How AI is Shaping the Creation of ‘Regulatory-Grade’ Real-World Data in Ophthalmology

Regulatory bodies, such as the Food and Drug Administration (FDA) and the European Medicines Agency (EMA), have placed growing emphasis on the use of real-world data (RWD) to drive clinical research and regulatory decisions. This shift highlights the potential of RWD to accelerate drug development, enhance patient outcomes, and complement traditional clinical trials. The key challenge, however, is transforming RWD into ‘regulatory-grade’ evidence that meets the rigorous standards for accuracy, completeness, and reliability.

In ophthalmology, this transformation is especially significant, as the field continues to leverage RWD and artificial intelligence (AI) to address complex conditions, such as cataracts and neovascular age-related macular degeneration (nAMD). These advancements offer a pathway to enhanced regulatory decision-making and improved patient care.

The Role of AI in Elevating RWD to Regulatory-Grade Evidence

AI can streamline the harmonization and curation of RWD, turning raw information into actionable regulatory-grade real-world evidence

Unlike controlled clinical trial data, RWD originates from sources, such as electronic health records (EHRs), medical claims, and imaging. While these RWD sources reflect real-world clinical experiences, their inherent variability and unstructured nature pose challenges for regulatory use. This is where AI benefits as a transformative tool.

Approximately 80% of clinical data is unstructured, yet these untapped insights are critical for understanding patient outcomes and long-term care trends. AI can streamline the harmonization and curation of RWD, turning raw information into actionable regulatory-grade real-world evidence (RWE). Key AI applications include:

• Data Standardization: Ensuring consistency across data sources by aligning data formats and terminologies, which reduces variability and enhances comparability.
• Predictive Analytics: Identifying trends and forecasting outcomes, such as which patients may qualify for clinical trials.
• Bias Mitigation: Correcting for biases that might otherwise compromise the validity of RWE.
• Transparency and Replicability: Developing models with clear training processes and performance benchmarks to meet regulatory standards.

Verana Health’s VeraQ^® population health data engine transforms raw EHR data into high-quality, research-ready datasets, or Qdata^®, enabling robust analyses that meet the scrutiny of regulatory standards

By integrating AI-driven methodologies, Verana Health has demonstrated how curating data from the American Academy of Ophthalmology IRIS^® Registry (Intelligent Research in Sight) can produce regulatory-grade evidence. Verana Health is the exclusive data curation and analytics partner of the IRIS Registry, which is one of the largest specialty clinical data registries in all of medicine, providing a rich source of RWD for ophthalmology. 

Verana Health’s VeraQ^® population health data engine transforms raw EHR data into high-quality, research-ready datasets, or Qdata^®, enabling robust analyses that meet the scrutiny of regulatory standards.

Regulatory Frameworks Guiding RWD in Ophthalmology

Recent regulatory guidance underscores the increasing importance of RWD. For instance, the FDA’s July 2024 guidance highlights how EHR and claims data can accelerate drug approvals and improve postmarket surveillance. Similarly, the EMA’s DARWIN EU initiative aims to standardize and harmonize data collection across Europe, ensuring that RWD is representative of diverse patient populations.

These frameworks emphasize:

• Data Quality and Completeness: Ensuring the data is accurate, timely, and comprehensive.
• Study Design: Mitigating biases and confounders to support causal inferences.
• Transparency: Maintaining clarity in data collection and analysis to enable replicability.

To view case study examples of how Verana Health’s regulatory-grade RWD is being leveraged to address key challenges in ophthalmology, including for conditions such as cataracts and nAMD, read the white paper.