Therapeutic drug monitoring and drugs of abuse testing on the cobas® 6000 analyzer series: analytical performance under routine-like conditions

Abstract

The analytical performance of the clinical chemistry module c 501 (cobas 6000 analyzer series) was evaluated for therapeutic drug monitoring and drugs of abuse testing using a spectrum of representative assays. Particular attention was paid to potential interactions between reagents using a simulated routine workload. Within-run and total imprecision were assessed using a selection of representative reagents. Deviation from a consensus mean was tested using samples from a proficiency testing scheme. Method comparison using routine samples was carried out against the MODULAR ANALYTICS SWA and COBAS INTEGRA 800 analysis systems. Total coefficients of variation (CV) ranged from 1.9% to 7.8% for individual drugs, and from 3.2% to 8.6% for drugs of abuse testing. Results from proficiency test samples were between 81% and 125% of the consensus mean for therapeutic drugs. Method comparisons (Passing-Bablok regression) showed overall good comparability to MODULAR ANALYTICS SWA and COBAS INTEGRA 800 systems, with slopes from 0.93 to 1.17 and correlation coefficients r > 0.98. Imprecision in a simulated routine run was tested using a total of 42 methods (10 therapeutic drug monitoring, 9 drugs of abuse testing, 3 enzymes, 12 substrates, 8 specific protein assays). Imprecision in the reference batch run ranged from 0.7% to 5.0% CV for therapeutic drug monitoring assays, except for digoxin (DIG) (7.3%), and from 0.9% to 7.7% for drugs of abuse testing. The CVs of general clinical chemistry and specific protein tests were within the expected limits of 2% and 4%. CV changes in the simulated routine run were within the expected limits for most assays. Negative DeltaCVs (> or = 2%) for DIG, digitoxin (DIGIT), cannabinoids (THC), and phencyclidine (PCP) may indicate improved performance when running these assays in a simulated routine operation. A positive DeltaCV (> or = 3%) was found for amphetamines (AMPHs). In conclusion, the cobas c 501 module seems to be well-suited for routine use as consolidated workstation. Except for a potential interaction with AMPH, as indicated by the positive DeltaCV, no significant interferences from different reagents could be observed during this study.