Performance of formulae based estimates of glomerular filtration rate for carboplatin dosing in stage 1 seminoma.

BACKGROUND

Single cycle carboplatin, dosed by glomerular filtration rate (GFR), is standard adjuvant therapy for stage 1 seminoma. Accurate measurement of GFR is essential for correct dosing. Isotopic methods remain the gold standard for the determination of GFR. Formulae to estimate GFR have improved the assessment of renal function in non-oncological settings. We assessed the utility of these formulae for carboplatin dosing.

METHODS

We studied consecutive subjects receiving adjuvant carboplatin for stage 1 seminoma at our institution between 2007 and 2012. Subjects underwent 51Cr-ethylene diamine tetra-acetic acid (EDTA) measurement of GFR with carboplatin dose calculated using the Calvert formula. Theoretical carboplatin doses were calculated from estimated GFR using Chronic Kidney Disease-Epidemiology (CKD-EPI), Management of Diet in Renal Disease (MDRD) and Cockcroft-Gault (CG) formulae with additional correction for actual body surface area (BSA). Carboplatin doses calculated by formulae were compared with dose calculated by isotopic GFR; a difference <10% was considered acceptable.

RESULTS

115 patients were identified. Mean isotopic GFR was 96.9 ml/min/1.73 m(2). CG and CKD-EPI tended to overestimate GFR whereas MDRD tended to underestimate GFR. The CKD-EPI formula had greatest accuracy. The CKD-EPI formula, corrected for actual BSA, performed best; 45.9% of patients received within 10% of correct carboplatin dose. Patients predicted as underdosed (13.5%) by CKD-EPI were more likely to be obese (p=0.013); there were no predictors of the 40.5% receiving an excess dose.

CONCLUSIONS

Our data support further evaluation of the CKD-EPI formula in this patient population but clinically significant variances in carboplatin dosing occur using non-isotopic methods of GFR estimation. Isotopic determination of GFR should remain the recommended standard for carboplatin dosing when accuracy is essential.