Open questions1 September 2020
It’s been a slow, piecemeal process, but in recent years regulators have begun to develop a framework for using real-world evidence (RWE), shining a light on the important role it can play in helping to design better clinical trials. Covid-19 has shown what still needs to be done. Irenie Forshaw talks to Greg Koski and Mary Tobin, CEO and chief strategy officer at the Alliance for Clinical Research Excellence and Safety respectively, and Jonathan Fishbein, founder of Veracuity, about how RWE is bringing different groups together in response to the pandemic, and what this means for future trials.
Over the past decade, researchers have started to look beyond the confines of traditional clinical trials and realise their limitations. Meanwhile, advances in technology have made real-world data (RWD) from sources including electronic health records and medical claims more accessible. Thanks to increasing standardisation, it is becoming easier to analyse the data and generate clinically instructive realworld evidence (RWE).
However, there is more to be done, and it is only when organisations come together to effectively integrate their databases that the full potential of using RWD in clinical trials can be realised.
This emphasis on interdependence is at the heart of the Alliance for Clinical Research Excellence and Safety (ACRES). The collaborative non-profit, which is distributing a Covid-19 RWD collection tool, is committed to bringing together stakeholders to share their expertise and build a global system for better clinical research.
“Clinical trials suffer from very limited inclusion and exclusion criteria that define a narrow population,” explains ACRES CEO Greg Koski. “As a result, they may have little or no relevance in a broader population.”
While there are certain questions that can only be answered through rigorous and controlled trials, Mary Tobin, chief strategy officer for ACRES, points to the data they leave behind. “There’s all of these real-world situations that affect patients and don’t get reflected in normal clinical trials,” she says.
For Jonathan Fishbein, co-founder of biopharmaceutical safety and analytics start-up Veracuity – which designed the collection tool – this is where RWE can be useful. “As you look at a larger cohort, trends start to emerge that you wouldn’t see in the very controlled homogenous population that a clinical trial typically has,” he explains. “RWE has a very important place in evaluating products and whether they’re making a positive contribution to patient care.”
Test and test again
Unlike standard randomised controlled trials that aim to establish efficacy, real-world studies are often used to assess effectiveness in large unselected populations that include patients with co-morbidities. “A drug can be efficacious, but you can find it’s not effective,” explains Fishbein. “To be effective is a combination of efficacy and safety. RWE is often used not so much in trying to determine efficacy of a drug in its investigational stage, but after the product is marketed.”
As well as evaluating effectiveness after trials are completed, RWE can help to establish a hypothesis and provide guidance to more traditional research. “If you make sure you’re asking patient-oriented questions in your RWD collection, then it allows you to design better protocols,” says Tobin.
Regulators are gradually showing more acceptance of using RWE to improve the efficiency of clinical trials. “I think the FDA has been very amenable in the 21st Century Cures Act,” says Fishbein. In that legislation, the agency “put together guidelines that encourage manufacturers and drug developers to bring more real-world data into their strategy”.
Still, Tobin highlights an important gap in the guidelines. “They are pretty technology driven,” she says. “If you’re looking at it from a patient side of things, there’s not too much in the FDA guidance that is really trying to follow the patient engagement model. For RWD collection and RWE to be truly effective, taking patient concerns into account is absolutely critical.”
For Tobin, it’s not just patients being disenfranchised that is a problem, but individual physicians too. “The same kinds of issues that plague patients in dealing with large external organisational structures affect healthcare professionals and clinical researchers,” she says. “That’s another thing that becomes important when trying to figure out how to connect these people together.”
This need for a wider structural shift makes Koski wary of overstating the role of RWE by itself. “I think it’s important to remember that it is just one small piece of the dramatic overall change we’re going to see in the entire realm of clinical research,” he says.
Moreover, with traditional randomised trials often seen as the gold standard for drug development, some researchers are cautious of embracing the use of RWE, and question whether it should be used in regulatory decision-making. “How we actually get RWD that is meaningful is one of our great challenges,” admits Koski. “If we truly want to be able to use RWE, we have to develop standards that will allow multiple databases to be connected and integrated. You need massive amounts of data to get really meaningful analyses.”
That can prove difficult. Institutions are often reluctant to share their data due to concerns around privacy and confidentiality. Even when data is collected, there can be issues with its accuracy and completeness because there is no standardised method of data entry. “The standards created by the Clinical Data Interchange Standards Consortium [CDISC] for clinical research were a step in the right direction,” Koski explains. “The standards created for medical electronic health records by Health Level 7 [HL7] were too. Unfortunately, the HL7 standards are not compatible with the CDISC standards and they’re not uniformly utilised across entire populations.”
Both Koski and Tobin are frustrated with the pace of progress when it comes to solving these issues. “The major problem is that we live in both a medical and pharmaceutical world in which there has been an enormous inertia to actually move towards the integration of technology and adoption of standards that is needed,” says Koski.
