Autosomal Dominant Polycystic Kidney Disease (ADPKD) is a genetic disorder affecting approximately 1 in 1000 Americans, and collectively represents a leading cause of renal failure. While over 90% of ADPKD cases can be diagnosed genetically through defects in ADPKD1 or ADPKD2 genes, the clinical course of an individual is highly variable. Investigations are ongoing to understand the biochemical basis for this variability, but other studies take a more pragmatic approach. They attempt to find biomarkers or features which can predict ADPKD progression. Perhaps the most notable recent such study is the Consortium for Radiologic Studies of Polycystic kidney disease (CRISP). CRISP was a longitudinal multi-institutional imaging study where the predictive value of several easily extracted image features were evaluated. The image features studied by CRISP are basic by design, permitting efficient extraction from large numbers of participants. Image features investigated by CRISP include total kidney volume (TKV), cyst volume, and % cyst volume. CRISP found TKV correlated with kidney function decline, measured by glomerular filtration rate (GFR) change over the course of a decade. Hundreds of standardized ADPKD imaging studies underwent TKV evaluations through the CRISP study, which includes Mayo Clinic. TKV is obtained via stereology, a technique which consists of a coarse grid overlaid on image slices;only points falling within a region of interest are labeled. Many stereology studies regard volume measurement as an end product and discard the raw grid, but in CRISP raw grids were preserved. Many more advanced features are possible, but due to high human interaction they were not previously considered feasible to obtain. We hypothesize raw stereology grids provide a rich source of a priori information, which can effectively constrain soft tissue segmentations of kidneys in MRI images, permitting efficient extraction of advanced analytic features that may better predict individual clinical prognosis and therapy requirements. This work will obtain segmentations of ADPKD kidneys using nothing but raw stereology grids, and optimize the results possible with this constraint. We are also evaluating the predictive value of a promising advanced feature called Cyst-Parenchyma Surface Area in a pilot study. Notably, these segmentations and advanced features will require no additional human interaction when applied to studies where stereology was routinely conducted, including the CRISP cohort.

Public Health Relevance

This groundbreaking research will open an entirely new field of study to predict disease progression, therapy responsiveness, and secondary risks of autosomal dominant polycystic kidney disease (ADPKD). The analysis builds upon data already routinely collected in ADPKD imaging studies, allowing broad retrospective application rather than requiring new initiatives. At time of submission proof-of-concept is complete and one promising metric has been found;the studies herein will bring these fledgling ideas to completion and full application.

Agency
National Institute of Health (NIH)
Institute
National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
Type
Individual Predoctoral NRSA for M.D./Ph.D. Fellowships (ADAMHA) (F30)
Project #
1F30DK098832-01
Application #
8521858
Study Section
Special Emphasis Panel (ZDK1-GRB-G (J1))
Program Officer
Rankin, Tracy L
Project Start
2013-01-01
Project End
2015-12-31
Budget Start
2013-01-01
Budget End
2013-12-31
Support Year
1
Fiscal Year
2013
Total Cost
$40,932
Indirect Cost
Name
Mayo Clinic, Rochester
Department
Type
DUNS #
006471700
City
Rochester
State
MN
Country
United States
Zip Code
55905
Kline, Timothy L; Irazabal, Maria V; Ebrahimi, Behzad et al. (2016) Utilizing magnetization transfer imaging to investigate tissue remodeling in a murine model of autosomal dominant polycystic kidney disease. Magn Reson Med 75:1466-73
Warner, Joshua D; Irazabal, Maria V; Krishnamurthi, Ganapathy et al. (2014) Supervised segmentation of polycystic kidneys: a new application for stereology data. J Digit Imaging 27:514-9
van der Walt, Stéfan; Schönberger, Johannes L; Nunez-Iglesias, Juan et al. (2014) scikit-image: image processing in Python. PeerJ 2:e453