Executive Roundtable: The Rise of RWD in Clinical Research

Earlier this year, a panel consisting of leaders from the clinical research industry convened for a virtual roundtable to discuss the emerging uses of real-world data (RWD) drug development.

The group discussed a variety of topics, including patient centricity, regulatory concerns, underrepresented populations in clinical trials, and their own experiences with using and/or managing RWD.

Andy Studna, Editor, Applied Clinical Trials: What is the top RWD use you have seen in practice in clinical trials this past year?

Gabriele Brambilla, CEO, Alira Health: From my experience, the use of RWD has been very important and more popular now in two areas.

The first one is the use in external control arms; the ability to use RWD to accelerate the clinical trial process using medical care data or data coming from the real world to be able to simulate the control arm.

And the second one, which is very exciting—but not a lot of people are talking about—is all the new software and algorithms coming out for biosimulation. What we use to train these systems and algorithms is RWD. It is electronic health record (EHR) data. And that allows the development of systems and algorithms that can help develop clinical trials or biosimulation for PK/PD (pharmacokinetics/pharmacodynamics). So that’s exciting, and I see more and more of that in the market right now.

Amy Cramer, Steering Committee Board Co-chair, HL7 FHIR Accelerator Vulcan: What I’m really excited about is an organization called Vulcan. It’s the HL7 FHIR (Fast Healthcare Interoperability Resources) Accelerator dedicated to both translational and clinical research. Vulcan is made up of 40-plus international members representing various stakeholders from the research community, including patients, sites, sponsors, regulators, vendors, and more. In order for RWD to be functional, there must be consistent use of standards because data mapping and harmonization is not scalable. Vulcan’s members and collaboration with other organizations is exciting to me because standardization of data is how we’re going to optimize efficiencies out of all the data that’s available.

Jill Pellegrino, Vice President of Recruitment and Real-World Evidence (RWE), CVS Health Clinical Trial Services: At CVS, we have access to millions of lives in our data, so access to pharmacy records and claims information on these individuals. And we’ve been using that data to reach out to the patients to invite them to participate directly into research and to trials. That’s been powerful because a lot of what I see in RWE and use of it is taking data sets from different sources and bringing them together to glean insights from it. But we’ve seen a lot of success in engaging the patient directly as a participant in that. It allows us to collect more information and then also to bring their voice into the research, which has been powerful. Further, it’s allowed us to bring people to trials and research studies that they wouldn’t have had access to before and help with enrollment challenges that we see in the industry.

Craig Serra, Head, Commercial Solutions and Partnerships (clinical research), Flatiron Health: The top use case that I’ve seen over the last year is figuring out what kind of patients those sites have historically seen and if there are current patients eligible for a trial at those sites. Then, when you do that last mile connection, you’re starting to see magic happen between patients and sites within the clinical research world.

Alecia Clary, Founder and CEO, Evidence to Practice: I’m most excited about watching people embrace remote trials, hub and spoke trials, and using RWD to identify not only where patients are but who are the patient populations that most need the therapeutic or the tool, and making sure that recruitment efforts align with those populations. This includes efforts to incorporate patients and data from community centers and other places where they patients may not traditionally have an opportunity to participate in trials.

What challenges and/or successes have you seen in some of these RWD practices?

Serra: From a pure RWD perspective, the timeliness of the data and the contemporaneous nature of it, are we getting to patients soon enough? Do we have enough “real-time data?”

Brambilla: We have a lot of partnerships with patient advocacy groups, who are really the ones helping to build that trust with the patient. But sometimes it’s not just the method of educating them on the importance to be part of the study. It’s educating them on how to get access on their data. For example, at Alira we believe that the patient is the carrier of the longitudinal information. Why don’t we focus on the patient and not on the provider?

Right now there is a new regulation from the [21st Century] Cures Act that patients can have access to their electronic medical records (EMRs), but not everybody understands what an electronic medical record is. Not everybody knows how to go on a patient portal and download it. Not everybody is willing to acquire it.

