Application of non-targeted analysis for the identification of uncommon or unreported chlorinated contaminants in human milk

Many known halogenated substances have been shown to accumulate in the human body and transfer into maternal milk, with potential adverse health consequences for the breastfeeding infant. However, few studies have applied non-targeted analysis techniques to identify additional, possibly unknown, halogenated contaminants in human milk. Using HaloSeeker, a powerful non-targeted data-mining software, several chlorinated-related features of interest were extracted from data obtained for 594 human milk samples collected in Canada (Montreal) and South Africa (Vhembe and Pretoria) during 2018 - 2019. In total, six compounds were confirmed at confidence level 1 using analytical standards: two antimicrobials (chloroxylenol and triclocarban), two pesticide-related chemicals (propanil and 4-hydroxy-2,5,6-trichloroisophthalonitrile), one ultraviolet filter (bumetrizole), and one antiretroviral metabolite (8-hydroxyefavirenz). These findings demonstrate the effectiveness of non-targeted approaches in identifying uncommon or previously unreported chlorinated contaminants and their metabolites in human milk, and underscore the need for further research to assess their potential health risks.