Back to Top

Acute to chronic extrapolations

2 posts / 0 new
Last post
hannaholmquist's picture
Acute to chronic extrapolations


In the paper by Rosenbaum et al. (2011) acute to chronic extrapolation factors are suggested (ED50 (kg/lifetime)=LD50 (mg/kg)/26 (non-cancer) and 129 (cancer). These factors were compared to factors derived by Kramer et al. and others so it is clear that the relationship is to be used directly as expressed, i.e. simply dividing an acute LD50 (in the unit mg/kg bw) by the respective factor (26 or 129) to arrive at a ED50 in the unit kg/lifetime) (please correct me if I am wrong).

The extrapolation factors were derived from regression analysis (fig 7 och 8). I am very interested in understanding how the regression was used to arrive at the extrapolation factors. I cannot follow the explanations in the figure texts completely. I have e-mailed the corresponding author of the paper with no reply and decided therefore to try here, as the factors are suggested to be used with USEtox and thus are relevant for the forum.

Thank you in advance.



Rosenbaum, R., M. J. Huijbregts, A. Henderson, M. Margni, T. McKone, D. Meent, M. Hauschild, S. Shaked, D. Li, L. Gold and O. Jolliet (2011). "USEtox human exposure and toxicity factors for comparative assessment of toxic emissions in life cycle analysis: sensitivity to key chemical properties." The International Journal of Life Cycle Assessment 16(8): 710-727.


USEtox Team
USEtox Team's picture
Extrapolation from LD50 effect information

We recommend to primarily use the non-cancer extrapolation of Fig. 8 in Rosenbaum et al. (2011) to derive the human lifetime non-cancer ED50. The related regression is carried out as log ED50 = a + b x log LD50. Since b=1 (as b is not significantly different from 1 in this regression), this equation becomes log ED50 = a + log LD50. Elevating both side to the power 10 yields 10^(log ED50) = ED50 = 10^a x 10^(log LD50) = 10^a x LD50, with 10^a = 1/26.

For cancer, the correlation of Fig. 7 in Rosenbaum et al. (2011) is only valid to derive a cancer ED50, if there are independent evidences of carcinogenicity (which cannot be based on acute LD50s), since the obtained correlation between cancer ED50s and non-cancer acute test primarily reflects the experimental design of cancer test, that is based a maximum tolerable dose determined using  non-cancer acute dataset. So in short: in most cases, this correlation does not apply unless there has been a chronic cancer test carried out. In general, if chronic tests have been carried out to show carcinogenicity, ED50s should be directly available in the USEtox database (which is itself based on the CPDB database, the Carcinogenic Potency DataBase, available on the NIH portal).