Mouse skin tumours and human lung cancer: relationships with complex environmental emissions

Mouse skin tumorigenesis has often been used to evaluate the tumorigenic effects of complex mixtures, including human respiratory carcinogens. This study examines the quantitative relationships between tumour induction in SENCAR mouse skin and the induction of respiratory cancer in man using four well established human respiratory carcinogens: emissions from coke ovens, roofing tar pots, diesel engines, and cigarettes. The tumour multiplicity data from mouse skin studies using extracts of particulates from each of the four emission sources are compared and their relative potencies to the coke oven sample are: coke oven:roofing tar:diesel:cigarette smoke condensate; 1.0:0.20:0.15:0.0011. Using the unit risk (the lifetime probability of respiratory cancer death due to a constant lifetime exposure of 1 microgram/m3 emissions in the inhaled air) to compare the potencies of the four emissions to human respiratory cancer, the relative potencies are: coke oven:roofing tar:diesel:cigarette smoke: 1.0:0.39:0.075:0.0024. When fitted to a linear regression, the mouse skin tumour data and human lung cancer risks were highly associated, with a correlation constant of 0.95 and a slope value of 0.89. The close association suggests that this comparative potency approach can be used in prospective analyses of human respiratory cancer risk to complex emissions, and several examples are proposed. In addition, the similarity of the chemical and toxicological characteristics of complex mixtures is discussed with regard to the prospective use of the comparative potency approach.