Extension of the Calvert formula to patients with severe renal insufficiency

The Calvert formula was derived from the study among patients with glomerular filtration rates (GFRs) of 33-135 ml/min, and it remains unclear whether the formula can be used to calculate optimal and safe dosages of carboplatin in patients with severe renal insufficiency. We evaluated the utility of this formula in patients with severe renal insufficiency. For pharmacokinetic analysis, we studied nine adult Japanese patients with advanced cancer who had an estimated GFR of lower than 30 ml/min/1.73 m(2), as calculated by the Japanese equation for estimating GFR, or who were receiving hemodialysis. The dose of carboplatin was calculated with the Calvert formula, in which GFR was measured by inulin clearance or was assumed to be 0 in patients requiring hemodialysis. Hemodialysis was started 23 h after the end of carboplatin infusion. Although there was a significant correlation between the estimated and measured carboplatin clearance, the estimated clearance was consistently higher than the measured clearance [mean prediction error +/- A standard deviation = 41.0 +/- A 26.3 %] in all seven patients with renal insufficiency (GFR, median 21.4, range 7.8-31.4 ml/min) and in the two hemodialysis patients. Actual areas under the concentration-time curve (AUC) (mg/ml min) were 5.4, 5.7, 6.2, and 9.0 for the four patients with a target AUC (mg/ml min) of 5; 5.7, 6.2, and 7.1 for the three patients with a target AUC (mg/ml min) of 4; and 5.1 and 8.7 for the two hemodialysis patients with a target AUC (mg/ml min) of 5. The measured clearance of carboplatin ranged from 23.0 to 51.3 ml/min in the seven patients not receiving hemodialysis. The pre-hemodialysis carboplatin clearance in the hemodialysis patients was 20.5 and 11.1 ml/min, respectively. For adult patients with severe renal insufficiency, the Calvert formula causes carboplatin overdosing by overestimating the carboplatin clearance.