Clinical studies using undiluted products raise both scientific and ethical concerns, so experiments have to be carefully controlled. In clinical studies, the test material (often very small volumes and/or diluted) is usually applied to the upper or lower conjunctival sac, as opposed to the
apex of the cornea as in in vivo rabbit studies ( Freeberg et al., 1986b). This in itself raises concerns about the comparability of the outcomes. TSA HDAC cell line In addition, human testing often investigates the “sting” more so than irritation ( Freeberg et al., 1986b). Studies performed in the 1980s compared results from hundreds of accidental human exposures with Draize and LVET tests ( Freeberg et al., 1984, Freeberg et al., 1986a and Freeberg et al., 1986b). In such a study using human volunteers, household substances commonly associated with accidental exposure (shampoo, hand soap, fabric softeners), exposure see more data was collected under known, controlled conditions to establish the relationship between in vivo animal tests and human exposure effects ( Freeberg et al., 1986b). It was demonstrated that Draize testing was a poor predictor of accidental human eye exposure, whereas LVET correlated well,
although still over-predicted results. Human studies are limited, and are usually comparing human responses with Draize or LVET, as proof-of-principle that LVET is more credible than Draize testing (Roggeband et al., 2000), and not as a comparison for the validation of alternative methods. A prevalent problem is that there is no human database for the development of the prediction models needed in validation studies, thus in vitro toxicity tests are still being compared to rabbit data ( Bagley et al., 2006). Ocular organotypic models are isolated systems that aim Cyclooxygenase (COX) to maintain short-term
normal physiological and biochemical function of the enucleated eye or cornea (Barile, 2010). The test material is often applied neat so is more relevant to industrial testing (Reader et al., 1990) and more faithfully represents accidental exposure. The protocols usually utilize opacitometric and spectroscopic methods for quantitative assessment of changes to the isolated cornea in response to a test material followed by histological analysis. Corneal opacity is also an in vivo corneal endpoint, although the data is observational, so often subjective. Corneal opacity acts as an indicator of protein denaturation, swelling, vacuolation and damage to the epithelium and corneal stroma ( Barile, 2010). Fluorescein retention/leakage of the cornea is often used as a measure of permeability ( Prinsen and Koëter, 1993), although in vivo the iris and the conjunctiva are also involved in ocular irritation, so corneal swelling and histological analysis are often included as additional endpoints in organotypic models ( OECD, 2009a), often to distinguish “borderline” cases.