Is TCDD a Threshold Carcinogen? A Quantitative Analysis of the Epidemiological Data. C. R. Kirman and S. M. Hays, ChemRisk, Cleveland OH; Lesa Aylward and N. J. Karch, Karch & Associates, Inc., Washington DC; and D. J. Paustenbach, ChemRisk, Alameda CA
The shape of the dose-response curve for 2,3,7,8-tetrachlorodibenzo(p)dioxin (TCDD) carcinogenesis in humans continues to be a subject of debate. Historically, cancer risk assessment for TCDD has been based on a high-dosed rodent study under the assumption that tumor- response is linear in the low dose region. However, a non-linear dose-response curve with a threshold for tumorigenesis is better supported by mechanistic and animal data. In our previous work, we assembled serum sampling data collected for human cohorts exposed to TCDD. Mean area-under-the-curve, average concentration, and peak concentration values were estimated from reconstructed lifetime serum lipid TCDD concentration versus time curves. The combined data set includes several thousand participants, and covers a broad range of exposures. Monte Carlo methods were employed to address the uncertainty and variability in the dose estimates (peak, average, AUC) and response estimates (SMRs for lung cancer and total cancer) in humans. All cancer mortalities were conservatively assumed to be related to TCDD exposure. The results of our analysis indicate that the epidemiological data for TCDD are consistent with a non-linear dose-response curve, and that with greater than 95% confidence a threshold exists for TCDD carcinogenesis in humans. Furthermore, the effective threshold for TCDD carcinogenesis in humans appears to be well above the background levels measured in the U.S. population. Implications to current risk assessment practices for TCDD based on liver tumor data in high-dosed rats will be discussed.
Go to . . .