Real-World Evidence- Supporting Neurology Drug Approvals

The development of neurological drugs has been known to be very tedious, costly, and difficult. Medical cases such as those of Alzheimer's disease, epilepsy, multiple sclerosis, and Parkinson's disease have been known to be complicated and involve heterogeneous patient populations and thus cannot easily be covered through traditional RCTs.This is where real-world evidence has become a game-changer for pharmaceutical companies, regulators, and researchers alike.

If you have been thinking, “What is real-world evidence?” do not worry; you are not the only one. Over the last decade, real-world evidence has evolved from being a regulatory buzzword to becoming an essential element of drug development strategy, particularly in neurology. In this article, we will define what real-world evidence really is and explore its difference from clinical data, the importance of the FDA real-world evidence framework, and how AI platforms help sponsors turn disorganised real-world evidence into submission-grade data.

What Is Real-World Evidence?

Real-world evidence (RWE) refers to clinical evidence about the usage, potential benefits, and potential risks of a medical product that is derived from the analysis of real-world data (RWD). Contrary to the data gathered in the highly controlled randomised clinical trial environment, real-world data can be sourced from electronic health records, insurance claims databases, patient registries, wearables, pharmacy dispensing data, and even patient-reported outcomes through mobile applications.

Now you know that real-world evidence is simply the evidence collected while a drug or therapy is used in routine practice instead of being used within the tightly regulated confines of a trial. The difference becomes very significant in cases of neurology, where patients have other diseases and are taking more than one medication, and their disease trajectory often spans years or decades, which many traditional RCTs may not fully capture.

Whenever someone asks the same question about RWE and RWD, it means that they are trying to understand the difference between the two and want to know how the RWD becomes the RWE. Real-world data is simply the unprocessed information that could be claims data, EHR data or registry data points. Real-world evidence, on the other hand, is the interpretation derived from the data that answers a clinical or regulatory question.

Why Real-World Evidence Matters More in Neurology

There are various reasons why the neurological disease population is well-suited for real-world evidence:

• Small and heterogeneous populations: Neurological diseases that are rare usually have small enough populations to support a traditional RCT.
• Disease duration: Diseases such as Alzheimer's disease and Parkinson's disease take many years, thus a short timeframe is inadequate.
• Comorbidity burden: Many neurological patients have additional medical problems which make them ineligible for the studies according to criteria.
• Subjective and highly variable outcomes: Measures such as cognitive decline and tremor severity or quality of life cannot be standardised, and thus the long-term experience is important.

Therefore, real-world evidence becomes necessary for the field of neurology.

How Real-World Evidence Differs From RCT Clinical Trials

To fully appreciate the value of real-world evidence, it helps to compare it against RWE clinical trial designs and traditional RCTs side by side.

Traditional RCTs

Randomised controlled trials continue to be the benchmark for determining causation. In RCTs, patients are randomised into the treatment or control groups, reducing bias and confounding factors. On the other hand, RCTs are:

• A costly and lengthy process to conduct
• Restrictive to only a few patient populations
• Too brief in time to evaluate safety and effectiveness
• Unrepresentative of the real-world patients

RWE Clinical Trials and Pragmatic Studies

RWE clinical trials, also known as pragmatic trials, combine components of controlled experiments with real-world scenarios. Such trials involve more extensive and diverse participant groups and evaluate results under the conditions of actual clinical practice, not an artificial experimental setting. Such a combination makes it possible to conduct RWE studies while retaining some level of scientific strictness.

Pragmatic RWE clinical trials are especially helpful for neurology since such trials make it possible for drug sponsors to see how their product works in different patient subgroups, like elderly patients, those with kidney impairment, or those receiving complicated polypharmacy treatment.

The Role of a Real-World Evidence Platform

The process of collecting useful evidence is not as straightforward as extracting information from the EHR system and considering it finished. The reality of real-world data collection is that this data is fragmented, inconsistent in its form and located in different healthcare systems. It is the moment when the use of the real-world evidence platform comes into play.

A reliable real-world evidence platform gathers information from various data sources, such as claims data, EHR, registry, genomics data, and patient outcomes, in order to create one structured dataset. Otherwise, sponsors risk making decisions based on biased or unvalidated datasets.

The key components of an advanced real-world evidence solution usually include:

• Tools for the ingestion and harmonisation of data by converting the different types of data into common data models
• Natural language processing (NLP) systems used to derive structured information out of unstructured clinician notes
• Modelling for statistical and causal inference that accounts for the confounders in the observed data
• Regulatory-grade documentation that can be used in regulatory submissions
• Patient tracking over the course of time to analyse disease progression and the effect of treatments

High-quality RWE platforms can help improve data quality and regulatory readiness, supporting stronger submissions

FDA Real-World Evidence Framework: What Sponsors Need to Know

The FDA real-world evidence framework has evolved significantly since the 21st Century Cures Act of 2016 directed the agency to evaluate how RWE could support regulatory decision-making. Since then, the FDA real-world evidence guidance documents have clarified how sponsors can use observational data to support both new drug approvals and label expansions.

