BOSTON — OM1 said its real-world evidence platform supported a large regulatory study used in the U.S. Food and Drug Administration’s approval of Hologic’s Aptima HPV Assay for primary cervical cancer screening.
The Boston-based real-world evidence company presented its role in the study at the Medical Device Innovation Consortium Evidence Summit on May 5. Hologic announced the FDA approval in February.
OM1 said the study included more than 650,000 women across multiple U.S. health systems, making it one of the largest real-world evidence studies conducted for an FDA submission.
The study used OM1’s platform to ingest data from electronic health records and laboratory information systems at participating centers. The platform de-identified, tokenized and mapped the data at scale, while using artificial intelligence and machine learning to extract key variables from unstructured clinical notes, cytology reports and pathology records.
The study also participated in the NESTcc, or National Evaluation System for health Technology Coordinating Center, program, which supports the assessment of real-world evidence for medical devices.
“Two characteristics really defined this study,” said Rich Gliklich, Founder of OM1. “One is the scale — 650,000 patients is large by any measure. The second is that a lot of the key variables lived in the unstructured data, the clinical notes and reports.”
Gliklich said OM1’s automation capabilities helped make the study feasible.
“The site Net Promoter Score averaged 87.5 — which is off the charts, and that probably reflects how much manual effort was reduced with automation,” Gliklich said.
OM1 said the study required processing unstructured clinical data at a scale that traditional manual abstraction methods could not support. The company’s platform automated quality control, provided real-time reporting dashboards for site investigators and maintained traceability and provenance of data throughout the process.
The company said all AI and machine learning models used for natural language processing of clinical documents were validated and documented to support regulatory review.
OM1 said the approval represents a significant example of real-world evidence being used in regulatory submissions. The company said its platform shows that automation, when paired with rigorous validation, can reduce the cost and time constraints of large evidence-generation studies without compromising data quality.
