The trend of using real-world evidence (RWE) in the healthcare industry continues to grow. But perhaps it has never been more important than it is now during the current pandemic. Precisions Dan Danielson and others discuss the benefits to be gained from using RWE in the search for potential diagnostics, drugs and vaccines for COVID-19.

Real-World Evidence Has Crucial Role To Play in Race to COVID-19 Approvals

While the use of real-world evidence (RWE) has grown within the health care industry, perhaps it has never been of more importance that it is in the race to find potential diagnostics, drugs and vaccines for COVID-19. And the FDA, no stranger to using RWE in the drug approval process, has signaled its willingness to use a wide range of data in its consideration of these products. Randomized controlled trials still must be conducted, say industry sources, but RWE can help speed that process.

Back in May, the FDA released a statement from Amy Abernethy, M.D., Ph.D., principal deputy commissioner of food and drugs in the agency’s Office of the Commissioner, on the FDA’s use of “diverse streams of data” to help assess and accelerate approval of potential COVID drugs, diagnostics and vaccines.

“The FDA continues to work around the clock to respond to the COVID-19 pandemic,” said Abernethy. “As part of this effort, we recognize the potential for diverse, real-world data sources such as electronic health records, insurance claims, patient registries and lab results to further inform our science-based, all-of-America response to this unprecedented public health emergency.

“In recent years, the agency has taken steps to leverage modern, rigorous analyses of real-world data — such as data from the health care setting — to inform our work,” she continued. “The COVID-19 pandemic has brought an urgency to these efforts and the FDA has worked quickly to advance collaborations with public and private partners to collect and analyze a variety of real-world data sources. Evaluation of real-world data has the potential to provide a wealth of rapid, actionable information to better understand disease symptoms, describe and measure immunity and understand available medical product supplies to help mitigate potential shortages. These data can also inform ongoing work to evaluate potential therapies, vaccines or diagnostics for COVID-19.”

FDA Is Working With Various Sources

Abernathy said the FDA had entered an agreement with Aetion Inc., a company with software that analyzes real-world data, to address questions around drugs and diagnostics used with COVID, as well as risk factors for the illness in various patient populations. And in addition to using programs already available before the pandemic, such as the FDA’s Sentinel program, as a data source, the agency is partnering with the Reagan-Udall Foundation for the FDA and Friends of Cancer Research on the COVID-19 Evidence Accelerator, a website where sources can share real-world data on drugs and diagnostics.

According to Nancy Dreyer, Ph.D., chief scientific officer for real-world solutions and a senior vice president at IQVIA, “real-world evidence can bring a couple additional dimensions to the process” of assessing potential COVID products. “Its use allows regulators to see what’s happening to people getting vaccines,” which can help provide a benchmark for what’s happened in that population vs. nonvaccinated people.

“RWE, when collected in a rigorous fashion (fit for use) and analyzed scientifically, can speed building out a set of evidence that can quickly describe the natural history of a disease — how does COVID spread, what are ways to interrupt the infection cycle, etc. — and determine whether or not currently available drugs are useful for treating SARS-2 CoV infections or managing COVID-19 symptoms and sequela,” points out Dan Danielson, R.Ph., senior director of the access experience team at PRECISIONvalue.

Matthew Reynolds, Ph.D., vice president of real-world evidence at IQVIA, says there are a couple of reasons why it’s important for the FDA to use RWE in assessing potential drugs, diagnostics and vaccines. “First, with some of the things that were treating COVID-19 symptoms early on, such as hydroxychloroquine, real-world evidence was really the only thing we had,” he points out.

RWE Can Help With Enrollment in Trials

“It allows us to ask questions in pretty much real time,” says Reynolds, who also is an epidemiologist. “It’s given us a lot of answers.” There are so many different types of RWE and such a large amount of data that investigators “can look at it in different ways and find sites, physicians and patients” based on where physicians are treating COVID patients. “This helps with patient enrollment in trials” and also can help with postmarketing commitments.

In addition, he says, “while rushing to get trials underway,…there will be populations not enrolled in studies, like pediatrics.” The use of RWE can provide insight into these patient populations.

