Administration of hormonal compounds as growth promoters in livestock farming was banned by Council Directive 93/22/EC, however, this kind of substances are sometimes reported within the framework of European monitoring residue plans. various analytical methods have been previously developed to screen for their misuse, and they are now especially efficient for monitoring the illegal administration of synthetic and semi-synthetic hormones. Nevertheless, proving an exogenous administration of hormones from natural origin (i.e. estradiol-17 beta or progesterone) still remains a challenging task for European authorities. As a result of their origin, these target compounds are indeed always present in the analytical matrix, and because the concentration levels of natural steroids are extremely variable from one animal to another, the establishment of reference thresholds appears very difficult. During this preliminary study, metabolomic data was acquired on a high performance liquid chromatography system coupled to high resolution mass spectrometer (HPLC-LTQ-Orbitrap). Serum samples were collected from dairy cows treated or not with sex steroid hormones commonly employed in animal husbandry: estradiol-17 beta (or its ester estradiol benzoate) and progesterone. After appropriate data processing and multivariate statistical analyses (OPLS-DA), it was possible to highlight significant metabolic modifications in serum consecutively to the administration of estradiol and/or progesterone. Those differences were used to build predictive models able to suspect illegal administration of these hormones in cattle. Potential biomarker candidates of estradiol and/or progesterone were pointed out, that remains to be structurally elucidated. (C) 2011 Elsevier B.V. All rights reserved.