Issn Print: 0163-4356

Publication Date: 2005/04/01


Print

Validation Of A Robust Cyclosporine LC/MS/MS Method And Inter-method Comparison To FPIA Using AUC Samples: 102

Napoli KL; Cooper DP

+ Author Information

Author Information: University of Texas Medical School at Houston, 6431 Fannin, Houston TX 77030 and

    loading  Checking for direct PDF access through Ovid

Excerpt

We developed a robust cyclosporine (CsA) LC/MS/MS method that uses cyclosporin D (CsD) as internal standard: to 100 ul of 10% methanol in 0.1 M ammonium bicarbonate add 20 ul whole blood, then 100 ul 0.1 M zinc sulfate and 200 ul methanolic CsD. Detect the [M + H]+ of the [M + NH4] + adducts with positive ESI. Calibration curves prepared in fresh (never frozen) and previously frozen whole blood, and Chromsystems CsA-free whole blood matrix were identical indicating that the method accurately quantitates CsA in various matrices. The linear range is 10-2000 ng/ml with LLOQ of 2.5 ng/ml. Extraction efficacy of CsA and CsD are 92 and 98%. Recovery of CsA added to C0 and C2 samples was 102%; using a 2X dilution with fresh whole blood results were 101% of the original value. Eighteen samples from the International Proficiency Testing Scheme gave results that were + 3% (−1 to + 9%) of the target values.
Sixty C0 and 59 C2 samples from outpatient and 12-hour AUCs on 27 inpatient renal transplant recipients were analyzed. Mean FPIA C0 was 98 (26-317) and LC/MS/MS was 63 (17-221) ng/ml, while mean FPIA C2 was 433 (34-1383) and LC/MS/MS was 351 (24-1169) ng/ml. Weighted linear regressions showed r2 values of 0.936 and 0.994. Using differences plots FPIA averaged 50% (29-165%, r2 = 0.046) and 20% (9-69%, r2 = 0.503) above LC/MS/MS for C0 and C2. By difference plot the FPIA AUCs averaged 47% (20-102%) above the LC/MS/MS values: by linear regression FPIA = 1.49 X LC/MS/MS - 28 (r2 = 0.952). By either method C2 was most poorly correlated with AUC (r2 = 0.507) while C4 and C12 were highest (r2 = 0.782). C0 + C4 estimated the AUC well (r2 = 0.925). Especially at these low CsA levels to avoid nephrotoxicity, FPIA provides highly inaccurate C0 information. Poor correlation of C2 with AUC indicates poorly timed sample collection. Such gross errors caused by method or sample timing make for inaccurate estimations of drug exposure. Use of LC/MS/MS and appropriate sample timing will make for better clinical decisions regarding CsA TDM.

Related Topics

    loading  Loading Related Articles