Cisplatin dose adjustment in patients with renal impairment, which recommendations should we follow?

BACKGROUND: Nephrotoxicity is the dose-limiting side effect of cisplatin justifying the assessment of renal function for dose adjustment. OBJECTIVE: To determine whether appropriate dose adjustment is made in patients with renal impairment using the Cockcroft-Gault (CG) or the abbreviated Modification of Diet in Renal Disease (aMDRD) formulas to estimate the glomerular filtration rate (GFR). SETTING: The study was conducted in a 1,000-bed university hospital. METHOD: Two years of cisplatin prescriptions were retrospectively compared to the 4 and 3 ranges estimated glomerular filtration rate (eGFR)-stratified dosing recommendations (4RR and 3RR respectively). MAIN OUTCOME MEASURE: Cisplatin dose in mg/m(2) based on kidney function and according to the dosing recommendations. RESULTS: Among 1,364 cycles of cisplatin, 156 (11.4 %) were prescribed for 70 patients with eGFR < 60 mL/min and a median age of 67.4 years. For 57 (36 %) of these cycles, doses were not reduced. When reduced, prescribed doses were not different than recommended doses according to 4RR using CG (% of protocol, 63 ± 12 vs. 64 ± 17) while it was significantly lower using aMDRD (% of protocol, 66 ± 12 vs. 81 ± 22, p < 0.01) and significantly higher according to 3RR using both CG and aMDRD (% of protocol, 63 ± 12 vs. 50 ± 3 and 66 ± 12 vs. 50.7 ± 4.0 respectively, p < 0.01). Prescription of at least one appropriate dose according to 4RR and using aMDRD was associated with a statistically significant higher median total cumulative dose (% of protocol, 89.9 vs. 75.1 % respectively, p < 0.01) without higher decrease of eGFR over time. CONCLUSION: Cisplatin dose adjustment in patients with renal impairment must be improved. Estimating GFR with the aMDRD formula and adding an intermediary level of dose reduction for patients with eGFR from 50 to 59.9 mL/min may result in a higher cumulative dose of cisplatin without higher renal toxicity, which may significantly impact on the effectiveness of the chemotherapy. A prospective evaluation remains needed to assess the benefit/risk ratio of this dose adaptation schedule, taking into account the variability of the GFR estimates.