Improving the evaluation of high-risk deceased donor kidneys for transplantation

About This Grant

PROJECT SUMMARY/ABSTRACT Kidney allografts are a life-saving but limited resource. Nearly ~5000 deaths each year can be attributed to the disproportionate number of patients who need an allograft compared to the number of available kidneys. The majority of available kidneys are from deceased donors. Astonishingly, ~2500 procured kidneys are discarded, often labeled as “poor quality” by a single biopsy or by a high kidney donor profile index (KDPI) score. However, measurements of kidney health from a biopsy are limited as they may not represent the variability in pathology within the kidney, and the KDPI lacks individual specificity. This proposal focuses on developing and validating imaging biomarkers to improve the accuracy and precision of the assessment of the discarded kidneys from deceased donors in order to address the critical need to increase the number of available allografts. Using discarded human kidneys, we will determine which imaging features derived by MRI can accurately detect whole kidney pathology. Then we will model those high-risk kidneys using a pig model to determine which features accurately predict outcomes. In Aim 1, we will use site-directed tissue biopsy to validate the individual image features to histopathologic scores for interstitial fibrosis, glomerulosclerosis, vascular sclerosis, and tubular atrophy. Then we will determine whether three-dimensional, multiparametric magnetic resonance imaging features in the donor kidneys can sensitively detect whole kidney pathology. We will apply statistical classification and machine learning to identify predictive imaging features. We will perform this analysis in discarded human kidneys from donors with a KDPI>85 or age>60 years. This population was chosen as it represents the largest population of discarded kidneys. We will also compare these to kidneys of lower KDPI. In Aim 2, we will model high-risk donor kidneys by inducing bilateral ischemia reperfusion injury in a pig. Once the transition to CKD is complete, the kidneys will mirror the pathology observed in human kidneys with a KDPI>85. These high-risk kidneys will undergo ex vivo imaging followed by an autotransplant of a single kidney. Glomerular filtration rate (GFR) will be measured weekly until euthanasia eight weeks after transplant. We will determine which individual or combination of imaging and histopathologic features predict kidney function after transplant. We will generate foundational evidence for the use of imaging features to overcome sampling bias and predict outcomes in high-risk kidney with the primary goal of increasing the number of kidneys for transplantation. This work directly addresses the urgent, recognized need for novel strategies to increase pool of available allografts. The long-term impact of this work will be to reduce inappropriate discards and provide biomarkers to evaluate the suitability of kidneys to expand the donor pool.