Beyond that, says Jeffrey Stoll, KPMG U.S. strategy leader for life sciences, “clinical trials don’t always capture how drugs and vaccines are going to react in the real world, where people often have co-morbidities and take medications for them. Many of the patients in clinical trials are often healthier than the patients who will actually need the medication once it is approved. Randomized control trials are not going to go away since that approach is the gold standard to determining the safety and efficacy of medications. Real-world evidence can be gathered and applied to these studies, much like how the data generated from these studies can be applied to determine which patients benefit the most from a particular course of treatment.”

Danielson agrees. “RWE allows the collection of data on the use of medication by patients who are not represented in the clinical trials that lead to the medication’s approval. RWE also allows the collection of data on medications that are in off-label use. This can highlight where existing medications may be safe and effective for new uses, as well as demonstrate where particular medications are not useful for diseases that may be highlighted in the lay press.”

“The ability to gather data from biosensors in real time and apply analytics to a variety of data streams gathered not just at clinical trial centers but via remote monitoring can uncover a number of ways medications can be made safer and more effective,” contends Stoll. “Genetic and biomarker information adds another variable that should be accounted for in clinical studies, adding a layer of complexity. However, that data can also act as a filter about the patients that could benefit the most or potentially suffer harm from a particular course of treatment.”

Dreyer maintains that RWE “bookends randomized controlled trials. It’s like the first and last chapter of a book.…If you think about COVID, there’s so much we don’t know, and this is where real-world evidence” is filling in the gaps. For example, while not a drug, RWE has shown that “putting patients prone is better than putting them on a ventilator.”

“Certain elements of a clinical trial could rely on the collection and analysis of data extracted from medical claims, EHRs [i.e., electronic health records] or laboratory and pharmacy databases — sources outside of those conducting the studies,” points out Stoll. “Researchers could collect other data specifically for the trial, such as results of exercise stress tests or radiographic analyses, using methods typical of a traditional clinical trial.”

Indeed, the FDA is no stranger to using real-world data in its approval process and at times has leaned heavily on it. For example, the FDA initially gave Pfizer Inc.’s Ibrance (palbociclib) accelerated approval in February 2015 for use in combination with letrozole to treat postmenopausal women with estrogen receptor-positive and human epidermal growth factor receptor 2-negative (HER2-) advanced breast cancer, and the agency has expanded its label multiple times. In April 2019, the FDA approved it for use with an aromatase inhibitor or fulvestrant in men with hormone receptor-positive, HER2- advanced or metastatic breast cancer. The approval was based on EHR data and postmarketing reports on the drug’s real-world use from IQVIA, Flatiron and Pfizer databases.

The application for EMD Serono, Inc., a unit of Merck KGgA, and Pfizer’s Bavencio (avelumab) for Merkel cell carcinoma included a comparator arm based on a retrospective analysis of data from McKesson U.S. Pharmaceutical and Specialty Health’s iKnowMed EHR. Amgen Inc.’s supplemental application for Blincyto (blinatumomab) for Philadelphia chromosome-negative relapsed or refractory B-cell precursor acute lymphoblastic leukemia also included patient-level data for 694 historical controls. And Janssen Pharmaceuticals, Inc., a Johnson & Johnson unit, received FDA approval to expand the label of its antipsychotic Invega Sustenna (paliperidone palmitate) to include real-world data on its use vs. seven other oral antipsychotics.

RWE Options, Timeliness Have Improved

“The FDA has been using real-world evidence for a long time; it’s not new,” says Reynolds. “What’s new, in my mind, is the number of options and the timeliness of the data,” which can be obtained within “days and hours.”

Some challenges, though, exist with RWE use, not the least of which revolves around patient privacy.

“Patient privacy is paramount. In working on a real-time patient registry of people with COVID,…the biggest issue we’ve had is patient privacy,” Reynolds tells AIS Health. All the entities involved, including pharma, the FDA and consulting groups, must protect people’s privacy. “I have yet to see a lapse in this.…We try to do the best science we can within the rules and regulations around patient privacy.”

According to Dreyer, “it can be frustrating when you can’t get at the data you want” because of patient privacy or other reasons “or because it’s not there. In underserved communities, such as the homeless and indigent populations, it’s hard to get data” simply because it doesn’t exist.