Also, EMRs are not available to everyone because many patient portals have not yet been updated to allow downloading of [electronic health records] and EMRs.

There’s also a knowledge barrier of where can I get my information, and how can I download my information? Now hopefully that will change in the next couple of years. But it’s a process for them; it’s a process for us, but an exciting one because the changes in the regulations are allowing patients to be able to get access to that very valuable information.

Pellegrino: One of the challenges that we’re working on is how to get more in-depth or robust data in populations with more complex disease states. Areas like oncology and immunology, you need a lot more data where you can reach out to patients and make sure the trial is relevant for them at that point in time.

Clary: One of the most pressing challenges for the work that I’m doing right now, is missing data.

We’ve found that, for example, race and ethnicity data are missing in 30% to 70% of clinical RWD sets. So, if you’re interested in doing a targeted outreach to a potential patient pool, there may be several patients that you’re missing. Also, are patients for whom race and ethnicity data are missing systematically different from patients who are willing to report that data, for example?

The only people who show up in these clinical data sets are people that have access to care at these sites. There are large amounts of patients who are probably not represented in these data sets and could benefit from and be interested in participating in clinical trials or having their data leveraged for the generation of RWE.

Cramer: Global differences in data collection, including variability in the documentation of data domains and some data coming from paper sources, as well as the inability or inconsistency of data interoperability are challenges to optimizing RWD. Country-specific and regional variances in regulations, data access, patient trust, and more creates different operating landscapes that are not always easily comparable. To enable more effective data comparisons and integration globally, we need to establish and adhere to a single set of standards across both clinical care and
clinical research.

How does RWD help advance approaches in patient centricity?

Clary: Using RWD really does allow more people to benefit from participating in research than people who typically participate in prospective clinical trials. And so, consequently, they also get to participate in the benefits of research.

Postmarketing safety studies are a well-established and robust example of how real-world data can facilitate our ability to understand the real-world performance of therapeutics for people, again, who have access to them. But, as several of my colleagues have mentioned, there really is an opportunity to improve our ability to link data sets; increasing interoperability is really an opportunity for us to not only link some of our clinical data sets but also to go beyond that and link more nontraditional data sets to begin to incorporate additional data about these patients.

Serra: When I think about patient centricity in clinical research, I’m thinking about studies designed and executed in a way where the sample represents the population with a particular disease. RWD is tremendously effective at helping us accomplish that, and inherently when a population uses that drug, you have exponentially more confidence that your safety and efficacy evidence is the highest quality possible. I also think about patients who may be eligible and want to be in a study, but the trial burden is too much; RWD is able to help that aspect tremendously as well.

Pellegrino: I found incorporating patient centricity is very hard to do on a global or protocol level. It’s best to make the experience and the options for patients most flexible. At CVS, we are bringing trial sites to our MinuteClinic footprint across the US. And the strategy and goal there is to provide an environment and an opportunity for more patients to participate in trials in a way that’s more flexible and convenient for them. In the data we’re collecting at the site level, there are insights that we can act on at that local level to help improve patient experience and increase the level of engagement.

Cramer: The promise of RWD allows for flexibility in study design such as decentralized clinical trials, local care models, or claims/clinical data in an interoperable way. When we look at protocol development, there must be options for patients in how they want to participate in clinical research. Some people enjoy coming into the site and having that interaction. But where is the site? Are participants willing to travel a long distance, or do they want to travel to a closer location?

If we’re really going to be patient-centric, we have to give flexibility at the level of patient interaction. It’s going to make it more nuanced, yes. But I think in the end, this approach will allow sponsors to provide a better patient experience, the study will be more inclusive because we can offer access to patients from underrepresented communities, and we’ll likely help boost retention in the study. RWD is an enabler to allow that flexibility.

Clary: That also brings to mind opportunities for remote patient monitoring, for example, using some digital tools. There are opportunities to collect some data from the patient’s home so that they don’t have to travel to the trial site no matter how far it is.

