Digital histomorphometry reveals the optimal bioptic sitein pre-transplantation kidney: preliminary results of a prospective study.

INTRODUCTION Kidney transplantation is the best treatment option for patients with end-stage renal disease.1 The success of this procedure is largely based on the donor nephron mass (NM), as the nephron represents the functional unit of the kidney.1 However, the NM shows an extraordinary inter- and intra-organ heterogeneity, with subsequent challenges in its clinical and histologic quantification.2 Furthermore, after transplantation the number of nephrons can be reduced by cold ischemia time, transplant trauma, and the potential nephrotoxicity of immunosuppressive therapy, that may result in kidney failure.1 To date, histopathological evaluation of core biopsies from donor kidney before transplant is the gold standard procedure to predict recipient’s outcome.1 Given the heterogeneous distribution of the nephrons across the kidney, there are not standard protocols to define the NM, either on core biopsies and on surgical samples. Therefore, the role of this parameter as a biomarker in renal transplant patients is unknown. GOAL Here we set out to define whether digital pathology can assist in the clinical workup of kidney transplant. To address this aim, we sought (i) to evaluate the NM by means of digital histomorphometric analysis and (ii) to evaluate the homogeneity of the nephron density across the entire organ. MATERIALS AND METHODS Five kidneys removed from brain-dead donors but found not to be amenable for subsequent implant (female n=2, male n=3, mean age of 62.5, range 21- to 83-year-old) were prospectively collected. For each case, 6 tissue wedges were sampled, encompassing distinct topographic areas of the kidney. Representative 4-μmthick sections were cut from formalin-fixed paraffin-embedded blocks of all samples and stained with hematoxylin and eosin using standard protocols. In order to minimize human-related biases, each stained slide was digitalized and blindly analyzed by three pathologists using a dedicated navigation software.3 Subsequently, a customized counting grid was employed to evaluate the glomerular density of the tissue slide. Using a specific add-on module, both the cortex thickness and glomeruli size were measured. Finally, the total volume of the renal cortex was calculated using bioinformatic Cartesian-based algorithms.4 RESULTS Case 1 Case 2 Case 3 Case 4 Case 5 Average Kidney volume (cm3) 178.1 88.2 66.1 111.7 145.2 117.9 Renal cortex thickness (mm) 6.7 4.4 5.1 4.9 8.0 5.8 Mean glomerulus size (µm) 223.86 174.96 185.30 205.46 174.26 192.77 Glomerular density (%) 8.10 8.30 7.40 8.74 9.19 8.35 Nephron mass 497,156 389,450 301,369 353,734 855,279 479,398 Table 1. Histomorphometric parameters of the cases included in the study. Superior segment Middle segment Inferior segment Renal cortex thickness (mm) 5.40 5.24 5.12 Mean glomerulus size (µm) 197.89 195.46 191.63 Glomerular density (%) 8.41 9.06 7.00 Table 2. Histomorphometric parameters across distinct anatomical sites of the study group. The overall glomerular density was significantly different in the three macro-regions of the kidney, while both the cortex thickness and the mean glomerulus size did not show any significant variation. CONCLUSIONS The present study is the first to investigate the histomorphometry of pre-transplantation kidneys by means of digital image analysis. Our findings suggest that a single preimplantation biopsy could be representative of the overall donor NM if performed in the superior kidney segment. The integration of traditional pathology with cutting-edge digital technologies and bioinformatic tools, could be extremely beneficial in the implementation of reproducible diagnostic schemes for renal transplantation. REFERENCES 1. Karam G, Kälble T, Alcaraz A et al. Guidelines on renal transplantation. European Association of Urology, 2014. 2. McMahon AP. Development of the mammalian kidney. Curr Top Dev Biol 2016;117;31-64. 3. Fusco N, Guerini-Rocco E, Del Conte C et al. Her2 in gastric cancer: A digital image analysis in pre-neoplastic, primary and metastatic lesions. Mod Pathol 2013;26;816-824. 4. Calhoun D, LeVeque R. A cartesian grid finite-volume method for the advection-diffusion equation in irregular geometries. Journal of Computational Physics, 2000;143-180.