Dose-response and threshold analysis of tumor prevalence after intratracheal instillation of six types of low- and high-surface-area particles in a chronic rat experiment

Over a period of 20 years, rat experiments have consistently shown tumorigenic responses after exposure to poorly soluble low-toxicity particles (PSP). We performed a rigorous dose threshold analysis of a large previous rat study. Seven hundred and nine rats were intratracheally exposed to five different PSP: carbon black and titanium dioxide of low and high surface area, diesel emission particles (low surface area), and one soluble dust (amorphous silica, high surface area), at varying instilled total mass doses ranging from 3.0 mg to 120 mg. A multivariable Cox model was applied to analyse lung tumor prevalence. The model was extended by a dose threshold or a dose saturation parameter. This statistical approach, which is new in animal studies, showed no better fit when using surface area or volume as dose metrics but found significantly higher tumor prevalence in animals instilled with high-surface-area dust particles. Interestingly, a dose threshold of about 10 mg mass dose (0.95 CI: 5 mg to 15 mg) emerged from our calculations. In addition, our statistical analysis demonstrated that tumor prevalence is saturated beyond 20 mg mass dose. In summary, our analyses showed that these data are compatible with earlier observations that high-surface-area particles induce more lung tumors and support the concept of a dose threshold for lung tumor after PSP exposures in the rat. However, collinearities in the data (particle type and dose were correlated by design) and the saturation phenomenon (506 out of 709 rats were exposed above the estimated saturation dose) limit generalization of these findings.