A review of current toxicological concerns on vanadium pentoxide and other vanadium compounds: gaps in knowledge and directions for future research.

Vanadium pentoxide (V(2)O(5)) and other inorganic vanadium compounds have recently been evaluated by several occupational exposure limit (OEL) setting (occupational exposure limit, OEL) committees and expert groups in response to the publication of several new studies, including the U.S. National Toxicology Program (NTP, 2002) carcinogenicity study of inhaled V(2)O(5) in rats and mice, which concluded that clear evidence of lung tumors was seen in mice of both genders and that there was some evidence of carcinogenicity in male rats. This study reviews the expert evaluations of several OEL committees and expert groups and attempts to understand the strengths and weaknesses in their scientific arguments. This study also evaluates some key studies relating to potential genotoxicity, carcinogenicity, and respiratory effects of vanadium compounds and discusses how they might elucidate the mechanism(s) by which V(2)O(5) induces lung cancer in mice. All expert groups appear to agree that the lung tumors induced in mice in the NTP (2002) study are a site-specific response and, in general, verify that existing in vitro and in vivo studies suggest that tumors were induced by a secondary mechanism (presumably non-genotoxic), which is supported, though not conclusively, by a mechanistic data set. As some vanadium compounds produce a range of DNA and chromosome damage, there is no consensus on which of these changes is critical for the carcinogenic process for V(2)O(5) or whether the findings for the lung tumors seen in mice exposed to V(2)O(5) can be extrapolated to other inorganic vanadium compounds. As such, the various expert committees used the evidence differently, some to read across, i.e., to predict an endpoint for a substance based on the endpoint information of another with similar characteristics (e.g., physicochemical properties [solubility, bioaccessibility, bioavailability], structure, fate [toxicokinetics], and toxicology) for carcinogenicity from V(2)O(5) to other inorganic vanadium compounds. It is noteworthy that the toxicity of metals does not necessarily relate to carcinogenicity in a direct manner; thus, no assumptions should be made a priori when trying to extrapolate from V(2)O(5) to other inorganic vanadium compounds. Recent studies evaluated in this review provided some further insights into possible mechanisms but do not cover all relevant endpoints, address only a limited number of vanadium compounds, and have not established no-effect thresholds for carcinogenicity or respiratory tract irritation. Thresholds need to be established in order for arguments to be made for setting a health-based OEL for non-genotoxic or secondary genotoxic carcinogens. In conclusion, important knowledge gaps preclude confident classification and risk assessment for all vanadium compounds. Evidence suggests that further research that may address some of these critical gaps is needed.