Real-World Evidence in Drug Development and Evaluation

Harry Yang, Binbing Yu, Harry Yang, Binbing Yu

• Publisher: Chapman and Hall/CRC
• Year: 2021
• ISBN:
  9780429676819
• Edition: 1
• Topic: Mathematik
• Subtopic: Wahrscheinlichkeitsrechnung & Statistiken

Real-world evidence (RWE) has been at the forefront of pharmaceutical innovations. It plays an important role in transforming drug development from a process aimed at meeting regulatory expectations to an operating model that leverages data from disparate sources to aid business, regulatory, and healthcare decision making. Despite its many benefits, there is no single book systematically covering the latest development in the field.Written specifically for pharmaceutical practitioners, Real-World Evidence in Drug Development and Evaluation, presents a wide range of RWE applications throughout the lifecycle of drug product development. With contributions from experienced researchers in the pharmaceutical industry, the book discusses at length RWE opportunities, challenges, and solutions.

Features

• Provides the first book and a single source of information on RWE in drug development
• Covers a broad array of topics on outcomes- and value-based RWE assessments
• Demonstrates proper Bayesian application and causal inference for real-world data (RWD)
• Presents real-world use cases to illustrate the use of advanced analytics and statistical methods to generate insights
• Offers a balanced discussion of practical RWE issues at hand and technical solutions suitable for practitioners with limited data science expertise

1

Using Real-World Evidence to Transform Drug Development: Opportunities and Challenges

Harry Yang

1.1 Introduction

In recent years there has been a growing interest in using real-world evidence (RWE) to support drug development, regulatory review, and healthcare decision-making. RWE, gleaned from real-world data (RWD), provides useful insights into disease prevalence, innovative trial design, comparative effectiveness and safety of treatment, and health economic value. Coupled with evidence from randomized controlled trials (RCTs), RWE enables drug developers, regulators, and healthcare providers to make more informed decisions. RWE has been at the forefront of pharmaceutical innovations, disrupting the way evidence is generated in the value chain of drug R&D and commercialization. The use of RWE is further powered by the new governmental policies and laws such as the 21st Century Cures Act in the United States and Conditional Approval (Martinalbo et al. 2016) and Adaptive Pathways in Europe (EMA 2016a,b). Leveraging RWE in regulatory decision is a key priority for many regulatory agencies. In the recently released U.S. Food and Drug Administration (FDA) guidance (FDA 2018a), it is stated that under the right conditions, data derived from real-world sources can be used to support regulatory decisions. When extracted from well-designed studies and appropriate analysis, RWE may constitute valid scientific evidence to support the early approval of a drug product, label change, or expansion. In this chapter we present the unprecedented opportunities and challenges of applying RWD and RWE in drug development and evaluation.

1.2 Traditional Drug Development Paradigm

1.2.1 Drug Development Progress

Drug development is a complex, lengthy, and resource-intensive process. Figure 1.1 presents a diagram of drug development.

FIGURE 1.1
Drug development process. Adopted from FDA website.

The process commences with drug discovery. Scientists utilize many technologies such as synthetic chemistry and genomic sequencing to uncover targets that are causes of diseases. When a lead new molecular entity (NME) is identified, it is advanced to a pre-clinical development stage where the NME is tested both in vitro in cells and in vivo in various animal species to determine its safety and efficacy. This is followed by the clinical phase of product development, which typically follows a well-established paradigm, with the primary aim of generating evidence of drug safety and efficacy in support of marketing approval by regulatory authorities. It consists of three phases, Phase I, II, and III trials, with a focus on clinical pharmacology and early safety, efficacy evaluation in targeted patient populations, and confirmation of drug's safety and efficacy, respectively. Potentially, Phase IV studies, often termed post-approval trials, may be required after marketing approval. It is aimed at gaining better understanding of either potential long-term adverse effects or rare adverse events associated with the product. This post-marketing evaluation again draws insights from clinical studies in controlled settings. Clinical trials are often carried out utilizing a mechanism in which subjects in treatment group(s) or a control group are randomly assigned, and usually referred to as RCTs. Randomization is used to rule out the effects of potential confounding factors and ensure comparable patients across the groups (Barton 2000). Patients in the studies are closely followed to ensure treatment adherence. Additionally, individuals including the medical monitors from sponsors, patients, and investigators are blinded about the patient treatment, and the way data are analyzed is pre-specified. Together, these measures ensure internal validity of the study design and conclusions.

1.2.2 Limitations of Traditional Randomized Controlled Trials

Use of RCTs for demonstrating drug safety and efficacy has been the gold standard for assessing a drug's safety and efficacy. The evidence generation process of an RCT is consistent with regulatory expectations. However, there are several drawbacks concerning the RCT methodology. First, the outcomes from RCTs may lack external validity as RCTs are often conducted under strictly controlled experimental conditions that are different from routine clinical practice. Consequently, RCTs often provide an estimate of the efficacy of the drug rather than the true measure of effectiveness in the real world (Black et al. 1996), resulting in a gap between the efficacy demonstrated in the clinical trials and effectiveness of the drug in real-world use. Factors that contribute to the gap include patient adherence, age, comorbidities, concomitant medications, and so forth. Because of these variations, findings from RCTs may not always translate into the performance of the product in the practical setting. Second, due to rising medical costs, healthcare decision-making by payers has increasingly relied on the balance of cost and benefit of a new treatment. However, despite the demand for demonstration of the value of medical products to justify payment, traditional RCTs offer little information because of the gaps between efficacy-effectiveness. The situation worsens for drug products that are approved based on single-arm studies, surrogate endpoints, or short-term outcomes. Last, because not every patient responds to the same treatment in the same way owing to heterogeneity of the patient population, it is of interest to both the patient and prescribing physician to understand potential effects of the treatment on an individual patient. In the traditional clinical trials, the efficacy and safety of a treatment is demonstrated through comparing the average outcomes of the treatment and control groups. Therefore, it does not provide the patient-specific assessments of efficacy and safety. That evidence generated from the traditional clinical research fails to guide patients, physicians, and health systems for real-world decisions, as noted by many researchers. Hand (2009) stressed that “the aim of a clinician is not really to work out whether drug A is superior to drug B ‘on average,’ but to enable a decision to be made about which drug to prescribe for the next patient who walks through the door, i.e., for the individual.”

1.3 Real-World Data and Real-World Evidence

1.3.1 Real-World Data

RWD are data that are collected from diverse sources, outside the constraints of conventional RCT. They often consist of observational outcomes in a heterogeneous patient population. Because the data are not collected in a well-controlled experimental setting and come from diverse sources, they are likely unstructured, heterogeneous, complex, and inherently variable. RWD can be derived from electronic health records (EHRs), claims and billing activities, product and disease registries, patient-related activities in outpatient or in-home use settings, and health monitoring. They also may be captured through social media and wearable devices thanks to the advances of digital technologies. These latter data provide opportunities to deliver...