Aligning real-world terminology: clarification from the FDA on generating real-world evidence from real-world data

“Use of consistent terminology can help the FDA to classify and quantify RWE and promote better understanding of reports by external entities regarding the use of RWE for regulatory purposes.” 

In a new open access commentary, Rahman and co-authors at the FDA seek to provide clarification on when studies with real-world data (RWD) can generate real-world evidence (RWE) on effectiveness and safety of new treatments to support regulatory decision-making. The article entitled ‘When can real-world data generate real-world evidence?’ was published October 19, 2023, in the journal Pharmacoepidemiology and Drug Safety. The article comes at a time when the FDA is seeing increased use of RWE in regulatory submission, but it is often used inconsistently and with stakeholders using varying explanations for RWD and RWE.  

The authors explain that despite several initiatives from the FDA to advance RWE since the passing of the 21st Century Cures Act of 2016, clinical study methodologies have not ‘fundamentally changed’. Randomized controlled trials (RCTs) are still the ‘gold standard’ to generate evidence on the efficacy and safety of new drugs and biological products; however other well-designed studies, such as nonrandomized studies, must also be considered when RCTs are not feasible or considered ethical, and to study more diverse populations, as the authors note: 

“We have encountered a misconception that only non-interventional (observational) research utilizes RWD to generate RWE—in other words, a dichotomy of RCTs versus RWE is said to exist. In reality, the spectrum of study design involves various combinations of data sources and design architectures.” 

The article details the type of study and when they should/should not be considered to generate RWE. For instance, a randomized, point-of-care trial can generate RWE when the primary outcome is measured using RWD. Similarly, in externally controlled trials where the comparator arm is from a RWD source, this will be considered to generate RWE. 

There are also several specific scenarios when RWE is not generated, despite utilizing RWD. For example, when RWD is used in RCTs to identify enrollment criteria, potential participants, or trial sites. Data collected by digital health technologies in RCTs is not considered RWD but is classified as RWD when collected outside the research setting, such as by personal use. 

To illustrate their point, the authors cite two regulatory examples where RWD was used to generate RWE. 

1. Approval of Prograf^® (tacrolimus) in combination with other immunosuppressant drugs to prevent organ rejection in adult and pediatric patients receiving lung transplantation in 2021. As previously reported, the regulatory decision was based on analysis of RWE from non-interventional, observational study, using RWD from a US-based registry. Catch up on Marc Berger’s view of this case here, in his Patti’s People episode.
2. Approval of Ibrance^® (palbociclib) for male patients with metastatic breast cancer in 2019. The approval relied on evidence of drug effectiveness from RCTs in women and safety in men based on RWD from electronic health record and medical claims for men with metastatic breast cancer.  

 The authors also outline the ongoing efforts by the FDA, such as published guidance and the Advancing Real-World Evidence Program, to ensure clarification and consistent terminology is used by all stakeholders. 

“Although the distinction between studies that generate RWE and those that do not may at times seem confusing to some in the stakeholder community, careful consideration of data and design elements can help sponsors and regulators better describe and characterize RWE.”