The use of real-world evidence (RWE) to support US regulatory submissions often suffers from a lack of data reliability or clinical relevance, but there have been some success stories where RWE was considered fit for use.
During a session at RAPS Convergence 2022, speakers shared cases where RWE supported regulatory decision making for marketing applications for a new indication or to answer post-marketing safety questions. These examples are not easily replicated but there are lessons about the importance of data quality and study design, according to John Concato, MD, associate director for real-world evidence analytics in the Office of Medical Policy at the FDA’s Center for Drug Evaluation and Research (CDER).
“It’s not easy but it’s fundamental,” he said. “Do the fundamentals right and you’ll be in much better shape.”
When FDA receives an application containing RWE, officials consider whether the real-world data (RWD) in the application is “fit for use,” whether the trial or study used to generate the RWE can provide adequate scientific evidence to answer the regulatory question and whether the study conduct meets FDA regulatory requirements, Concato said.
Concato offered an example in which FDA approved a supplemental New Drug Application based on a non-interventional RWE study. Prograf (tacrolimus) was originally approved for prophylaxis of organ rejection in patients receiving liver transplants based on randomized controlled trial evidence, and the drug was widely used in clinical practice for lung transplants. In seeking a new indication related to lung transplants, the sponsor, Astellas Pharma US, relied on RWE since a randomized controlled trial would not be feasible.
The RWE study was a non-interventional treatment arm versus historical control data, looking at cumulative incidence of graft failure or death at one year as the primary endpoint. The data came from the Scientific Registry of Transplant Recipients, which has data on all lung transplants performed in the US from 1999-2017. The sponsor also submitted the analysis plan and the patient-level data to the FDA.
“The Scientific Registry of Transplant Recipients had very strong data reliability and relevance and we felt it was similar to what would have been collected if an interventional study would have been done,” Concato said. “We determined this to be, for the first time at CDER, an adequate and well-controlled study that wasn’t a clinical investigation.”
It may be hard to replicate this study exactly, Concato said. In general, the agency has seen issues with real-world data sources, such as a lack of relevance to the clinical question, data reliability issues, missing or mistimed data, a need for linkage to other data sources and a lack of suitable capture of endpoint. Other problems may relate to non-randomized study design, including the threat of residual confounding, problems with the index date, and use of an inappropriate comparator.
Additionally, Concato said there is some skepticism around non-randomized studies that do not have sufficient confirmation of a prespecified protocol and analysis plan. “We need confidence from you all that we’re not seeing the 100th analysis,” he said.
Orencia
FDA also approved a new indication for Orencia (abatacept) based in part on RWE from a registry-based observational study. Orencia was first approved to prevent acute graft versus host disease (aGVHD) after hematopoietic stem cell transplant (HSCT). In December 2021, FDA approved a supplemental Biologics License Application (sBLA) for a new indication for the prophylaxis of aGVHD, in combination with a calcineurin inhibitor (CNI) and methotrexate (MTX), in adults and pediatric patients at least 2 years old undergoing HSCT from a matched or one allele-mismatched unrelated donor.
The supporting evidence for the application included a phase 2 trial (ABA2) with two cohorts, including a randomized controlled trial and a single-arm trial that was compared with a prespecified matched cohort from the Center for International Blood and Marrow Transplant Research (CIBMTR) registry. Additionally, the application included a retrospective observational study drawing on data from the CIBMTR that validated the ABA2 results. The efficacy results from the real-world observational study were included in the clinical studies section of the product label.
“Data is paramount,” said Kerrie L. Nagrod, a senior director and global regulatory lead at Bristol Myers Squibb (BMS), who presented the Orencia case. The CIBMTR registry includes data from every allogeneic transplantation performed in US, allowing validation of data reliability by comparing to patients in the phase 2 clinical trial.
Communication and transparency were also essential, and BMS had multiple formal and informal interactions with FDA during the process. “We were able to share what our data was and what it wasn’t,” Nagrod said.
Heplisav-B
In 2017, FDA approved Heplisav-B, a two-dose hepatitis B vaccine containing a novel adjuvant. Though the vaccine showed a higher seroprotection rate over the traditional three-dose vaccine in clinical trials, one trial also showed a non-statistically significant numeric imbalance in acute myocardial infarction (AMI) events in those receiving the two-dose vaccine. The sponsor, Dynavax, proposed a large observational post-marketing vaccine surveillance study to resolve the AMI safety issue.
FDA determined that conventional clinical trials, spontaneous post-marketing surveillance, and the FDA’s pharmacovigilance system would be insufficient to assess the risk at the time of the product’s approval and required a post-market RWE safety study, explained Michelle Pernice, PharmD, vice president of regulatory affairs at Pardes Biosciences, who was working at Dynavax at the time of the Heplisav-B application.
The final study design was a pragmatic, prospective, cohort, noninferiority study that collected electronic health record data on vaccine distribution as part of routine care, looking at the outcomes of patients who received the Heplisav-B vaccine and those who received the three-dose vaccine. There were more than 30,000 participants in both vaccine groups, and the researchers observed that the two-dose vaccine was not significantly associated with an increased risk of AMI.
When considering whether there is potential to use RWE to contribute to regulatory decision making, think about the endpoint, Pernice advised. “When I see a hard endpoint – a gold-standard endpoint like death, like acute myocardial infarction, like hospitalization – things the electronic medical record is going to pick up, to me that’s fodder for generating real-world evidence,” she said.
To continue reading this article please go to RAPS .