Understanding and Reporting PFAS Exposure: Wide-scope Retrospective Suspect Screening of High-resolution Mass Spectrometry Data from Exposomics Studies

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 MS2 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 MS2 similarity, followed by spectral quality filtering and retention time-index regression models based on isotopically labeled standards, reinforcing identification confidence. The reanalysis confirmed up to 80% of previously reported PFAS while uncovering overlooked high-confidence carboxylic, sulfonic, and sulfonamidoacetic acid homologs, and >30 tentative PFAS series, encompassing isomeric clusters (alcohols and ethers) and structurally distinct subclasses (alkylpyrimidine PFAS). The reanalysis framework also revealed shared homolog patterns across exposure contexts. By framing PFAS exposure in chemical space and providing a reusable repository of confirmed and tentative structures, this framework enables reinterpretation of HRMS biobanked data and improved cross-population PFAS assessment.