Background

Randomized Controlled Trials (RCTs) have long served as the standard benchmark in drug development. This approach, which involves randomly assigning patients in a controlled environment to assess the efficacy and safety of a drug, is generally considered reliable. However, patient recruitment can be costly, and in areas with limited patient populations, such as rare diseases, the design itself can be challenging. A disconnect from real-world clinical practice is also a concern. Because studies are conducted on a limited number of carefully selected patients, it is difficult to fully reflect the actual responses of diverse patient populations with varying ages and comorbidities in everyday clinical settings.

To overcome these limitations, there is a growing interest in Real-World Data (RWD), which is generated from routine clinical practice. Previously, RWD was primarily used for post-market surveillance (PMS) to monitor adverse events after a new drug is launched. However, with the accumulation of large-scale Electronic Health Records (EHRs) and health insurance claim data, and the development of analytical techniques, the landscape is changing. Recently, Real-World Evidence (RWE), which is derived from RWD, has emerged as a key component in drug approval and clinical trial design.

Key Findings

A recent report published in Nature Medicine suggests that RWE has evolved beyond a mere reference tool to a stage where it can simulate existing RCTs or even conduct real-time clinical trials. A notable example is the case of 'Iwilfin' (active ingredient: eflornithine), a treatment for high-risk pediatric neuroblastoma approved by the U.S. Food and Drug Administration (FDA). Instead of a traditional double-blind RCT, the FDA used an Externally Controlled Trial (ECT) comparing a treatment group of 90 patients with a historical control group of 270 patients, using propensity score matching to verify the drug's efficacy. The analysis showed that the Iwilfin group had a hazard ratio (HR) of 0.48 for event-free survival (EFS) compared to the control group, indicating that the risk of recurrence was reduced by more than half. This is considered a landmark case in which external control group data was used as the primary basis for obtaining new drug approval in the field of oncology.

The rapid expansion of RWE is due in part to the growth of large EHR data platforms such as Epic Cosmos. Target Trial Emulation (TTE), a technique that analyzes data from hundreds of millions of patients, helps to infer causal relationships without randomization. Researchers are also introducing medical Artificial Intelligence (AI)-based models that convert prescription history and clinical outcomes into a sequential series of events and use this data for learning. This technique is being used to simulate patient outcomes and predict drug response rates, thereby enhancing the clinical trial design phase.

Significance and Prospects

The use of RWE is expected to facilitate the development of new drugs in areas where patient recruitment is difficult, such as rare diseases and pediatric cancers. It is also expected to reduce clinical costs and time, thereby improving patient access to treatment. Regulatory agencies in various countries are also releasing relevant guidelines to increase its institutional acceptance.

However, there is also cautious skepticism that RWE cannot completely replace existing clinical trials. The incompleteness of EHR data, which is not standardized in its collection methods, and hidden confounding variables may distort the results. In fact, many RWE-based analysis plans are being questioned for their reliability due to unclear design requirements and analytical errors. Experts point out that it is difficult for RWD analysis to completely replace the bias control of randomization, and that it is desirable to establish a dual model structure in which RCTs and RWE complement each other. Establishing a causal roadmap to ensure transparency from the data collection stage is also a challenge.

Why It Matters

The institutional adoption of RWE is likely to accelerate the realization of personalized precision medicine in clinical practice. In particular, its application is noteworthy in multi-ethnic countries and in super-aged societies with a large number of patients with complex chronic diseases. It is possible to use an RWE-based monitoring system to analyze in real time the long-term survival trends of patients aged 70 or older with diabetes complications, in whom adverse events or efficacy were not fully identified in the clinical trial phase, when a specific immune checkpoint inhibitor is prescribed. Pharmaceutical companies can use this data to quickly expand the approved indications, and clinicians can establish precise prescription guidelines tailored to individual patients. From the perspective of national health insurance finances, there are also practical benefits in reducing the waste of healthcare resources by rationally adjusting the coverage of drugs based on RWE analysis of actual treatment effectiveness.