Recommendations on dose level selection for repeat dose toxicity studies.

Prior to registering and marketing any new pharmaceutical, (agro)chemical or food ingredient product manufacturers must, by law, generate data to ensure human safety. Safety testing requirements vary depending on sector, but generally repeat-dose testing in animals form the basis for human health risk assessments. Dose level selection is an important consideration when designing such studies, to ensure that exposure levels that lead to relevant hazards are identified. Advice on dose level selection is provided in test guidelines and allied guidance documents, but it is not well harmonised, particularly for selection of the highest dose tested. This paper further builds on concepts developed in a technical report by the European Centre for Ecotoxicology and Toxicology of Chemicals (ECETOC) which recommends pragmatic approaches to dose selection considering regulatory requirements, animal welfare and state of the art scientific approaches. Industry sectors have differing degrees of freedom to operate regarding dose level selection, depending on the purpose of the studies and the regulatory requirements/legislation, and this is reflected in the overall recommended approaches. An understanding of systemic exposure should be utilised where possible (e.g., through toxicokinetic approaches) and used together with apical endpoints from existing toxicity studies to guide more appropriate dose level selection. The highest dose should be limited to a reasonable level, causing minimal but evident toxicity to the test animals without significantly compromising their well-being. As the science of predictive human exposure further develops and matures, this will provide exciting and novel opportunities for more human-relevant approaches to dose level selection.

Dosimetry by means of DNA and hemoglobin adducts in propylene oxide-exposed rats.

The main purpose of the study was to establish the relation between exposure dose of propylene oxide (PO) and dose in various tissues of male F344 rats exposed to the compound by inhalation. The animals were exposed to 0, 5, 25, 50, 300, or 500 ppm PO in the air for 3 days (6 h/day) or 4 weeks (6 h/day, 5 days/week). Blood, nasal respiratory epithelium, lung, and liver were collected. 2-Hydroxypropylvaline (HPVal) in hemoglobin was quantified using the N-alkyl Edman method and gas chromatography/tandem mass spectrometry. 7-(2-Hydroxypropyl)guanine (7-HPG) in DNA was quantified using (32)P postlabeling. The levels of 7-HPG in DNA of nasal respiratory epithelium and lung increased linearly with concentration as measured both after 3 days and 4 weeks of exposure. Similarly, 7-HPG in liver DNA and HPVal in hemoglobin showed a linear increase with PO concentration in the 3-day exposure group, whereas a deviation from linearity was observed above 300 ppm in the 4-week exposure group. The new results confirm previous observations of a dose difference between tissues with the highest dose present in the nasal respiratory epithelium. The measured adduct levels were used for calculation of adduct increments and corresponding tissue doses per unit of external exposure dose. For this purpose, the buildup of adducts was modeled considering the different kinetics of formation and elimination of adducts with DNA and hemoglobin, respectively, and also considering the increasing body weight of the animals. The half-life of 7-HPG in vivo, as well as tissue doses, could be solved from DNA adduct data at the 3rd and 26th days. Within the range of concentrations where the dose-response curves for adduct formation are linear, the relationship between exposure dose and resulting tissue doses could be based equally well on adduct data from the short-term exposure as on adduct data from the prolonged exposure.