Abstract Neurodegenerative properties of acrylamide were studied in vitro by exposure of differentiated SH-SY5Y human neuroblastoma cells for 72 h. The number of neurites per cell and the total cellular protein content were determined every 24 h throughout the exposure and the subsequent 96-h recovery period. Using kinetic data on the metabolism of acrylamide in rat, a biokinetic model was constructed in which the in vitro toxicity data were integrated. Using this model, we estimated the acute and subchronic toxicity of acrylamide for the rat in vivo. These estimations were compared to experimentally derived lowest observed effect doses (LOEDs) for daily intraperitoneal exposure (1, 10, 30, and 90 days) to acrylamide. The estimated LOEDs differed maximally twofold from the experimental LOEDs, and the nonlinear response to acrylamide exposure over time was simulated correctly. It is concluded that the integration of the present in vitro toxicity data with kinetic data gives adequate estimates of acute and subchronic neurotoxicity resulting from acrylamide exposure.