Talking The Same Language
A universal standard for real-world evidence no longer seems quixotic. But how are companies going to do it?
At a time when capitalism is a dirty word, the notion that private enterprises are willing to muck in and forge a universal standard for real-world data — whereby it can be judged by regulators and used to support regulatory decision-making — is a pure shot of philanthropy.
The value this would bring to healthcare systems is writ large. Real-world insights into the effects of different drugs and diseases could help find a needle in a haystack, build a 360-degree portrait of specific patient populations and make therapies more targeted; supporting and potentially even supplanting the clinical trial process.
Pipe dream? We think not. In our previous article, we caught up with some of the architects of this enterprise and explored the industry-wide implications of achieving it.
Before we get ahead of ourselves, there is a teeny-weeny technicality that needs to be worked out: how the heck are companies going to do this?
“This is a work in progress but when you think about data quality, it must be fit for purpose and so depending on the individual data field that we are talking about, the way to validate quality could be different as well,” says Jacqueline Law, Global Head, Real World Data Science, Roche/Genentech.
She continues. “To give you an example, we work with Flatiron to validate the quality of mortality outcomes. For that validation we use the national death index data from the US, which is considered to be the gold-standard of mortality for oncology patients in the US. We compare flatiron’s mortality data with the national death index data to measure the validity and reliability of the data.”
While data comparison is a general approach, the way “we characterize the data quality would depend on the context of the data and the intended use.”
“This is indeed a technical question, one we’ve spent years working on,and an area that we continue to advance,” says Shane Woods, Vice President, Life Sciences at Flatiron Health.
Quality criteria need to encompass the entire process to generate RWE, from data sources and processing to defining appropriate use cases, he says.
“We have a proprietary method that includes both technology and skilled clinicians based in the US in which we abstract, normalize and de-identify data from the EHR. Our technical approach and centralized oversight abstraction processes ensure that our data are of high quality and that regulators can trust in our methodology and ultimately the product so that findings from our datasets can be used to inform regulatory decisions.”
Engaging in a “rational dialogue” with payers and regulators should inform your methodology, says Anjan Chatterjee, Global Head RWE - COE, Epidemiology and Big Data, Boehringer Ingelheim. Understanding their expectations and devising a methodology that delivers on those expectations is how the industry will move the needle.
Speed bumps ahead.
The obstacles will not be easy to surmount.
The opportunity but also the challenge is repurposing the data that are collected in the real world, says Law. “The data are collected for routine care so may not necessarily have the research quality needed for regulatory decision-making.”
To overcome this problem, “we must ensure that thoughtful statistical analysis plans are developed proactively and that the data source employs mechanisms related to data quality and integrity,” says Woods.
Flatiron also wants to continue to see studies validating the use of RWE to recapitulate known findings - evaluating consistency of controls from clinical trials to real-world external controls, and use of Flatiron and Foundation Medicine’s clinico-genomic database to validate the occurrence of outcomes associated with known mutations, etc.
We must pay heed to the “feedback and eventual formal guidance from regulatory agencies (e.g., 21st Century Cures Act). And finally, we need to be sure that our partners know how to get the most out of our regulatory-grade data sets,” says Woods.
Maintaining close collaborations with key industry players, including researchers, clinicians, regulators, government agencies, health authorities and advocacy groups, to ensure that we are making progress with regulatory-grade data, will also help us reach the top of the hill, he says.
Progress will plod along until one company takes a leap of faith and changes the game for everyone, says Chatterjee. Entrenched processes stifle innovation. Carving out a regulatory-grade standard for RWE will require risk-taking and a radical new way of thinking.
Anjan Chatterjee’s views are his own and do not represent the views of Boehringer Ingelheim.
Since you're here...
... and value our content, you should sign-up to our newsletter. Sign up here