Algorithm for Identification of Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) and Metabolic Dysfunction-associated Steatohepatitis (MASH) detection

This technology is a tool for rapidly discovering patients at risk for Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) and Metabolic Dysfunction-associated Steatohepatitis (MASH) at all stages including cirrhosis using existing information in an Electronic Health Record (EHR) database.

Unmet Need: Screening for undiagnosed MASLD and MASH within Health Systems

MASLD remains underrecognized within health care databases and among health care providers. Higher risk groups, such as those with diabetes or obesity, would benefit from earlier detection, stratification, and an opportunity to intervene by implementing lifestyle changes and specific treatments. Although some noninvasive tests have been developed to more accurately stratify MASLD, few methods leverage already-collected patient data to detect existing MASLD upstream of endstage disease outcomes.

The Technology: Rapid, scalable algorithm to detect MASLD and MASH using EHR data

This algorithm leverages multiple data elements stored in existing EHR databases to identify potential MASLD/MASH patients at all stages of disease. Data elements such as radiology reports, diagnoses of existing prediabetes, diabetes or obesity, abnormal enzymes levels, demographic history, medication history, and past/present disease diagnoses can be used to identify patients at-risk for MASLD or its more inflammatory and progressive phenotype, MASH. Furthermore, most advantageous is the utilization and application of noninvasive calculators to further detect patients at risk for advanced fibrosis, an area of immediate therapeutic interest given its association with clinical outcomes such as cirrhosis, liver cancer, decompensation, need for liver transplantation or liver related death.

Applications:

• Clinically validated foundation for machine learning and AI approaches
• Pre-diagnostic tool for additional procedural testing (such as imaging, biopsy)
• Referral tool used by general practitioners to identify patients who should seek a specialist’s opinion
• Screening for metabolic risk suitable for intervention (weight loss medications, diabetes screening) not captured through other means
• Screening method to identify patients not-yet-diagnosed with Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) and Metabolic Dysfunction-associated Steatohepatitis (MASH)
• Screening tool for clinical trials identification of potential subjects
• Research tool for studying the link between certain risk factors and the development of MASLD or MASH
• Research tool to study the link between family history of an illness and the presentation of an illness in subsequent generations
• Identification of rare phenotypes/misdiagnoses for genetic testing

Advantages:

• Developed and validated in an ethnically diverse population
• Fits into existing standard of care
• Augments clinical care, not a tool to replace clinical decision making
• Rapid and scalable platform to reach more patients
• Earlier disease detection
• Cost-effective
• Considers a diverse array of patient characteristics

Lead Inventor:

Anna Basile, Ph.D.

Patent Information:

Patent Pending (US 20220181028)

Related Publications:

• Basile A, Verma A, Tang LA, Serpa M, Scanga A, Farrel A, Destin B, Carr RM, Anyanwu-Ofili A, Rajagopal G, Krikhely A, Bessler M, Reilly MP, Ritchie MD, Denny J, Tatonetti NP, Wattacheril J. “Rapid Identification and Phenotyping of Nonalcoholic Fatty Liver Disease Patients Using an Automated Algorithmic Approach in Diverse, Urban Healthcare Systems” medRxiv. 2023 April.

Tech Ventures Reference:

• IR CU19281

• Licensing Contact: Kristin Neuman