B-288 Automated Chromatography Review of a Comprehensive 113-Drug Panel in Urine by LC-MS/MS

Background When simultaneously measuring a panel of drugs using liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based methods, multiple-reaction-monitoring (MRM) is used to generate a chromatograph for every precursor-to-product ion m/z transition. The data must be reviewed, verifying that all acceptance criteria (eg, correct ion ratios and retention times) are met before reporting. With drug panels that have >100 target compounds, the number of chromatographs to review per run is >10,000 per 96-well plate. Our objective was to modernize the data review process for our new comprehensive drug LC-MS/MS screening method. Methods The acquisition method consisted of 113 drugs belonging to 12 different categories: anticonvulsants, antidepressants, antipsychotics, barbiturates, benzodiazepines, local anesthetics, opioids, over-the-counters, muscle relaxers, sedatives/dissociative anesthetics, stimulants, and 11-nor-9-carboxy-delta-9-tetrahydrocannabinol (THCCOOH). A linear 3-point calibration model was used. The dilute-and-shoot method included hydrolysis and contained 13 deuterated compounds as internal standards (ISTD). The data was acquired using a SCIEX 5500+ mass spectrometer. All analytes were monitored using 2 MRMs and 1 ion ratio for acceptance. One MRM was used for each ISTD. Automated data processing used Ascent software (Indigo Bioautomation, Carmel, IN). Ascent automation functions with 2 general methodologies; processing and assessing. Processing the chromatography has end-user adaptability to allow refinement in peak selection. During the validation stage, processing parameters were adjusted with multiple instrument data sets and a large data set of patient results. Assessing rules were layered on top of the processing method to bring any peak outside of defined limitations to the users’ attention with flags. Results Optimizing processing parameters reduced the number of irrelevant peaks selected and the customized assessing rules improved the number of flags from thousands to tens per batch for missed peaks, known drug metabolism, potential carryover, quality control, and other reporting criteria. Comparing the legacy library scan method manual review times against Ascent processed runs showed substantial time savings. A run with the legacy method had a processing time of 135 minutes per batch. The Ascent processing time was <15 minutes per batch, a time savings of 2 hours. The new MRM comprehensive method was also processed in SCIEX Multiquant, and review time was similar to legacy method processing times. Using Multiquant to view each chromatograph and then reviewing run summaries to ensure metabolism rules, interference monitoring, and other review criteria was too complex and time consuming for a high-throughput laboratory environment. Conclusion We successfully modernized the data review process for our new comprehensive drug LC-MS/MS screening method using Ascent software optimized for processing and assessing MRM chromatograms. Ascent allows the processing method designers to customize and layer rules and settings so that a specimen with no peaks present or with peaks present within established ranges for all criteria to require no further action from the user. Alternatively, when there are peaks present that do not pass the established rules, those calibrators, QCs or patient specimens will be flagged for manual review before reporting. These rules made chromatography review significantly more manageable and allowed for more rapid training of additional processing technicians.