When President Obama described the “Cancer Moonshot 2020” initiative during his State of the Union address this month, it was apparent that this would be an ambitious, complex, and highly collaborative undertaking. Last week, Dr. Tom Coburn, a former U.S senator from Oklahoma, weighed in on the issue of what it will take to bring Cancer MoonShot 2020 to fruition in an op-ed in the WSJ.
Coburn emphasized the need to leverage big data culled from the vast, yet largely untapped, treasure trove of genetic and clinical data from hospitals and doctors. “Harnessing that information—‘big data’—would allow us to personalize prevention and treatment based on the genetic characteristics of a patient’s tumor, family history and personal preferences, while minimizing unwanted side effects,” he wrote.
When it comes to the ability to leverage big data, the stakes are high, according to Ashish Singh, partner and head of Global Healthcare at Bain & Company in New York City. "In the future, only companies that use real-world data to demonstrate superior outcomes for real drugs will generate attractive returns," he told BioPharma dive in an interview.
In fact, although the biopharma industry has been slow to adopt wholesale digitization, there has been rapid uptake of such technologies in the last couple of years. In 2014, 30% of U.S. doctors were using electronic medical records (EMRs), whereas this year, 75% of doctors will be using EMRs. By 2018, an estimated 93% of U.S. doctors will be using EMRs.
Big data analytics are relevant everywhere
This rapid adoption curve represents an opportunity for pharma to systematically leverage big data to not only improve outcomes, but to also justify specific interventions. However, this is just the tip of the iceberg, and according to Singh, vast quantities of healthcare data—in fact most of it—is underutilized.
Big data analytics is relevant in every aspect of managing a biopharmaceutical enterprise, from R&D, to sales and marketing; to health economics, interaction with payers, regulatory affairs, patient outcomes, and clinical trial oversight. However, there’s a significant barrier stopping many companies from taking advantage of the big data opportunity.
Big data is hard to get
"At this point, data is still hard to access for pharma companies, because the companies don’t own a lot of the healthcare- and drug-related data that’s out there," said Singh. "Many companies have tried to upgrade their big-data strategies, but industry initiatives to this point are mostly scattered."
Singh concedes that the industry has made the most traction in two specific areas—working with available real-world data and building a good customer experience—but because of a lack of access to a comprehensive data set and a fractured digital strategy, companies are "not able to utilize data to link it with analytics, workflow and connectivity across the value chain."
Singh also pointed out that in an optimal situation, "Big-data analytics enable pharma companies to harness real-world data to accelerate drug discovery while also providing valuable health economics and outcomes data for access, pricing and potentially, safety."
Using big data analysis to improve communication with payers
There is an undeniable trend towards payers leveraging their ability to contain costs by conducting strong-arm negotiations with pharma companies. Singh uses the hepatitis C drugs marketed by Gilead and AbbVie as examples. Both treatments are highly efficacious and have completely revolutionized the treatment paradigm for hepatitis C; however, their prices have been discounted by about 50%.
The takeaway here is that because payers have more access to data, they are better able to segment segment and leverage it to justify lower prices. At the same time, pharma companies don’t have access to all of their own data, and therefore lose the chance to leverage analytics to support broader use of their product and justify pricing.
In addition, the role of post-marketing surveillance (PMS) is shifting, according to Singh. PMS is generally used to track long-term safety outcomes, but in the age of payer restraint, it is starting to become a tool to track efficacy–related outcomes. "Post-marketing surveillance is currently regulation driven, but it will become more voluntary and driven by pharma companies, who will use it to track outcomes," Singh observed.
In his Wall Street Journal op-ed, Coburn made the point that HIPAA is a major barrier to being able to leverage data. "We’re handicapping ourselves in the war on cancer, in part because of a web of privacy regulations like the Health Insurance Portability and Accountability Act," he wrote. "HIPAA makes it difficult for researchers to tap into large caches of clinical and genomic data shared across multiple institutions or firms, and then share their findings more broadly."
But there are ways around this challenge. And as Singh said, "Digital health is a broad and complex field, but it’s not a black box. There are solutions emerging from various digital healthcare companies, including data aggregators, such as FlatIron, Orion, and Explorys, which are able to blind and aggregate data, and make it available to biopharma companies for analysis."
Which companies are leveraging big data to their advantage?
There are some examples of pharma companies that are leveraging big data to their advantage. One is Novartis, which has partnered with Walgreens to recruit participants in clinical trials, drawing on the pharmacy’s 100 million-patient database.
In another case, when Bayer needed to generate real-world data for its anticoagulant Xarelto after concerns about drug-related bleeding surfaced, the firm was able to use data from patient registries to show that 96 out of 100 patients did not experience any major side effects, including bleeding.
And in yet another example, Novo Nordisk is collaborating with IBM's Watson supercomputer to use large amounts of blinded data to profile patients with diabetes and gain unique treatment-related insights that would otherwise be unavailable.
One step at a time
Admittedly, the need to capture and leverage big data on an enterprise-wide level is a complicated undertaking. "Healthcare data is very complex," Singh said. "The temptation is to use other industry approaches, but that won’t work. Biology complicates things dramatically."
Therefore Singh suggests selecting a few high-priority areas that represent the highest return for the lowest cost, risk, and time investment, with the goal of eventually implementing an overarching digital strategy across the company.
While we’re still five to 10 years away from a fully digitalized industry with built-in big data analytics, Singh emphasized that there are ways to make the transition easier, while salvaging what’s already working. "It’s not a zero-sum game," he concluded.