There can be no doubt over the transformative potential that real-world data (RWD) and the real-world evidence (RWE) derived from them can bring to the industry today and especially in the near future. It is a space that is a hot topic of discussion and this discourse only continues to increase as the toolset’s potential comes to be more fully realised with time. Dr Elliott Levy, Senior Vice President of Global Development at Amgen, has been one of the tool’s most vocal proponents. Speaking to Pharmafocus, he extolled its many merits and stressed the breadth of benefits its use can bring to the industry: “RWD and RWE are poised to transform how we conduct clinical development,” he said. “They have the potential to dramatically accelerate the acquisition of useful knowledge about the performance of our products in different types of patients. They will help us to get new drugs on the market more quickly, expand their approved indications more quickly, and inform the safe and appropriate use of medicines. This is already starting to happen, and the trend will accelerate in the years ahead.” While the use of RWD is already quite prevalent in the industry and has been, in some quarters, for quite some time, wide adoption is still nascent and much of its true potential is still to be fully realised. It certainly could mean big things in the future, but where does RWD fit on the industry’s horizon? A study by the Tufts Center for the Study of Drug Development (CSDD) set out to find out just that. The study, published as part of the Tufts CSDD Impact Report November/December 2017, sought to quantify and illuminate the RWD intentions of major industry players over the next three years until 2020. Pharmafocus spoke to Dr Mary Jo Lamberti, Senior Research Fellow at Tufts CSDD and lead investigator on the study, to get the details of what was uncovered: “We conducted a web survey and each respondent represented a different company. We had 17 large sponsors and CROs and the remainder were mid-sized and small companies. We also had 14 of the top 30 pharmaceutical companies represented,” she explained. “The key findings were that 29 out of 30 companies in the survey reported that they have a RWE function, and most companies indicated that their RWE functions were increasing size, so that shows that the uptake is growing. “From a post-approval regulatory and labelling perspective, there were two primary areas for the use of RWD to generate evidence: for post-approval safety studies, including decreasing the severity of a label warning or to support risk evaluation and mitigation strategies. The other was to inform post-marketing studies. “A lot of companies are using some of the data sources in support of a new drug application – demographic data, claims data, patient-reported outcomes data, biomarkers or genomics data. That helps with how a new molecule might work for different patients, and at different stages of a disease, or in conjunction with an existing therapy.” To determine exactly how the respondents planned to expand their RWE operations, the study looked at a number of predictive metrics which reveal an illuminating picture of how the area will develop in the near future: “What we did is that we asked about full-time equivalents (FTEs) and asked companies to project out in terms of their staffing needs,” Dr Lamberti outlined. It was found that respondents planned to expand their FTEs devoted to RWD by 25.1% overall in the next three years – by 34.5% in large companies and 16.8% in small and medium-sized companies. “We also looked at the skill sets that companies are requiring or are looking at in terms of the most important training for real world evidence staff, and the top areas were epidemiology, statistics, mathematics and economics,” she continued. “But there was also qualitative research and communications skills. So it’s an interesting mix of quantitative and qualitative skills that organisations are looking at when they’re talking about adding to their staff. “We also looked at budgets,” she added. “We didn’t get exact numbers, but we looked at how budgets were allocated, and what we found was that their budgets were increasing and they were looking at spending money in a number of different ways: in analytics, as well as research studies and data.” These findings are enlightening to say the least, but while some of the findings were perhaps to be expected in some regards, the report also unveiled plans from some respondents to expand their RWE use in a lesser-talked-about and, so far, only tentatively explored areas. “Social media is seen by the companies in our survey as an area where the use of data sources to support an NDA to a regulatory authority will increase,” Dr Lamberti revealed. “The way we see that is that it would really help give patients more of a voice and capture their experience, whether it be patient information sharing networks or Facebook. But that’s really a projection for the future because there are lots of challenges. For example, the data is in an unstructured format, so it’s hard to mine or abstract. And then there are issues with data validation and privacy issues – there are still obstacles there.” Reinventing the wheel ** While RWD and RWE have a number of well-utilised applications that can reveal insights which are sometimes just not possible through traditional clinical testing, the humble randomised clinical trial is celebrated by the industry as the gold standard in research and development efforts and will remain so. Leveraging the two tools in harmony can allow for much more robust evaluation of safety and efficacy, but the insights gleaned from the analysis of RWE can also be applied to the design of the clinical trial itself, as Ludovic Helfgott, Global Vice President, Cardiovascular, Renal and Metabolism at AstraZeneca, explained to Pharmafocus: “If you are using RWE in a very in-depth, epidemiological study, you can certainly design your clinical trial, and especially the inclusion criteria, in a much smarter way because you have a much better and up-to-date standard of care,” he remarked. “In the development of randomised clinical trial design, it is very important to know exactly what will be the population that you will include in your trial, and to do so you need to find the best patients that will benefit the most from your medicine or treatment. These data can change quite fast: if you look at developing markets such as the Chinese market where the prevalence of an incidence of metabolic disease has changed over the past 20 years, RWE gives you an accurate sense for these data. “Likewise, in the more established markets, the prevalence of diseases after the bulk of the population is treated with a good standard of care might change your assessment of individual risk. If you want to design your randomised clinical trial smartly to identify the right sub-population that can really benefit versus those that can’t benefit, having your RWE helping you can be a real plus.” Amgen’s Dr Levy agrees: RWD and RWE can be utilised to great effect to refine and hone the investigational frame of a clinical trial to streamline operations and ensure the greatest utility of the findings it will seek to generate, as he explains: “RWD helps us to optimise eligibility criteria for studies, speeding up the