Kidney Week 2025 Annual Meeting-Histological Lessons from Modern Biopsy Cohort Studies

Video Transcription

Video Summary

The session opens with housekeeping announcements about disclosures, recording availability in the ASN eLearning Center, evaluations, and room etiquette. The moderators then frame the theme: large kidney biopsy cohorts (e.g., KPMP, NEPTUNE, CureGN) are combining re-reviewed histology with digital pathology and molecular methods (spatial/proteomics, single-cell RNA-seq) to better understand kidney lesions and clinical outcomes.<br /><br />Andrew Rule (Mayo Clinic) presents quantitative morphometrics for prognostication rather than diagnosis. Using whole-slide annotation (glomeruli types, inflammation, IFTA, arteries), stereology, and CT-derived cortical/medullary measures, his group estimates nephron number and nephron size. Key points include age-related nephron loss that may occur with minimal visible sclerosis/IFTA, the value of age-adjusted “normal” thresholds, and strong outcome prediction from nephron number/size and IFTA foci density. In living donors, these measures predict post-donation GFR, albuminuria, and hypertension. In transplant recipients, imaging—especially medullary pyramid number—predicts graft failure, and “ischemic” (deflated) glomeruli and IFTA foci density outperform conventional chronicity scores. He describes AI models that automate these metrics and add features like peritubular capillaries and inflammation foci density.<br /><br />Laura Barisoni (Duke) explains computational pathology and “pathomics” to extract high-dimensional, subvisual features from segmented functional tissue units (glomeruli, tubules, interstitium, capillaries). She highlights collaborative intelligence (human + AI), discovery of new quantitative spaces (e.g., interstitial fractional space), and examples where pathomic signatures predict progression or treatment response (including IgA nephropathy), with biologic validation via gene expression.<br /><br />Niko Kozakowski (Vienna) reviews the evolving Banff concept of microvascular inflammation (MVI) in antibody-mediated rejection, including C4d-negative phenotypes, transcriptomic correlates, and emerging work on infiltrating cell types (NK, T, B/plasma cells), multiplex methods, EM insights, and AI-assisted capillaritis assessment.<br /><br />Kerry Phillips (Indiana) discusses 2D/3D visualization of glomerular neovascularization (“polar vasculosis,” “hairy glomeruli”)—common in diabetes but also other diseases—its potential adaptive role, links to angiogenic signals, and possible modulation by SGLT2 inhibitors, emphasizing how AI can also quantify inter-pathologist scoring variation.

Asset Subtitle

Moderator(s): Shana Coley, Vanessa Moreno

Presentation(s):

Quantitative Morphometrics and Nomograms for Predicting Kidney Transplant Outcomes - Andrew Rule
Applying Advanced Computational Analysis from NEPTUNE, CureGN, and KPMP Biopsies to Clinical Practice - Laura Barisoni
Patterns of Microvascular Inflammation in Kidney Transplantation and Emerging Implications - Nicolas Kozakowski
Novel Vascular Lesion in Diabetes Glomerulopathy - Carrie Phillips

Meta Tag

Date 11/8/2025

Pathway 1 Pathology

Pathway 2 Kidney Biology and Physiology

Session ID 507884

Keywords

kidney biopsy cohorts

KPMP

NEPTUNE

CureGN

digital pathology

spatial proteomics

single-cell RNA sequencing

quantitative morphometrics

nephron number estimation

IFTA foci density

whole-slide imaging annotation

AI-assisted pathology

pathomics

microvascular inflammation Banff

glomerular neovascularization polar vasculosis