Proctor D, Suh M, Thompson C, Hixon, G. 2015. Inhalation cancer risk assessment of titanium dioxide. Presented at the Society of Toxicology’s 54th Annual Meeting, March 22-26, San Diego, CA.
Abstract
Titanium dioxide (TiO2) is characterized as a poorly soluble particulate (PSP) with low toxicity. It is well accepted that low toxicity PSPs such as TiO2 induce lung tumors in rats when deposition overwhelms particle clearance mechanisms. Epidemiologic studies of workers in TiO2 production have not shown evidence of elevated lung cancer risk, and despite the sensitivity of rats to PSPs, the relevance of PSP-induced tumors to humans has not been completely discounted. TiO2 is listed as a possible human carcinogen, but currently, no environmental toxicity criteria for TiO2 are available. Because TiO2 particles agglomerate, TiO2 is predominantly found as inhalable course particles (PM10) rather than as fine (PM2.5) or ultrafine (nanoscale) particles in the environment. We derived cancer-based toxicity values for fine TiO2 based on data from chronic inhalation studies in rats using multiple approaches including different dose-metrics (regional deposited dose ratios and lung surface area burden), different modeling techniques (benchmark dose modeling, smoothing spline regression, and bilinear modeling), and different extrapolation approaches. Based on empirical evidence and mechanistic support for a mode of action (MOA) for TiO2-induced lung tumor in rats, involving chronic inflammation and cell proliferation, we derived reference concentration (RfC) values, ranging from 14.7 to 220 μg/m3. Despite the questionable relevance of the particle overload MOA in rats for humans and the empirical evidence of a threshold, certain regulatory applications (Proposition 65 in California) necessitate the development inhalation unit risk (IUR) values. Hence, we derived a IURs based on rat lung tumors ranging from 0.0027 to 0.0045 (mg/m3)-1. Risk-based concentrations at a 10-5 theoretical risk and continuous exposure (70 years lifetime, 24 hours per day) are lower than the RfC values but above background. These toxicity values should be useful for regulators interested in setting health-protective standards for exposure to fine TiO2.