recruitment process and ensuring the study results are more broadly relevant. We use the data to estimate the appropriate size and duration of trials, and the data also drive decisions around where to recruit investigators and set up study sites so that we can execute studies more quickly and efficiently. “In theory, comparator data based on RWE has the potential to be more generalisable than the comparator data collected in randomised clinical trials,” he added. “That’s because high-quality RWE reflects the experience of a much larger and more diverse group of patients in the real world of medical practice. But there’s still a lot of work that would need to be done to ensure the validity and acceptance of that approach.” Dr Levy also noted another crucial benefit which the tool can bring to the table: “RWE will also help to solve one of our greatest challenges: the high cost of clinical development. By lowering development costs, RWE will increase the return on R&D investment. More importantly, it allows us to invest in more of the exciting opportunities emerging from human genetics and other research into the molecular roots of disease.” This is of particular importance in the past months, where the issue of cost/benefit ratios and their impact on the industry’s ability to innovate has become even more of a major talking point than usual, culminating with the release of the White House’s white paper Reforming biopharmaceutical pricing at home and abroad in February. The report explains the Trump administration’s position on the issue, asserting that the United States’ unparalleled medical costs are due to the insistence of foreign nations to provide affordable healthcare, meaning that the US is left to foot the remainder of the bill. While this is a different debate entirely, the report does note that this results in pharma firms being unable to make an adequate return on investment, stretching budgets and greatly reducing the feasibility of exploring new research and developing new innovative treatments. If RWD were to open the door to cutting some of these costs and free up more cash to inject into new research, particularly in therapeutic areas such as depression and other mental health areas where innovation has ground to a halt because there is no money to be made in newer treatments, it could prove revolutionary for the industry and the patients it serves. What’s standing in the way? Of course, unlocking the potential of any promising tool like RWD is never an affair bereft of challenges, and the Tufts CSDD study sought to pick the brains of its respondents to discover what they foresee as the biggest obstacles on the road ahead. Perhaps unsurprisingly for a tool at such a relatively embryonic stage, availability of data was voted the biggest challenge. “When you acquire data, it can have a lot of gaps, and you don’t always have the full data set, especially when you’re looking to integrate different types of data sets, as well as compiling them across systems,” Dr Lamberti explained. “The standardisation of data is also a challenge between systems. There are a lot of areas which have to be overcome in order to use multiple sources of data. Linking the data – for example, electronic health record or claims data and unstructured data – trying to have all of these disparate data sources come together is a real challenge.” Also high on the list were issues including lack of stakeholder trust in RWE and RWD on both the regulatory side and the payer’s, cost of integration, and notablyquality and reliability of registry data. When asked, Helfgott singled out the latter as one of the most crucial issues in the effective use of RWD and RWE. “The RWE space will only grow and be credible if the data quality is up to standard,” he remarked. “If you don’t have that, it’s going to be difficult.” “In order for RWE, whether retrospective or prospective, to be reliable, you need to make sure the quality of the data being used is pristine and top-notch. We are studying more than 50 real world evidence registries in the world of cardiovascular, renal and metabolic disease alone. To make sure these 50 are high-quality, reliable and highly credible, we are very cautious about the quality of the data that we’re using. We are adopting exactly the same standards in terms of data quality for RWE as we do for randomised clinical trials. Essentially, our statisticians and our clinical specialists are using the same criteria to certify the data. Very often, the top investigators of RWE are also investigators of clinical trials, so they know exactly the standards of quality that they have to adhere to. “You look at the size of the registry; you look at the institution that is holding the registry; you look at the connection between the primary institution and the centralised institutions,” he continued. “When a registry doesn’t seem totally reliable, or it’s not big enough, or we don’t have the option to verify its quality, we just won’t use it. So data quality is a must, and as a consequence, if you are just as rigorous in your RWE as you are in your randomised clinical trials, you will pick the best institutions to get access to the necessary data.” Dr Levy adds: “The scientific community still needs to establish a consensus on data quality standards, both for data elements and analytic methods. Within industry, most of our drug development processes and systems are still based on randomised clinical trials, and most people are trained in them, not observational research.” Into the future As the Tufts CSDD study shows, the prevalence and utility of RWD and RWE is only going to expand in the coming years, as the industry moves to better embrace the tools and their many uses. The study illuminated the quantifiable ways in which industry players will pursue this expansion over the next three years, but what are the predictions for the longer-term? “I fundamentally believe that the use of RWE will move from confirmation – validating, in real life, what you see in a randomised clinical trial – to more exploration: investigating, either through registries, the impact of medicines, or to help us understand new fields of medicine that have been radically changed by previous standard of care,” Helfgott mused. “I really feel that, in that sense, the randomised clinical trial will become even more important in the overall toolkit. This is why at AstraZeneca we are investing so much in this field, growing the number of registries with which we work year after year, as well as our topics of interest, where we can observe, either from a disease perspective or a medicine perspective, what reality has to tell us.” Dr Levy also had some exciting forecasts to share: “In the long-term, a lot will depend on future regulatory policies. But I’m confident that as trust in this evidence increases, we’ll be able to use these data in more situations where randomised clinical trial data was once the only option” he remarked. “I think it’s only a matter of time before we see the first label extension of a product based purely on RWE. I also think we’ll see more use of hybrid studies, where we run a traditional trial to characterise the safety and efficacy of a new product, and we use RWD to show the safety and efficacy of the therapeutic alternatives.” More Information On **Pharmafile