The accuracy of estimating equations for the evaluation of kidney function in adult survivors of unilateral, nonmetastatic, non-syndromic Wilms tumor: A pilot study from the St. Jude Lifetime Cohort Study

Background

Adult survivors of unilateral, nonmetastatic, non-syndromic Wilms tumor (WT) treated with whole abdomen radiation therapy (WART) are at risk for impaired kidney function. The impact of bias and accuracy on estimated glomerular filtration rate (eGFR) among adult survivors of WT has not been well documented.

Procedure

We clinically evaluated male and female WT survivors with creatinine and cystatin C, calculated eGFR using the Chronic Kidney Disease—Epidemiology equations with and without cystatin C, and measured ^99mTc diethylenetriamine pentaacetic acid (DTPA) plasma clearance. WT survivors treated with unilateral nephrectomy (UN), non-nephrotoxic chemotherapy (NNC) and WART or treated with UN, no radiation therapy, and NNC were enrolled. Correlations between ^99mTc DTPA clearance and eGFR were calculated. Bias and the percentage of eGFR calculations that differed from the ^99mTc DTPA clearance by 10% or less (P₁₀) or 30% or less (P₃₀) (accuracy) were calculated.

Results

Among female WT survivors, none of the eGFR calculations was statistically significantly correlated with ^99mTc DTPA clearance. Among both unirradiated and WART-treated male WT survivors, ^99mTc DTPA clearance correlated well with eGFR calculations that included creatinine. eGFR calculations that included creatinine were positively biased among female participants compared to ^99mTc DTPA clearance, and no P₃₀ was greater than 90% among either irradiated males or females.

Conclusions

Among female survivors of unilateral, nonmetastatic, non-syndromic WT who have undergone UN, eGFR is poorly correlated with, is positively biased, and lacks sufficient accuracy, compared to ^99mTc DTPA clearance.