Literature data of epidemiologic studies, experimental in vivo studies and in vitro studies were surveyed. The analysis concentrated on the validation of the results of in vitro genotoxicity studies with inhalable fibrous and granular inorganic dusts against the results from epidemiologic and long-term animal studies. The in vivo data implied to form three classes of carcinogenic potency across the dust groups. There were also positive results ("effects") in some tests for nearly all types of dusts that were tested by in vitro genotoxicity tests with mammalian cells (e.g. diesel soot, SiO₂ crystalline and amorphous, TiO₂ fine and ultrafine, carbon black fine and ultrafine, iron oxide fine and ultrafine). Overall, more than one half of the in vitro tests showed positive results. Positive results were most consistently obtained with fibrous dusts (about 70 %). However, no clear correlation of the probability of a positive in vitro test with the in vivo potency classes is seen across all dusts and studies. Statistical analysis of 179 data sets of “GBP, nanomaterials and other dusts” shows a correlation with the type (public/private) of funding or laboratory rather than with physico-chemical particle properties. Determination of sensitivity and specificity of the in vitro methods strongly depends on how the in vivo data are assessed, i.e., which reference information is used. Nearly all types of substances have shown to be carcinogenic in at least one in vivo system, as far as they were investigated. If these findings are considered as valid, i.e., if they are used as the reference "true positive", then the proportion of about 60 % positive in vitro results approximates the "sensitivity" of the in vitro methods; this means, sensitivity is about 60 % then (on average across all test models and substance types). Specificity cannot be estimated meaningfully because no adequate in vivo data are available which exclude a carcinogenic potential with sufficient certainty. However, for nearly all investigated dust types, some doubts have been published in the literature regarding the relevance of the in vivo data for current workplace conditions. Doubts associated, e.g., with claims of the so called overload phenomenon are particularly widespread for (nano-)materials of the GBP type. Some doubts, at least in the sense of a possible or practical threshold, have also been published for known carcinogens like nickel compounds, quartz, diesel soot and chrysotile asbestos. Against this background, the meaning of statistical significance is discussed extensively in this report. Furthermore, recommendations are given here for planning future carcinogenicity studies and in vitro genotoxicity tests. With regard to an overall evaluation based on the current state of knowledge, it is explained that the data on nano-GBP and mechanistic considerations are compatible with a dose-response relationship without threshold. Given the severity of a neoplastic disease, it is prudent to consider the findings in rats and the observed increased risks at former workplaces as a benchmark for acting at lower exposures.

Please download the complete report "Relevance of in vitro methods for the evaluation of chronic toxicity and carcinogenicity of nanomaterials, fine dust and fibers" (in German only).