Carboplatin dosing based on measurement of renal function--experience at the Peter MacCallum Cancer Institute.

BACKGROUND Carboplatin is used in treating many types of cancer. Because renal excretion is the major variable determining the pharmacokinetics of this drug, a dosing formula based on glomerular filtration rate (GFR) has been proposed and is being increasingly used in carboplatin dosing. This method of dosing is critically dependent on accurate measurement of GFR. AIMS To report the experience at a single major oncology centre of carboplatin dosing based on GFR, and comparisons of different methods of measuring renal function for use in this dosing method. METHODS An initial group of patients (n = 24) was studied where GFR was measured by Cr51EDTA clearance and compared to Tc99mDTPA clearance, measured 24 hour urine creatinine clearance and the Cockcroft and Gault formula. The carboplatin area under the plasma concentration versus time curve (AUC) was calculated using total platinum measured in a single blood sample assayed by flameless atomic absorption spectrophotometry. A subsequent patient group (n = 16) was then studied using Tc99mDTPA clearance to measure GFR. RESULTS Carboplatin dosing using Cr51EDTA clearance to measure GFR was accurate (< 25% difference between planned and measured AUC) in 87% of samples. Estimation of renal function using the Cockcroft and Gault formula correlated with Cr51EDTA clearance only in patients with GFR < 100 mL/minute. The measured 24 hour urine creatinine clearance did not correlate with Cr51EDTA clearance. Using Tc99mDTPA clearance to measure GFR, carboplatin dosing was accurate in 81% of samples. Across a GFR range of 42-239 mL/minute, the Cr51EDTA and Tc99mDTPA clearance were closely correlated (r = 0.98, slope of regression line = 1.02). CONCLUSIONS Carboplatin dosing using a pharmacological formula based on GFR produces accurate targeting of the carboplatin AUC. Tc99mDTPA clearance can be used to measure GFR instead of Cr51EDTA clearance, which is both more convenient and has potential cost savings. Estimates of renal function using the Cockcroft and Gault formula or measured 24 hour creatinine clearance are insufficiently accurate to use for carboplatin dosing.