Improved predictions of total kidney volume growth rate in ADPKD using two-parameter least squares fitting

Area of Science:

• Nephrology
• Radiology
• Genetics

Background:

• Autosomal dominant polycystic kidney disease (ADPKD) is a genetic disorder characterized by kidney cyst formation.
• The Mayo Imaging Classification (MIC) predicts kidney growth using a single MRI/CT scan, assuming exponential growth.
• Optimal methods for combining multiple imaging scans to improve growth prediction in ADPKD are not well-defined.

Purpose of the Study:

• To evaluate the accuracy of MIC in predicting kidney growth trajectory in ADPKD patients with longitudinal imaging data.
• To compare MIC predictions with alternative methods, including least squares fitting, using multiple available scans.
• To identify patient subgroups where MIC performance may be suboptimal.

Main Methods:

• Retrospective analysis of ADPKD subjects with at least 8 years of imaging follow-up (mean 11 years).
• Established ground truth kidney growth trajectories from serial MRI/CT scans.
• Compared annualized mean absolute error of MIC predictions against 1- and 2-parameter least squares fitting models.

Main Results:

• 2-parameter least squares fitting demonstrated superior accuracy over MIC in predicting total kidney volume growth rate when 4 measurements were available.
• With 3 measurements, a combination of MIC and averaging showed improvement, but 2-parameter fitting remained more accurate.
• Male sex and PKD2 mutations were associated with poorer MIC performance.

Conclusions:

• In ADPKD patients with 3 or more imaging scans, 2-parameter least squares fitting provides more accurate kidney volume growth rate predictions than MIC.
• MIC's predictive accuracy is particularly limited in males and individuals with PKD2 mutations.
• Longitudinal data and advanced fitting methods can enhance growth prediction accuracy in ADPKD management.