4 Approaches To Data Access Strategy — And How To Apply Them Throughout Drug Development
- Posted by: LSTI-Editor
- Category: Clinical,
The data landscape is evolving rapidly, and some pharmaceutical companies are taking strong positions to gain privileged and preferential access to data. Oncology is at the forefront of this movement and is heralding a wave of activity across other therapy areas. Looking forward, it will be increasingly important for pharma companies to employ ingenious strategies to access data already existing in the healthcare environment, moving away from solely building and owning data sets. These new approaches will bring significant new opportunities and help companies to further differentiate their medicines from competitors and accelerate their products to market.
The Data Landscape Is Evolving
Data is increasingly prevalent. Genomic, clinical, and behavioral data have all become progressively collected, codified, and digitized into electronic medical records (EMRs), driven by advances in technology and government policy (e.g., Meaningful Use Act in the U.S.).
Data analytics capabilities are advancing rapidly. Increased processing power means larger and more complex data sets can be analyzed and has enabled a shift from statistical to machine learning analytics, driving improvements in prediction and pattern recognition capabilities. On top of this, natural language processing is enabling codification of free-form clinical text for use in analysis.
External companies are innovating to provide new ways for pharma to access data. Data can now be accessed in an increasingly cost-effective way beyond traditional trials and registries, with external players employing innovative business models, such as data as a service and direct-to-consumer diagnostics. Additionally, companies such as Nebula Genomics are emerging, offering patient-owned blockchain genomic data, which, in the future, could change how patient data is accessed altogether.1
Real-world evidence (RWE) from routine clinical care is increasingly being used in regulatory settings. Regulatory bodies are becoming more comfortable with innovative applications such as external and synthetic control arms to supplement placebo or current standard of care arms in randomized clinical trials.
Pharmaceutical companies are taking a stake in the data ecosystem and a lot of corporate activity is being seen in the space. The Roche and FlatIron acquisition2 and the strategic partnership between GSK and 23andMe3 are examples of activities creating more privileged access to data. Pharma’s influence could irreversibly change the direction that data holders take, meaning outsiders may lose the freedom to fully access the data they need (for example, GSK’s influence on 23andMe post Parkinson’s collaboration, leading to the latter’s partnership with the Michael J. Fox Foundation4).
Forming A Successful Data Access Strategy
For pharmaceutical companies, it is important to consider multiple areas when designing a data access strategy. The key question to answer is what are our priority use cases? Answers can range from discovering new targets to differentiating in-line medicines, and these answers are vital to inform follow- on questions. Next, data types and analytics required need to be considered for each use case, i.e., “what are the right data types for our use cases?” and “how will we gain insights from this data?” Lastly, the current situation needs to be assessed, i.e., “what capabilities/data do we currently have?” and “how will we access this data and/or analytics capability?”
Our work analyzing recent developments and discussing them with industry partners has led us to identify four distinct approaches that forward-thinking companies are employing to enhance their development activities. Each of these approaches leverages valuable data already in the healthcare ecosystem, but in vastly different ways. The approach a company selects will greatly impact how it brings its drugs to market and, conceptually, can be used by ambitious companies as aspirational targets to base future strategies around.
- “The hardcore scientist” – Focuses on broad and deep data sets, with bespoke analytics to allow a tailored approach to development. These companies have a medium- to high-risk approach to data access and are prepared to make significant investments, such as strategic partnerships, with the expectation of greater data ownership, value add, and potential for revenue generation.
- “The therapy hacker” – Accesses real-world data sources to make rapid impact in the market. Their eyes are open for novel partnership opportunities to access data and analytics that will help them get to the patient quickly. With companies offering ad-hoc access to high-quality clinicogenomic data, they can take a medium-risk approach, using smaller one-off but high-impact investments to accelerate products through development.
- “The frugal thinker” – Looks for flexibility and cost reduction, accessing data through open access and public resources to limit risks and increase profits. This low-investment approach reduces risk significantly but also leads to lower value add from the data. The frugal thinker will always be a data accessor rather than a large data set owner, reducing downstream commercialization opportunities.
