PathAI presents results at The AASLD Liver Meeting 2021 showing that panel scoring of NASH histological features can reduce bias and increase reproducibility
Research study shows high concordance in scores within panels of three NASH-trained pathologists and between independent panels
Boston, Massachusetts - PathAI, a global leader of AI-powered technology, today announced results showing that scoring of NASH histopathological features in phase 3 clinical trials is highly concordant within a panel of 3 expert hepatopathologists and between two independent panels. These results will be shared in the poster, “Minimizing Variability and Increasing Concordance for NASH Histological Scoring in NASH Clinical Trials” (Abstract #1620), at The AASLD Liver Meeting 2021, November 12-15, 2021.
NASH histology is central to evaluation of clinical trial enrollment criteria and endpoints; however, pathologist assessment is subject to intra- and inter-reader variability. A pilot study based on a subset of data from a Phase 3 clinical trial assessed concordance of NASH histology scoring within a panel of 3 board-certified, expert hepatopathologists (who underwent proficiency training for NASH Clinical Research Network (CRN) scoring) and between two independent panels. Within each panel, scores were highly concordant for all components of the NASH CRN scoring system, and highest for steatosis (Panel A, κ(95% CI): 0.81 (0.69-0.81); Panel B, κ(95% CI): 0.87(0.79-0.87)) and lowest for lobular inflammation (Panel A, κ(95% CI): 0.61(0.23-0.61); Panel B, κ(95% CI): 0.57(0.38-0.57)). Interpanel concordance was similar to that in previously published studies.
In alignment with a recent FDA regulatory proposal recommending scoring of NASH histological trial endpoints by two hepatopathologists and a third for discordant cases, this study shows that panel scoring can support the reproducibility of NASH scoring and reduce bias, as demonstrated in biopsies obtained in a phase 3 clinical trial. PathAI has curated a network of board-certified pathologists that is central to model training and development and the validation of PathAI’s AI-based drug development tools.
PathAI is a leading provider of AI-powered research tools and services for pathology. PathAI’s platform promises substantial improvements to the accuracy of diagnosis and the efficacy of treatment of diseases like cancer, leveraging modern approaches in machine and deep learning. Based in Boston, PathAI works with leading life sciences companies and researchers to advance precision medicine. To learn more, visit pathai.com.