Brambilla: What kind of information can I give back? What type of technology can I give to them (patients) that will help them manage their condition while they’re going through a collection or they’re providing data for clinical research? That is the right approach because if you use them only as, “Oh I need them to collect data,” we’re not going to get their compliance. So thinking about what can you deploy as a technology for them that can help them also manage that period of their life, their emotions, their conditions, that’s the key to get compliance, to get participation. That’s what companies should focus on.

How does RWD help improve clinical trials among underrepresented populations?

Pellegrino: The goal at CVS was to bring trials to underrepresented populations that don’t always get a chance to participate. We rely on RWD to inform that strategy in terms of the locations and patients with disease areas that would benefit from trials.

Serra: With the advantage of having an oncology-specific EHR embedded in hundreds of community practices, of which there’s a good portion of them that also practice research, we’re simply able to get to where patients actually get care. And many of these patients simply can’t make the trek to a big-named university, academic medical center, or large hospital system that’s a couple hours from them, despite being diagnosed with cancer. Logistics aside, our historical study design and execution ecosystem vastly overestimates capabilities and knowledge of clinical research that your average patient has. With RWD, we meet patients where they are at.

What are the steps we need to take to unify data sets?

Cramer: I believe there are three main components. One, we need to agree on standards because right now clinical research and clinical care don’t typically use the same standards. Second, we need to reduce variability in the way that data is collected—semantic interoperability. What I mean is when we collect data, we need to have the same definition of what we’re collecting. What does a rescue drug mean when I’m a nurse as opposed to what does a rescue drug mean when I’m in a clinical trial? We need alignment, and we need to make sure that we agree on the way data is collected. How many years have you been a diabetic vs. what was the year that you became a diabetic? It’s a small thing, but it makes a big difference in the end. We also need to collect data at the point of care so we’re not requiring another step for patients to complete.

Lastly, we must enable multidirectional data flow. Data should come from the physician to the [principal investigator], to the sponsor, to the patient, and to the regulator—flowing in multiple directions in near real-time. Imagine, a triangle with the site, sponsor, and patient at each corner with data flowing in multiple directions and regulators positioned in the center of the triangle.

Clary: The subtext to the conversation about linking data sets and interoperability is also the data holder’s knowledge or interpretation about existing laws and regulations governing it. My experience with multi-sponsored studies suggests that data partners may participate in the same study but interpret laws and regulations differently even when they’re operating within the same jurisdiction. While we’re aligning on definitions and the data elements that we’re collecting, we also need to align on how we interpret laws and regulations that governs data sharing.

What do you think is lacking to further industry adoption of RWD?

Cramer: I think it’s just time and change management. We have the pieces of the puzzle, and what isn’t mature yet is being developed. We have stakeholders across the research community who are interested in making RWD the industry norm. What I think is really interesting and exciting is that we’re taking learnings and best practices from recent experiences like the COVID-19 pandemic and driving toward a standards-based interoperable exchange of RWD. The next step is change management and incentives to drive adoption.

Clary: I share Amy’s optimism and enthusiasm and if we can keep that same energy up around the use of RWD as a whole and clinical research as a whole—having more cross-industry conversation about what works and what doesn’t work; having conversations about what is acceptable; and where we have some work to do in terms of acceptability—there are a large amount of opportunities for stakeholders.

Pellegrino: I think we’ve seen a huge increase in pharmaceutical companies that are trying to use real-world data to make clinical trials more efficient to capture greater insights. There’s a lot of momentum and motivation to do this. It’s just a matter of working through the challenges and trying to navigate the data options that are available and finding sufficient data to do those analyses.

Brambilla: Right now in pharma, the amount of money spent in real-world evidence research matched the actual amount of money spent in clinical research in 2021, which is great. The amount of money spent on real-world evidence will be more than the amount spent in clinical research. The importance of proven effectiveness and not just safety and efficacy, that has become a very important point through COVID. I share the positivity and the trend is there. You see pharma companies are spending a lot more money in this particular discipline. That’s a key factor for sure.