A new study from Roche finds concerns around real-world evidence as a comparator to single-arm trials may be unfounded.
Over the past decade or so, real-world evidence (RWE) has emerged as an extremely important tool for bringing innovative treatments to patients.
Just like with any new concept, however, regulatory and reimbursement stakeholders have asked for clear evidence that the reliability of studies based on RWE can equal (or exceed) that of existing research methodologies, most notably the randomized controlled trial (RCT).
New research from Roche, one of the largest and most influential life science companies in the world, is making a very strong argument in RWE’s favor.
In a relatively rare subpopulation of patients carrying the ALK driver mutation in advanced non-small cell lung cancer (NSCLC), Roche performed a single arm trial for the targeted therapy alectinib. In oncology, it is relatively hard to recruit patients with driver mutation positive disease into trials, so the purpose of Roche’s trial was to try to provide this medicine faster to patients who urgently need it.
Whilst this trial led to regulatory (FDA and EMA) approval, comparative evidence was needed for health technology assessment (HTA) for reimbursement. Roche provided this evidence to HTA agencies, but it was rejected due to concerns of unmeasured confounding and missing data in RWE.
Revisiting the decision, in a recent study, the Roche team used quantitative bias analysis to assess the degree to which unmeasured confounding or missing data in RWE could nullify the finding of comparative benefit of alectinib. The researchers were able to show that implausible levels of bias would have to exist for alectinib to not provide benefit to patients. The study received an editorial from Prof. Julian Hong at UCSF, who described how the Roche team had gone to painstaking efforts to address concerns with RWE, giving great confidence in the results.
This is big news for RWE.
The FDA and European HTA Agencies want to be as careful as possible when using new research methodologies that will have a direct impact on patient safety, but Roche’s findings should give them a gentle nudge to reevaluate their confidence in RWE. If we minimize the role of RWE in regulatory decision making, we risk delaying the delivery of beneficial treatments for patients.
The ability to quantify bias present in RWE has value and utility in fast, accurate, safe, and effective clinical research. In the future, using this technique will make it possible to see how RWE studies could become a viable substitute for some interventional research.
Study author Sreeram Ramagopalan shared,
“Rarely do industry sponsored RWE studies receive any special attention, so I am really proud to see our work applying quantitative bias analysis to a single-arm trial real-world data comparison receive such broad recognition. This was a lot of hard work from some of the smartest people I’ve ever worked with, so I am very happy to see it rewarded.”
At COTA, we are excited about these implications. We see a starring role for RWE in the future of clinical research and are glad to see that other leaders in the field are working to build up support for RWE in the United States and around the world.
