google, and Toshiba. “The term big data is used in a variety of ways, but its developers claim some common qualities and efforts,” he notes. “Those include attempts to standardize oncology clinical data, improve quality of care through the more widespread collection of patient and disease information, and manage the burgeoning results of molecular and genetic testing.”
To understand why the link between cancer and big data is such an interesting case, it pays to ponder one particular insight, namely the importance of getting over the misconception that “more is better.” With big data, oncologists are able to obtain an unprecedented amount of information about their patients. It is tempting to look at the sheer volume of it all and to be impressed. However, more data does not equal better data.
As Mulcahy observes, no less a figure than Robert Weinberg, Ph.D.—who is credited with discovering the first human oncogene and the first tumor suppressor gene—said in an interview earlier this year that there has already been a great deal of mining of cancer data, but “relative to the effort that’s been put into it, there’s been little in take-home lessons” for clinicians.
Such skepticism is understandable. In fact, the most rigorous quantitative patient profile is merely a collection of numbers. It is useless until interpreted properly. It is wise to remember that as the ultimate personalized disease, no two cases of cancer are ever identical. One implication of this is that a particular cancer case might not be able to be discussed meaningfully in abstract terms, or serve as a predictive tool without reference to the specific patient.
This conclusion derives its force from the limitations and realities of our knowledge of cancer today and is not to be taken for a case of Luddite thinking in medicine. The huge range in the origin and progression of each form of cancer requires a level of decision-making that might never be reducible to algorithms. A separate issue involves the tolerance for false positives and false negatives. In healthcare and especially oncology, a false positive—when it comes to predicting an individual’s response to targeted therapies—might trigger significant amounts of anxiety, or worse, unnecessary and often radical or toxic treatment. A false negative is equally harmful, offering the illusion of good health that might cause a patient to pass the point at which treatment could have been life-saving. Unless and until the tools for analyzing cancer and the metabolic pathways it exploits are much more refined—a process that could take decades—there will be a need for the judgment of humans in the interpretation of big data. In no other area of medicine is this so clear as in oncology.
It is imperative that we continue to refine the use of big data in cancer care; however, we should be aware that there is more to proper insight than what an algorithm might manage on its own. How might the human element be integrated into a big data approach for oncology that could inform therapeutic decision-making? One strategy has already been designed and is being implemented: a series of web-based expert system apps that are designed to enable oncologists to learn which tests should be considered for tumor molecular analysis and how to use the results of those tests to evaluate therapy options, including approved and investigational drugs and clinical trials. This is part of a fully automated and scalable medical informatics solution to inform patient treatment planning by seamlessly pairing the results of genetic sequencing tests with clinically actionable and continuously updated knowledge.
Since the cancer treatment landscape changes quickly, the app has been designed to help oncologists navigate the complex landscape of oncology therapeutic options by offering in-depth explanations of different treatment options for different cancer types.
Explanations and advice come from leading experts in the field, with users able to research drugs and therapies based on tumor profiles, the different stages as well as treatment history. In addition to weighing the pros and cons of different treatment options, the app can help physicians identify the most relevant clinical trials that offer investigational therapies. This app also provides summaries and links to clinical findings published in top-tier medical journals. All of the knowledge and research found in the app is curated by a highly qualified expert advisory network of leading physicians, scientists, and researchers from the nation’s most renowned academic and medical institutions.
Regarding the ultimate value of big data to oncology, the jury is still out. However, it appears reasonable to suggest that no computer will ever expel human experts from the picture entirely.
Thomas R. Mika (firstname.lastname@example.org) is president & CEO of CollabRx.