This is something he partly attributes to conflicting interests in the industry. “There’s a resistance to change,” he says. “It’s trying to achieve a balance between proprietary interests and the mission. If the mission really is going to make medicines available and use information in the most effective way possible, this can come into direct conflict with the mission of making drugs to sell and make money.”
“Not only is the industry saddled with inertia,” adds Tobin, “but it is also relatively inflexible, and so it can’t respond quickly to changes.”
At times of crisis, this lack of agility is problematic as things need to move quickly when responding to rapidly developing situations. “There is a great deal of administrative activity to get a clinical trial started,” explains Fishbein. “At times like this, when we’re learning about Covid-19 week by week, it’s a real advantage collecting RWD as you can look at so many different scenarios. When you’re using RWE, you can adapt your study design as you learn more things. It allows you that flexibility.”
In fact, with little guidance from health authorities about the kinds of treatment and care that patients should be receiving, a space has opened up for collecting RWD to help fill knowledge gaps.
“It sounds terrible,” says Tobin, “but I think there is an opportunity with the Covid-19 virus. For example, there was already a move towards virtual clinical trials. Coronavirus has sped it up tremendously. So Covid-19 is moving certain parts of change forward more quickly.”
That is why ACRES and Veracuity have launched an online survey tool that captures RWD to generate evidence that can enhance researchers’ understanding of the Covid-19 pandemic. Though Veracuity initially designed the survey to collect data about adverse drug events in Covid-19 patients, it quickly dawned on Fishbein and his colleagues that there was an opportunity to find out more.
“We put together a fast, easy-to-complete survey collecting a good deal of critical information about these patients,” explains Fishbein. “We decided to look at risk factors, how patients are presenting, timings between symptoms and hospitalisation, treatments, supportive care and the actual outcomes. It really was mission creep.”
There is no commercial intention to the study, and the tool is publicly available to everyone including researchers, physicians and medical institutions. The tool does not collect identifying information, which alleviates issues of confidentiality, but its development has not been easy.
“A major challenge has been generating awareness,” says Fishbein. “The institutions that we need to fill this out are extremely busy and it’s difficult to get past the gatekeepers. The other problem is that institutions typically want to keep hold of their own data. The standards are not uniform at different institutions, so the idea that we can pool everybody’s data isn’t realistic, because it’s all recorded differently.”
To tackle these concerns, Koski arranged a meeting with Amy Abernethy, the principal deputy commissioner of the FDA, who suggested it would be more effective to work with a consortium to develop a common data and analytic model. This led to the project becoming affiliated with the Covid-19 Evidence Accelerator, an initiative by the Reagan-Udall Foundation for the FDA, in collaboration with Friends of Cancer Research. Its goal is to leverage RWE to speed up the pace at which data can be brought together to help patients afflicted with Covid-19.
“The fact that the FDA is leading the effort to put together this kind of consortium to do RWE speaks to its recognised value,” says Koski. “I think one of the great positives of this terrible situation is that putting together this coalition for the data Evidence Accelerator is something that will serve not just for Covid-19, but the next pandemic that comes along, as well as other health problems.”
Every week, contributing researchers come together and attend meetings to present and discuss findings from different data sources. “I’ve been really impressed with the Evidence Accelerator,” says Tobin. “There’s learning and sharing of the right kinds of data and the mechanisms for doing so.”
“We’ve learned to mobilise and make a joint, rather than competing, effort,” adds Fishbein. “It’s significant that such a wide variety of organisations and agencies have come together cooperatively and are working together relatively quickly. When there is another crisis, the process will be much smoother.”
It is this kind of collaborative approach to clinical research that Koski hopes to see more of in the future. “All of the stakeholders need to come together and work towards fully integrated solutions and unifying standards,” Koski continues. “Of course, that’s what ACRES is all about – that’s what we’re trying to do, build a system that enables that.”
There is a long road ahead to rolling out the standardised methodologies that would allow RWE to fulfil its potential in helping to design more efficient clinical trials and ultimately improve patient care. Large-scale expansion will depend on further regulatory support and multi-stakeholder action. For now, the launch of ACRES and Veracuity’s RWD collection tool and the FDA-led Covid-19 Evidence Accelerator demonstrate that times of crisis can spark action, and that change is indeed possible.
ACRES and Veracuity’s RWE survey
Already validated by a focus group of physicians actively caring for Covid-19 patients, the survey can be completed quickly by healthcare professionals as it requires minimal text entry and uses well-established rating scales that are routinely used in clinical practice. The survey tool captures real-world data to generate evidence to address the most pressing knowledge gaps relating to the treatment of patients affected by Covid-19, such as:
• the impact of the time lag between the onset of symptoms, the time of testing, hospitalisation and the ultimate outcome
• the sensitivity of testing
• treatment outcomes
• risk factors for contracting Covid-19 and dying. Source: ACRES