Rapid semi-automated simplified liquid extraction (SLE) and data processing for 43 drugs of abuse in oral fluid

Abstract

Oral fluid (OF) is commonly used as a non-invasive biological specimen to screen for the presence of drugs. Since 2004 in Victoria, OF has been used by law enforcement to randomly screen drivers on the roadside for the presence of methylamphetamine, MDMA and tetrahydrocannabinol. Over time, the drug-testing program has expanded to 150,000 random tests per year and produces approximately 12–14,000 positive roadside detections for laboratory confirmation. This sample load is a significant burden for manual sample preparation techniques. The need for high-throughput detection methods to maintain strict turnaround times has driven procurement and development of a cost-effective benchtop liquid handing system (LHS) for a rapid and robust sample preparation protocol at a fraction of traditional LHS expense. Rapid automated sample preparation was performed using the Integra Assist PLUS variable-channel width pipetting system. This system allowed for the development of a modular, programmable simplified liquid extraction (SLE) process using Biotage Isolute+ 96-well plates (400 μL). In addition to the three target drugs in these OF specimens, a further 40 drugs of abuse were targeted for detection and quantitation (1–50 ng/mL), including classes of amphetamines, benzodiazepines, cannabinoids, stimulants, and opioids. For extraction, 100 uL OF (analysed in duplicate) was added to a 2 mL polypropylene tube and placed on the LHS. The custom designed automated program then added 200 uL ammonium carbonate buffer (2 M, pH9), mixed the samples and transferred to a 96-well SLE plate. Positive pressure nitrogen gas was applied to ensure samples transferred into the SLE sorbent before a second LHS program added 900 μL chloroform:isopropanol (9:1) to each well. Eluate was collected into 1 mL glass inserts containing 50 μL 0.1 M hydrochloric acid and evaporated to dryness. A final LHS program added 100 μL methanol to each well for reconstitution. The 96-well plate was then placed on a Shimadzu Nexera X2 liquid chromatography system and 1 μL injected onto an analytical column [Kinetex C18, 4.6 × 50 mm, 2.6 μm, mobile phase of 50 mM ammonium formate buffer (pH 3.5)/0.1% formic acid in acetonitrile, 5 minute gradient separation]. Analytes were detected using a SCIEX 4500 Triple Quadrupole mass spectrometer in positive electrospray MRM mode with the use of carbon-13 isotopes to assist in decreasing detector response and aid MRM ratio confirmation. Data was processed using MultiQuant software utilising custom queries for rapid data review, assessment of duplicate concentrations and report generation. The method was fully validated in accordance with international guidelines for OF collected in StatSure and Quantisal buffers. Matrix effects and extraction efficiencies were acceptable with most analytes showing > 80% response and low variation (less than 5% RSD). Precision, accuracy and repeatability were typically within 6% RSD for all analytes, except for EDDP (methadone metabolite) and norbuprenorphine which showed poor precision in StatSure buffer, while the 7 aminobenzodiazepine metabolites showed poor precision in Quantisal buffer. The Integra Assist PLUS automated pipettor was sourced for approximately one tenth the price of a full LHS system, and only occupies a fraction of the bench space. Further, the reliable automated process reduces the risk of sample well errors or spills. The modular set-up of the technique required minimal manual handling and was able to prepare a complete 96-well plate for analysis within 90 minutes. Further, the use of a 3D printer allowed for design and implementation of custom equipment within the workflow. Finally, automated custom queries used for assessment of results ensured tight turnaround times are maintained. The developed fast and reliable extraction technique combined with rapid LC-MS/MS analysis of a broad range of drugs in OF allows for the opportunity of greater throughput, decreased turnaround times and lower risk from manual handling processes.