High Resolution Image Download MS PowerPoint Slide Per- and polyfluoroalkyl substances (PFAS) comprise thousands of persistent and structurally diverse chemicals that contribute to chronic human exposure. Yet, routine biomonitoring targets only a limited set of regulated legacy compounds, underestimating overall exposure. We present a retrospective, wide-scope suspect screening analysis framework to extract PFAS information from archived LC–HRMS exposomics datasets. A curated list of 1,666 PFAS, filtered for reversed-phase LC compatibility and enriched with structural descriptors, predicted retention indices, and MS 2 fragments, was used to reprocess two human biomonitoring studies involving firefighter serum pools and population-based plasma samples. Candidate features were prioritized using precursor accuracy, isotopic patterns, and MS 2 similarity, followed by spectral quality filtering and retention time-index regression models based on isotopically labeled standards, reinforcing identification confidence. The reanalysis recovered up to 80% of previously reported PFAS while revealing overlooked high-confidence carboxylic, sulfonic, and sulfonamidoacetic acid homologues, and >30 tentative PFAS series, encompassing isomeric clusters (alcohols and ethers) and structurally distinct subclasses (alkylpyrimidine PFAS). The reanalysis framework also revealed shared homologue patterns across exposure contexts. By framing PFAS chemical space and providing a reusable repository of confirmed and tentative structures, this framework reinterprets HRMS biobanked data, improving PFAS pattern assessment.
Renai et al. (Tue,) studied this question.