859 and 0.911, respectively). The plot for AD is shown in Fig. 2. In contrast, weak correlations exist between TEWL and AD (R2: 0.598) and maxKp (R2: 0.451) as well as TEER and AD (R2: 0.386) and maxKp (R2: 0.479). The find more quality of fit was not related to the skin preparations used, meaning that good, moderate and poor correlations were obtained with excised human skin, reconstructed human skin as well as excised rat skin. Finally, to assess and compare the variabilities of the integrity tests (TEER, TEWL,

TWF and BLUE), the overall, inter-donor and intra-donor or method variabilities were calculated. The results are given in Table 8. For instance, TEER resulted in CVs of 65%, 45% and 43%, respectively. Furthermore the method variability of the in vitro dermal absorption experiments (45% and 33% regarding AD and maxKp, respectively) and the ISTD (30% and 38% regarding AD and maxKp, respectively) are given. The independency of 3H- and 14C-analytics

was proven by the quantification of 14C-testosterone standards in presence of 3H-testosterone at two dose levels in comparison to 14C-testosterone standards without 3H in the matrix (Fig. 3). The R2 was 0.9991 and the slope 1.0077. No general influencing effects were apparent. This holds also true with 3H-testosterone levels measured without the addition of the 14C-labelled steroid and following the addition of this label at a high and low amount. Then the R2 was 0.9998 and the slope 1.0008 (data Org 27569 not shown). In the very low Bq range (<200 Bq) of

3H-testosterone the presence of 14C increased PD-332991 the variability of 3H-testosterone data. To assess the co-absorption of test compound and internal reference standard, Table 9 lists absorption characteristics for three 14C-labeled test compounds in absence and presence of a 3H-labeled ISTD. Except for a significantly different lag time for 14C-testosterone with and without 3H-caffeine all endpoints of dermal absorption were close and the ISTD did not influence the absorption of the test compound. TEER, TEWL and TWF are widely used skin integrity tests, each with a large historical dataset (Bronaugh et al., 1986, Davies et al., 2004, Diembeck et al., 1999, Elkeeb et al., 2010 and Meidan and Roper, 2008). Nevertheless there are still discussions about the experimental performances, limit values and fields of application (Chilcott et al., 2002, Meidan and Roper, 2008 and Netzlaff et al., 2006). Impairment of the skin barrier identified by these methods is expected to allow excessive penetration and permeation of the test compound and therefore yield invalid results. Usually cut-off values are used to distinguish impaired from intact skin preparations with no intermediate stages: a skin sample is either valid or invalid. This is helpful in case of a pre-test that rejects inappropriate samples for absorption testing.