Population-specific transcriptomic bioactivity assessment of perfluorooctanoic acid using human liver spheroids and...

Perfluorooctanoic acid (PFOA) has been widely used in industrial and consumer products and was recently classified as a Group 1 carcinogen by the International Agency for Research on Cancer. However, uncertainties remain regarding transcriptomic bioactivity thresholds associated with liver effects and their variability across clinically relevant populations. In this study, previously published transcriptomic data obtained following 24-h and 10-day exposure of primary human liver spheroids (LS) to PFOA were integrated into a physiologically based toxicokinetic (PBTK), population-specific framework to estimate transcriptomic human equivalent doses (HEDs). In addition, exploratory concordance analysis was performed between PFOA-responsive genes and liver hepatocellular carcinoma (LIHC)-associated transcriptomic profiles. Median transcriptomic HEDs ranged from 283 to 412 µg/kg body weight (bw)/week and were substantially higher than the estimated HEDs corresponding to the 95th percentile serum PFOA concentration in the U.S. population (0.0099 µg/kg bw/week) and the mean PFOA concentrations reported in human liver tissue samples (0.035 µg/kg bw/week). Obesity class III, cirrhosis Child-Pugh classes A and B, and non-alcoholic fatty liver disease reduced transcriptomic HEDs by approximately 31-41% compared with the healthy population, although HEDs remained substantially above reference values. Gene-level analysis identified only a limited number of genes within the bioactivity exposure ratio range of 1-100. More than 50% of PFOA-responsive genes exhibited concordant expression patterns with LIHC-associated transcriptomic profiles. Although the biological significance of these overlaps remains uncertain, concordance significantly exceeded that expected by chance from a background gene set and was driven mainly by down-regulated genes. Overall, this study provides a population-informed framework for integrating transcriptomic bioactivity data with PBTK modeling in human-relevant assessment of PFOA-associated liver effects.