Methylclostebol Metabolism Discovery by Untargeted Analysis

ABSTRACT Rationale This work introduces an alternative experimental approach by integrating high‐resolution mass spectrometry (HRMS) with multivariate statistical analysis for metabolite detection and identification. The integration of these tools maximizes information extraction from data, improving accuracy and reducing the risk of false identifications. Methods Seven volunteers' urine samples were collected before and after oral administration of 10 mg of methylclostebol (4‐chloro‐17β‐hydroxy‐17α‐methylandrost‐4‐en‐3‐one, ClMT) and assigned to three excretion time intervals. Analyses were carried out on a GC–HRMS system (Agilent 8890 GC coupled with 7250 GC/QTOF), utilizing low‐energy electron ionization (< 18 eV) to preserve the native molecular skeleton, thereby simplifying mass spectrum interpretation, with acquisition in full scan mode. Raw data were then processed and subjected to multivariate analysis. Results The orthogonal partial least squares‐discriminant analysis (OPLS‐DA) was employed to emphasize differences among specific sample conditions, and features that significantly contribute to classification in the OPLS‐DA can be identified as important biomarkers. Samples from the three excretion intervals demonstrated clear separations, occupying distinct areas within the model's defined space. From this approach, the S‐plot displayed seven features identified as biomarkers related to methylclostebol ingestion, comparing their mass spectra with an in‐house library of LE‐EI mass spectra. Conclusions The application of this approach is demonstrated to enhance the identification of new markers related to the intake of prohibited substances in the anti‐doping field, such as methylclostebol. Its application proved to be an alternative strategy that allows for gathering a more comprehensive range of information in the antidoping field.