Identification of characteristic heroin metabolites in urine based on data-mining technology and multivariate statistics analysis combined with a targeted verification approach for distinguishing heroin abusers.

A common phenomenon shows that ingestion of opium poppy shell-containing drugs can result in a "false-positive" urinalysis test result for mandatory or workplace heroin abuse screening. Owing to the short detection window (8 h in urine) of the characteristic heroin metabolite 6-monoacetylmorphine (6-MAM) confirmation or exclusion of heroin abusers still presents major challenges for toxicologists. In this work, we developed an ultra-performance liquid chromatography-time-of-flight mass spectrometry method (UPLC-TOF-MS) with online data acquisition and multiple post-data-mining technologies combined with a multivariate statistical and batch validation analysis workflow to assess the characteristic urine metabolites of heroin abusers. Based on the proposed methods, 28 characteristic metabolites were structurally identified, and their fragmentation patterns and metabolite pathways were also summarized. Correlation analysis was used to investigate the internal relationship and similarities among the identified metabolites, and seven representative metabolites were selected as "Target-metabolites". Multi-batch urine of samples of heroin abusers were certified based on the UPLC-MS/MS method for further validation of the practicability of using this method for routine analysis. Overall, the target-metabolites can be utilized as assistant "biomarkers" in workplace or mandatory drug screenings. This approach encourages further studies on the development of the "false-positive" identification system.