Clinical validation of an AI-based pathology tool for scoring of metabolic dysfunction-associated steatohepatitis

Hanna Pulaski; Stephen A Harrison; Shraddha S Mehta; Arun J Sanyal; Marlena C Vitali; Laryssa C Manigat; Hypatia Hou; Susan P Madasu Christudoss; Sara M Hoffman; Adam Stanford-Moore; Robert Egger; Jonathan Glickman; Murray Resnick; Neel Patel; Cristin E Taylor; Robert P Myers; Chuhan Chung; Scott D Patterson; Anne-Sophie Sejling; Anne Minnich; Vipul Baxi; G Mani Subramaniam; Quentin M Anstee; Rohit Loomba; Vlad Ratziu; Michael C Montalto; Nick P Anderson; Andrew H Beck; Katy E Wack

doi:10.1038/s41591-024-03301-2

Clinical validation of an AI-based pathology tool for scoring of metabolic dysfunction-associated steatohepatitis

Nat Med. 2024 Nov 4. doi: 10.1038/s41591-024-03301-2. Online ahead of print.

Authors

Hanna Pulaski^#¹, Stephen A Harrison^#², Shraddha S Mehta^{1

3}, Arun J Sanyal⁴, Marlena C Vitali^{1

5}, Laryssa C Manigat¹, Hypatia Hou¹, Susan P Madasu Christudoss^{1

6}, Sara M Hoffman^{1

7}, Adam Stanford-Moore¹, Robert Egger¹, Jonathan Glickman^{1

8}, Murray Resnick^{1

9}, Neel Patel¹, Cristin E Taylor¹, Robert P Myers¹⁰, Chuhan Chung¹¹, Scott D Patterson¹², Anne-Sophie Sejling¹³, Anne Minnich¹⁴, Vipul Baxi¹⁴, G Mani Subramaniam¹⁰, Quentin M Anstee¹⁵, Rohit Loomba¹⁶, Vlad Ratziu¹⁷, Michael C Montalto^{1

18}, Nick P Anderson¹, Andrew H Beck¹, Katy E Wack¹⁹

Affiliations

¹ PathAI, Inc., Boston, MA, USA.
² Pinnacle Clinical Research, San Antonio, TX, USA.
³ AbbVie, Inc., Irvine, CA, USA.
⁴ Virginia Commonwealth University, Richmond, VA, USA.
⁵ Whoop Inc., Boston, MA, USA.
⁶ Invicro, Needham, MA, USA.
⁷ Harvard Medical School, Boston, MA, USA.
⁸ Department of Pathology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA.
⁹ Department of Pathology and Laboratory Medicine, Rhode Island Hospital, Brown University, Providence, RI, USA.
¹⁰ OrsoBio, Inc., Palo Alto, CA, USA.
¹¹ Inipharm, Inc., Bellevue, WA, USA.
¹² Gilead Sciences, Inc., Foster City, CA, USA.
¹³ Novo Nordisk, Bagsvaerd, Denmark.
¹⁴ Bristol Myers Squibb, Princeton, NJ, USA.
¹⁵ Translational and Clinical Research Institute, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK.
¹⁶ UCSD School of Medicine, San Diego, CA, USA.
¹⁷ Sorbonne Université, ICAN Institute for Cardiometabolism and Nutrition, Assisstance Publique Hôpitaux de Paris, INSERM UMRS, Paris, France.
¹⁸ Amgen, Thousand Oaks, CA, USA.
¹⁹ PathAI, Inc., Boston, MA, USA. [email protected].

^# Contributed equally.

PMID: 39496972
DOI: 10.1038/s41591-024-03301-2

Abstract

Metabolic dysfunction-associated steatohepatitis (MASH) is a major cause of liver-related morbidity and mortality, yet treatment options are limited. Manual scoring of liver biopsies, currently the gold standard for clinical trial enrollment and endpoint assessment, suffers from high reader variability. This study represents the most comprehensive multisite analytical and clinical validation of an artificial intelligence (AI)-based pathology system, AI-based measurement of metabolic dysfunction-associated steatohepatitis (AIM-MASH), to assist pathologists in MASH trial histology scoring. AIM-MASH demonstrated high repeatability and reproducibility compared to manual scoring. AIM-MASH-assisted reads by expert MASH pathologists were superior to unassisted reads in accurately assessing inflammation, ballooning, MAS ≥ 4 with ≥1 in each score category and MASH resolution, while maintaining non-inferiority in steatosis and fibrosis assessment. These findings suggest that AIM-MASH could mitigate reader variability, providing a more reliable assessment of therapeutics in MASH clinical trials.