Validation of high-performance liquid chromatographic assay methods for the analysis of carboplatin in plasma ultrafiltrate

Validation of two HPLC assays for the quantitation of carboplatin in human plasma ultrafiltrate is described. Both assay methods employed a YMC ODS-AQ 3.9×150 mm (3 μm) column for the chromatographic separation. The first method utilized direct UV detection, the second method utilized UV detection following post-column derivatization with sodium bisulfite. Structural analogues of carboplatin were synthesized and used as internal standards for the assays. With direct UV detection, sample clean-up using solid-phase extraction on amino cartridges was required prior to injection, with extraction recoveries ranging from 80 to 90%. This extraction procedure was not necessary with the post-column reaction method, which employed a more selective analytical wavelength. Unfortunately, instability of the post-column reagent was a problem and led to greater variability in predicted concentration values. For standard curves, a weighted (1/y2) regression approach was used for plots of peak area or peak height ratio (carboplatin/internal standard) vs. carboplatin concentration. The limit of detection of both assays was 0.025 μg/ml and both were validated for carboplatin concentrations from 0.05 to 40 μg/ml. Accuracy and precision data were generated using three batches of validation samples, each batch consisting of a standard curve and five sets of quality control samples. Stability of carboplatin in blood, plasma, plasma ultrafiltrate, and reconstituted extracts was evaluated. The assay methods were employed for the pharmacokinetic analysis of blood samples drawn from a pediatric patient that received a 400 mg/m2 dose of carboplatin.