“Finding data fit for purpose” can be a challenge, agrees Reynolds. And even if investigators can get the data, what if they want to ask questions about it, such as what a person’s blood pressure was or whether he or she was hospitalized? “That information is not there” sometimes. “Long-term safety and effectiveness data may require a different data set” for a patient. “It’s kind of an art to find the right data sets.” It’s not “one-size-fits-all.” It comes down to “understanding the differentiation between data types.…It’s impossible to know every single data set.”

Randomized controlled trials remain the gold standard within the product approval process. But they also have their limitations.

“Clinical trials can be costly,” which in turn can “drive up the cost of medications,” says Dreyer, who also is an adjunct professor of epidemiology at the University of North Carolina School of Global Public Health.

In addition, “clinical trials are valuable, but they are slow,” she tells AIS Health. For example, “the Hong Kong flu of 1968 posed a big pandemic threat. The FDA had approved the drug amantadine as a prophylactic” for the H2N2 flu in 1966, but the pandemic strain of influenza was H3N2, so further clinical trials were required for amantadine for prevention of the H3N2 strain. “By the time those trials were done, the pandemic was over,” she notes.

Conversely, with the swine flu in 2009, “an early emergency use authorization” was given to Tamiflu, which “proved to be very useful.”

“Clinical trials are very well-defined and have very focused environments. They have lists of inclusion and exclusion criteria,” states Reynolds. “They are well-controlled environments.” RWE, however, can demonstrate what a product looks like in a broader population. In addition, “trials only tend to go for a period of time. Real-world evidence allows us to follow people for longer and evaluate many more people.” But clinical trials “are critical for the approval of products; they’re paramount in the drug approval process.”

Asked if there is anything that various stakeholders can do to speed up the use of RWE in the FDA’s assessment of potential COVID products, Stoll responds that contract research organizations “have emerged as fairly sophisticated users of real-world data, and they can have a role in helping not only with the trial design but also to help drugmakers decide whether or not a trial should continue. Drugmakers want their failures to end early in the process instead of becoming an extremely expensive failure in Phase III or a product needing to be recalled.”

Clarity Is Needed on Expectations

In responding to the same question, Dreyer points out that “in the new world, no one knows what will be expected of them once vaccines are launched” because “regulators have been silent on what they’re looking for” in terms of post-launch studies. “We want to have those expectations in place well before a vaccine comes to market.…If regulators, states and others would be clearer on what kind of evidence they want to see, pharma would be all over it,” but there is the “absence of a roadmap.”

“Putting together real-world data takes time to get through; some steps can’t be compressed more than they are now,” or you could “lose data quality,” adds Reynolds. “There are so many unknowns,” he says, including how providers will be billed, what entities will provide vaccines and therapeutics and what kind of codes will be used to bill these products. “Right now, it’s hard to know where to find data” that will be most useful in determining such issues.

Following the FDA’s willingness to step up its use of RWE during the pandemic, might this impact the agency’s use of it post-pandemic? RWE will “not replace standard trials, but the FDA and pharma companies will continue to use it,” says Reynolds. “There may be new opportunities.”

“As the FDA gains experience in managing and using RWE during the pandemic, it will learn how to develop better data structures that can be readily used to address new clinical questions, develop processes that ensure the transparency and sharing of RWE among collaborating organizations and agencies and assure appropriate use,” Danielson tells AIS Health. “Hopefully, it will also strengthen the relationships among collaborating groups.”

The experience, says Stoll, may result in the FDA’s becoming “more receptive [to] different clinical trial and approval strategies. COVID has allowed for more creativity and out-of-the-box learnings.” However, he cautions, “we should note we are still mid-process [in] this experience. If any of the vaccines or therapeutics that have received accelerated approvals ends up causing systemic issues, the FDA may swing back and become much more conservative again.”

Ultimately, he says, “much like telehealth got a big boost as a result of the need to change care delivery from the pandemic, clinical trials will need to undergo [their] own transformations to use more virtual care and real-world evidence when the science and patient circumstances require it.”

Contact Danielson via Tess Rollano at trollano@coynepr.com, Dreyer and Reynolds through Nicole Cohen at nicole.cohen@ogilvy.com and Stoll through Bill Borden at wborden@kpmg.com.

by Angela Maas

AIS Health