Comparison of Cancer Slope Factors for Using Different Statistical Approaches
In the past, the cancer slope factor has been calculated as the upper 95% confidence limit on the coefficient (q1*) of the linear term of the multistage model for the extra cancer risk over background. The U.S. EPA's draft final cancer guidelines, released in 2003, however, presents thecancer slope factor as the slope (s*) of the linear extrapolation from tjhe 95% lower confidence limit on dose at the lowest prescribed risk level supported by the data to the origin of the dose-response curve. To what extent do the two approaches differ in practice? We addressed this issue by calculating s* and q1* for a number of datasets drawn from the IRIS database, using the constrained multistage model to fit the dose-response data. For 82% of the datasets, s* is within 0.95 to 1.1 of q1* at the 0.1 response level, and with two exceptions, the ratio s*/q1 is less than 3.0 for the remaining datasets. The analysis validates the argument posed by other authors that the two approaches are likely to give nearly similar results when the multistage model is used. For the vast majority of chemicals, the prescribed E
PA methodology forthe cancer slope factor is neither more nor less protective of public health in comparison with the traditionally estimated q1*. We also examined, for various levels of risk quantiles, the extend to which s* exceeded a slope factor, sc (sc - is in subscript), derived from a point of departure based on the central estimate of the dose response. At the 0.1 response level, the median of the ratio, is s*/sc, is 1.6.