Dose response for formaldehyde-induced cytotoxicity in the human respiratory tract

Human studies of the sensory irritant effects of formaldehyde are complicated by the subjective nature of some clinical endpoints. This limits the usefulness of such studies for quantitative noncancer risk assessment of airborne formaldehyde. Objective measures of the noncancer effects of formaldehyde, such as the rate of regenerative cellular proliferation (RCP) secondary to cytolethality, can be obtained from laboratory animals but present the challenge of interspecies extrapolation of the data. To the extent that uncertainties associated with this extrapolation can be reduced, however, dose-response data obtained in laboratory animals are a viable alternative to clinical studies. Here, we describe the extrapolation of dose-response data for RCP from F344 rats to humans. Rats inhaled formaldehyde (0, 0.7, 2.0, 6.0, 10, and 15 ppm) 6 h/day, 5 days/week for up to 2 years. The dose response for RCP was J-shaped, with the rates of RCP at 0.7 and 2.0 ppm below but not statistically different from control, while the rates at the higher concentrations were significantly greater than control. Both the raw J-shaped data and a hockey-stick-shaped curve fitted to the raw data were used for predicting the human dose response for RCP. Cells lining the nasal airways of F344 rats and rhesus monkeys are comparably sensitive to the cytolethal effects of inhaled formaldehyde, suggesting that the equivalent human cells are also likely to be comparably sensitive. Using this assumption, the challenge of rat-to-human extrapolation was reduced to accurate prediction of site-specific flux of formaldehyde from inhaled air into the tissue lining the human respiratory tract. A computational fluid dynamics model of air flow and gas transport in the human nasal airways was linked to a typical path model of the human lung to provide site-specific flux predictions throughout the respiratory tract. Since breathing rate affects formaldehyde dosimetry, cytotoxicity dose-response curves were predicted for three standard working levels. With the most vigorous working level, the lowest concentrations of formaldehyde predicted to exert any cytotoxic effects in humans were 1.0 and 0.6 ppm, for the J-shaped and hockey-stick-shaped RCP curves, respectively. The predicted levels of response at the lowest effect concentrations are smaller than can be measured clinically. Published literature showing that the cytotoxicity of inhaled formaldehyde is related to exposure level rather than to duration of exposure suggests that the present analysis is a reasonable basis for derivation of standards for continuous human exposure.