Nawal Roy is founder and CEO of Holmusk, a company dedicated to building one of the world’s largest real-world evidence platforms for neurosciences. Here, Nawal Roy discusses the potential for real-world evidence to optimise drug life-cycle
Use of real-world evidence (RWE) can modernize the process of drug research, development and delivery.
The pharmaceutical industry spends a tremendous amount of resources to bring treatments from “bench to bedside” with the aim of addressing unmet medical needs and ultimately for financial return on investment. Globally, almost $160B was spent on pharmaceutical R&D in 2016 and is projected to exceed $180B by 20221. This process of drug discovery to commercialization is an arduous journey. By some estimates, bringing a drug successfully to market has direct costs up to $1.4B (USD)2, takes approximately 10 years2, and has roughly 10%3 probability of success in completing clinical development.
Regulatory agencies such as the FDA, delicately balance the goals of expediting effective treatments to patients while meeting the safety concerns of the new medications. The FDA published guidelines in 2016 for the use of real-world evidence (RWE) to modernize the drug development and approval process. The objective was to help accelerate the discovery, development and delivery of medical products to patients who need them faster and more efficiently.
RWE is obtained from real-world data (RWD), generated during clinical practice outside the context of randomized controlled clinical trials. RWD sources include electronic health records (EHR), digital applications, observational studies, prescription records, etc. RWD accounts for the vast majority of patient data: as much as 95% compared to only 5% from the actual clinical trials.
Tangible ways that pharmaceutical companies can use RWE to increase effectiveness in product life-cycle management:
- RWE for new indications – with little or even no additional clinical trials. Typically a new indication for an already approved drug requires both phase 2 and 3 clinical trials to establish efficacy and safety. With the new FDA guidelines, RWE may be used to obtain approval based on evidence collected from physician usage and clinical observations. Similarly, this provision can potentially be used for approval of new formulations of the original drug, novel delivery mechanisms for the drug, or new population of patients (e.g., different ethnicities, different severity levels of the disease).
- RWE for post-approval requirements (pharmacovigilance, phase 4 studies) – for efficient collection of supporting data. Typically, the FDA requires the monitoring of medical drugs after they have been licensed for use, to identify and evaluate previously unreported adverse reactions. It is only after the drug is marketed and hundreds of thousands of patients have been exposed to it that long-term safety is fully understood. In the real world, physicians use newly approved drugs in a wide range of patients and conditions: different ethnicities, various disease comorbidities, range of concomitant medications, or with hepato-renal dysfunction (impacts drug metabolism and excretion). This FDA provision will allow for more efficient gathering of pharmacovigilance data to meet the post-approval requirements and avoid lengthy and expensive post-approval studies.
Diseases that are “messy” and inherently “complex” because of comorbidities and polypharmacy treatments, such as mental health disorders and certain chronic diseases (cardiovascular, metabolic), can especially benefit from analytic techniques to generate RWE.
RWD is “messy” and includes significant unstructured data which requires processing for use in analysis and evidence generation. The challenge of converting RWD to meaningful evidence is more difficult with certain diseases that have inherent “complexities”. For example, patients with mental health disorders or certain chronic diseases (cardiovascular, metabolic) are plagued with comorbidities and treatment regimens often involve multiple drugs (polypharmacy). Furthermore, the progression of these diseases are characterized by a complex network of feedback pathways, and the management of one disease can significantly impact the outcomes of other coexisting diseases.
Generation of RWE will require the application of advanced data analytics for certain diseases. For example, analysis of outcomes of mental disorders to a drug in isolation will be of limited impact, unless the dynamics of the various comorbidities affecting mental disorders are also taken into account.
In these situations, identifying the meaningful signal from RWD will benefit from specialized analytic tools including machine learning techniques that can take into account the myriad comorbidities, polypharmacy and feedback loops. Use of the right analytical methods with RWD is key to extracting the correct insights and generating RWE.
Thus, optimizing the use of RWE, especially with data analytic enhancements, for approval of new indications and post-approval requirements can reduce cost and time burdens for pharmaceutical companies, enhancing their profitability. Moreover, direct and concrete benefits can be experienced by the patients by the accelerated delivery of medical products to those that need them. Given the highly competitive nature of the industry and the huge hurdles of drug R&D (in cost, time, risk), applying the latest tools to generate RWE will bolster competitive advantages, a path encouraged by the FDA.