• Liver histology is the reference standard for predicting therapeutic benefit in clinical trials with patients with nonalcoholic steatohepatitis (NASH)
• Histological assessments are limited by sampling variability and subjectivity of interpretation, even among individual expert pathologists, leading to inadequate intraand inter-reader concordance
• The United States Food and Drug Administration (FDA) recommends liver histology assessments for phase 3 clinical trials
• The NASH Clinical Research Network (NASH CRN) approach whereby multiple expert pathologists meet in-person to review samples and reach consensus is the current standard for histology assessments
• Convening a histopathology committee achieves high concordance, but can be logistically challenging and subject to bias (eg, impact of dominant voices within the committee)
•A recent FDA-issued regulatory perspective proposes using at least 2 pathologists trained in evaluating liver biopsy, with involvement of a 3rd pathologist for discordant readings, as a potential approach to ensure that histological endpoints are reliable and consistent Sanyal et al.
• Histological assessments are limited by sampling variability and subjectivity of interpretation, even among individual expert pathologists, leading to inadequate intraand inter-reader concordance
• The United States Food and Drug Administration (FDA) recommends liver histology assessments for phase 3 clinical trials
• The NASH Clinical Research Network (NASH CRN) approach whereby multiple expert pathologists meet in-person to review samples and reach consensus is the current standard for histology assessments
• Convening a histopathology committee achieves high concordance, but can be logistically challenging and subject to bias (eg, impact of dominant voices within the committee)
•A recent FDA-issued regulatory perspective proposes using at least 2 pathologists trained in evaluating liver biopsy, with involvement of a 3rd pathologist for discordant readings, as a potential approach to ensure that histological endpoints are reliable and consistent Sanyal et al.