- “The technical engineer” – Looks beyond data to owning the technology itself, whether diagnostic or digital. While the investment may be high, in their eyes the long-term payoff from continuous data streams and ready-to-use tools justifies the approach. Furthermore, the ability to commercialize the capabilities and data they obtain can diminish the risk of the higher investments involved.
Approaches Applied Across The Development Life Cycle
Whichever approach is taken, all of them can be successful, but in different ways. Moving through the development cycle, each approach has different benefits, and a company is not bound to one. In fact, companies can switch between approaches, depending on the development stage they are at, the market they are moving into, and the indication or type of medicine they have.
Drug Discovery And Early Development
- “The scientist” will enter into consortia or collaborations. Examples include BMS’ partnership with the Parker Institute5 or other players partnering with specialist centers like Memorial Sloan Kettering Cancer Center (MSKCC). Although often costly, these collaborations allow the sharing of broad and deep data sets that give greater understanding of disease and markers. They will leverage bleeding-edge analytics capability to analyze these multi-dimensional data sets by partnering with academic institutions.
- “The frugal thinker” takes a different approach and, to minimize costs, takes advantage of the troves of publicly available free data from databases such as Genomic Data Commons6 and cBioPortal.7 They rely on their in-house data science capability to see novel patterns or signals in the data.
- “The engineer” focuses on how to get the right patient on the drug to achieve better outcomes or to intercept diseases altogether. They will tend to concentrate on developing or investing in (companion) diagnostics or prognostic technologies and selecting therapies that would benefit from a diagnostic element like an associated genetic or protein biomarker. This approach is exemplified in J&J’s investment in GRAIL8 and Roche’s acquisition of Foundation Medicine.9
Mid- To Late-Stage Clinical Development
- At this stage, “the scientist” leverages detailed biomarker data to determine specific inclusion and exclusion criteria based on markers of resistance or efficacy. They select sites in a tailored fashion to gain access to specific patient populations, partnering with relevant companies, such as what Clovis Pharmaceuticals has done with Strata Oncology.10 Finally, they build bespoke value propositions by employing novel intermediary endpoints in trials.
- “The hacker” will aim to speed up approval times using RWE as well as using alternate data types to build their value proposition. They will partner with innovative data companies to access RWE and produce synthetic comparator or placebo control arms or gain data on compliance to therapy through mHealth solutions. FlatIron’s pharmaceutical collaborations and partnership with the FDA around real-world control arms appear to follow such a strategy.11
Market Opportunity Optimization
- “The frugal thinker” will look to gain one-off access to EMR data and use this to show how much market penetration they have. The EMR data will help them understand how the therapy is now being used and demonstrate positive outcomes. They will maximize their investment by utilizing internal analytics capability to mine the data they have bought to deliver valuable insights. Public clinicogenomic data sources such as AACR GENIE12 and other sources with genomic prevalence are key to keeping costs down.
- On top of demonstrating clinical outcomes, “the hacker” will look to utilize technology, including wearables and mobile apps, that can provide a large quantity of quality of life outcomes data to further differentiate their medicines.
- “The engineer’s” commitment and consistency pay off in this stage of the chain. Their investments mean they have access to diagnostic tools that aid physician decision-making and drive use of their drug, as well as proprietary data to drive market access and marketing efforts.
We are in the middle of a paradigm shift in the access, management, and utilization of healthcare data across the pharmaceutical value chain. Keeping ahead will hinge upon being flexible but targeted in development activities, paying close attention to competitors’ moves, understanding the growing data marketplace, and identifying emerging trends. Achieving the maximum value from the opportunities presented will depend on rapidly selecting and executing the most relevant approach and proactively shaping the landscape through deals and partnerships.
About The Authors:
Jamie Cartland is a life sciences expert at PA Consulting. He works with clients to understand future market landscapes and build strategies to respond to them. You can follow him on LinkedIn or Twitter.
Philip Winkworth is a life sciences expert at PA Consulting. He works with clients to understand the product and technology landscape, how to achieve portfolio optimization, maximize product launch, and how to optimize life-cycle management. You can follow him on LinkedIn.