Nina C Wieland, Noa Dominicus, Jonatan Dias, Hans Mol, Ad M J Ragas, Paul T J Scheepers, Frans G M Russel
Glyphosate is a widely used non-selective herbicide to control weeds, and also used as desiccant before harvest. This study aimed to integrate human kinetic data into a physiologically-based pharmacokinetic (PBK) model to improve risk assessment through reverse dosimetry. Kinetics were investigated in a human volunteer study following oral or dermal administration of glyphosate. Glyphosate and its metabolite, aminomethylphosphonic acid (AMPA), were analyzed in plasma, urine, and feces over a period of 96 h. Following oral exposure, peak plasma concentrations were reached at 2.5 h, while urinary concentrations peaked at 161 µg/L after 13 h. Fecal analysis indicated a substantial unabsorbed fraction, and dermal exposure resulted in minimal systemic uptake reflected by low urinary and blood plasma levels. The urinary excretion kinetics observed here are consistent with previously published findings. Using these data, a PBK model was developed and validated using an external dataset. Reverse dosimetry was then applied to an independent oral dosing study, accurately reconstructing the administered doses with a maximum deviation of 0.4-fold. While the model demonstrates high accuracy and utility for population biomonitoring, uncertainties about glyphosate kinetics remain. Knowledge gaps include glyphosate's interaction with blood proteins and its ionization behavior at physiological pH, which makes it difficult to predict distribution to bone tissue. Despite these limitations, this human dosing study advances the understanding of glyphosate kinetics and provides a validated PBK model capable of converting urinary biomonitoring data into accurate dose estimates, thereby improving the accuracy of glyphosate risk assessment.