© 1984 Oxford University Press
research-article |
The Importance of Delivered Dose in Estimating Low-Dose Cancer Risk from Inhalation Exposure to Formaldehyde
Department of Epidemiology, Chemical Industry Institute of Toxicology P.O. Box 12137, Research Triangle Park, North Carolina 27709
The Importance of Delivered Dose in Estimating Low-Dose Cancer Risk from Inhalation Exposure to Formaldehyde. STARR, T. B., AND BUCK, R. D. (1984). Fundam, AppL Toxicol. 4, 740–753. Data have recently been obtained on the concentration of formaldehyde covalently bound to the respiratory mucosal DNA of Fischer-344 rats following two 6-hr inhalation exposures to gaseous formaldehyde. These data provide a direct short-term measure of the delivered formaldehyde dose in target tissue as a function of the formaldehyde concentration in ambient air. They also demonstrate that the delivered dose/administered dose relationship is significantly nonlinear. Since chronic inhalation exposure of Fischer-344 rats to high concentrations of gaseous formaldehyde induces squamous cell carcinomas of the nasal cavity, and since widespread concern exists that formaldehyde exposure may also pose a cancer risk for humans, the implications of this nonlinearity for low-dose risk extrapolation were investigated. The incidence of nasal squamous cell carcinomas in a chronic formaldehyde inhalation bioassay was reanalyzed with several low-dose extrapolation models, using the estimated concentration of formaldehyde covalently bound to respiratory mucosal DNA as the measure of exposure. For this purpose, it was assumed that the short-term observations of covalent binding were representative of steady-state conditions during the course of the chronic study and further, that the covalent binding of formaldehyde to target tissue DNA is an important factor in nasal tumor induction. Resulting maximum likelihood risk estimates and upper 95% confidence bounds were unilaterally lower than the corresponding risk measures based on administered dose, irrespective of the dose-response model employed. Reductions in estimated risk ranged from a factor of 2.5, for the multistage model upper 95% confidence bound, to over 10 orders of magnitude, for the probit model upper 95% confidence bound. These results indicate that the concept of delivered dose can have a significant impact on estimates of low-dose risk and should therefore at least be considered as an alternative dose measure in assessments of human cancer risk from formaldehyde exposure