Histopathological Analysis of Renal Biopsies With Dynamic Full-field Optical Coherence Tomography, a Comparison to Conventional Histopathological Findings in Kidney Transplant Recipients

Banff classification stands as gold standard and international consensus for the identification, diagnosis and categorization of renal allograft pathology. In addition to multiple revision dedicated to the clarification and refinement of these diagnostic categories, a special attention is now dedicated to the development of automated protocols serving the use of artificial intelligence in digital pathology. To be integrated in the actual early post-transplant monitoring procedure, such original approach is considered to match various growing expectations of clinicians and pathologists regarding the future of transplant nephropathology : decrease inter-observer variability, reduce human errors and limit time-consuming analysis of kidney biopsy. Among these, an accelerated reading and access to Banff diagnostic criteria could help initiating both appropriate and immediate treatment to improve graft survival in kidney transplant recipients. Yet conventional histopathology still requires the preparation of a paraffin block, sections as well as subsequent colorations that altogether delay the final pathological diagnosis and urge the need for additional diagnostic modalities. Designed to overcome this critical limitation, the HARBOR study intends to test the performance of direct histopathological examination of fresh kidney biopsy with full-field optical coherence tomography for the identification of Banff elemental lesions and diagnostic categories.