Key components of FDA real-world evidence include:

• Data standards that are fit-for-purpose: The key principle of the FDA real-world evidence guidance is that the data sources need to be valid, credible, and appropriate for answering the particular research question at hand.
• Label expansion: There are several examples of neurological treatments where real-world evidence was used to support expanded indications. In some cases, registry data and other real-world evidence have supported regulatory decisions, including label expansions when combined with appropriate clinical evidence.
• Safety surveillance: Another application of the FDA real-world evidence framework is its use in Phase IV commitments, where real-world evidence can be used to monitor long-term safety concerns after approval.
• External control arms: In the case of rare neurological diseases when randomisation is not feasible due to ethical and practical reasons, the FDA real-world evidence pathway enables sponsors to use real-world data as an external control arm rather than placebo.

Knowing what to expect from the FDA in terms of real-world evidence early in the development program allows sponsors to develop their data collection strategy accordingly.

How NeuroDiscovery AI Powers Real-World Evidence Generation

Starts With Longitudinal Real-World Data

For RWE to be considered reliable, the data has to depict real-world clinical practice. At NeuroDiscovery AI, researchers have access to a substantial database of real-world neurology data which consists of records for millions of patients. This data has many clinical dimensions like structured diagnoses, medication history, laboratory tests, imaging results, and clinical notes. It is also noted that this data is longitudinal, enabling researchers to examine patient journeys rather than isolated clinical encounters.

Uses NeuroLLM™ for Clinical Context Extraction

One of the difficulties with the RWE generation process is that key clinical information may not be available in structured formats. NeuroDiscovery AI's NeuroLLM™ is a neurology-specific multimodal large language model that processes neurological data. NeuroLLM™ pulls together all information from the clinical records as well as from other data sources and assists in extracting the information from unstructured and multimodal data for research purposes. According to NeuroDiscovery AI, NeuroLLM™ includes EHR, billing and claims, imaging, and other clinical data.

Connects Cohort Intelligence With NeuroEvidence

NeuroEvidence is placed under NeuroDiscovery AI’s description of itself as a platform for clinical research intelligence, which provides clinically-traceable answers based on the scientific literature. NeuroEvidence is a resource described by NeuroDiscovery AI as a way to sift through massive amounts of scientific literature in order to answer clinical and research questions using evidence-based approaches.

Extends Evidence Generation to Health Technology Assessment With NeuroHTA

The concept of RWE can have applications beyond just clinical development. Evidence might be necessary in order to better understand the value and impact of therapies. NeuroDiscovery’s AI technology has identified NeuroHTA as one of its infrastructure tools for real-world evidence in the field of neurology. The platform aims to serve as a platform for evidence generation which extends to health technology assessment.

Real-World Applications: Real-world Evidence in Neurology Approvals

Some concrete real-world case studies help demonstrate how real-world evidence has already changed drug approvals in neurology:

• Label extensions in epilepsy: Registry-derived real-world evidence has contributed to label expansion decisions when evaluated alongside clinical trial evidence.
• Safety monitoring in multiple sclerosis: real-world evidence claims-based analysis has been employed for assessing the long-term risks of cardiovascular events and infections related to disease-modifying drugs.
• External control arms in Alzheimer's disease: External control arms have been explored in Alzheimer's disease research and are increasingly used in rare neurodegenerative diseases when randomised controls are not feasible.
• Functional outcomes in Parkinson's disease: real-world evidence generated by wearable sensors on gait severity and tremor allows getting additional objective assessments of disease progression besides those that are clinician-based.

All these cases show the importance of real-world evidence for neurology drug approval that has become more than just complementary to clinical trials.

Challenges in Generating Reliable Real-world Evidence

Although promising, the process of obtaining valid real-world evidence faces certain difficulties, such as:

• Quality of data and completeness: Omissions of certain data points and lack of consistency in recording data may lead to biases.
• Confounding variables: Real-world evidence research, unlike randomised clinical trials, needs to account for the presence of confounders.
• Interoperability: Inconsistencies between coding standards for different EHR databases and registries require their harmonisation.
• Regulatory acceptance: Not all real-world evidence qualifies as "fit for purpose," so it is important to ensure from the very beginning that the design of the study follows FDA guidelines on real-world evidence.

A real-world evidence platform designed properly, together with solid methodology, is vital for overcoming these barriers.

The Future of Real-World Evidence in Neurology

The future of neuroscience is not far away, and in the coming years, it will continue to evolve with the advancement of wearable devices, digital biomarkers, and analytics based on AI algorithms. Continuous monitoring devices such as smartwatches that record tremor frequency and apps that help to measure speech patterns for early signs of cognitive decline are providing entirely new sources of real-world evidence.

NeuroDiscovery AI is among the companies that are ahead in the use of technology and the advancement of platforms for the analysis of data to help transform real-world data into ready-to-submit real-world evidence. With the advancement of the FDA real-world evidence framework, sponsors can gain an edge over their competitors through investing in the real-world evidence framework.