Archives

  • 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • 2022-09
  • 2022-10
  • 2022-11
  • 2022-12
  • 2023-01
  • 2023-02
  • 2023-03
  • 2023-04
  • 2023-05
  • 2023-06
  • 2023-07
  • 2023-08
  • 2023-09
  • 2023-10
  • 2023-11
  • 2023-12
  • 2024-01
  • 2024-02
  • 2024-03
  • 2024-04

    • As shown in Table the results of

    2022-06-24

    As shown in Table 3, the results of individual prediction modeling with chemicals at 0.4 mM, 0.6 mM, and 0.8 mM would lead to lesser sensitivity of 60.0%, 65.0%, and 65.0%, respectively, in microsome-unused group, and 85.0%, 85.0%, and 90.0%, in microsome-used group. If we consider only the results from either microsome-used group, the chemicals, such as propyl gallate and lauryl gallate, would also be considered as non-sensitizers, and the accuracy would be lesser because of the effect of incorrectly judged chemicals. However, if we consider at least one positive in either microsome-used or -unused group as a sensitizer, the sensitivity would be improved to 95.0%, 95.0%, and 95.0% with the chemical concentrations of 0.4 mM, 0.6 mM, and 0.8 mM, respectively (Table 4). Thus, a combination model in which either one positive result from either microsome-used or -unused group was regarded as a sensitizer would be our final model rather than individual models. As shown in Supplementary Table 3, the current method was compared with the OECD-adopted methods, such as DPRA, KeratinoSens and h-CLAT, and 2 out of 3 w weight of evidence (WoE) approach (Clouet et al., 2017; Urbisch et al., 2016). We found that test chemicals, such as methyl eugenol, (R)-(+)-limonene, tween 80 and 4-aminobenzoic CP-673451 receptor were not studied in all the above in vitro methods. In addition, test chemicals, such as metol, isoeugenol, 3-aminophenol, farnesol, aniline, geraniol, and salicylic acid showed variable results in different studies. Because of the insufficient data, a final decision could not be made with test chemicals, such as methyl eugenol, (R)-(+)-limonene, tween 80 and 4-aminobenzoic acid. Meanwhile, other 23 test chemicals were correctly predicted among 27 compared chemicals. In addition, the sensitivity, specificity and accuracy of 31 test chemicals of the current study were compared with the OECD-adopted methods. As shown in Supplementary Table 4, the highest predictivity was obtained with KeratinoSens assay with sensitivity, specificity and accuracy of 89.5%, 90.0%, and 89.7%, respectively. In addition, comparable predictive capacity, such as sensitivity of 88.9%, specificity of 88.9% and accuracy of 88.9%, were obtained when applying the 2 out of 3 WoE approach with 27 chemicals overlapped. In comparison to other methods, the least accuracy was obtained with DPRA, where several chemicals, such as 3-aminophenol, methyl eugenol, farnesol, aniline, geraniol, and salicylic acid, were incorrectly classified. In the current study, the accuracy was 80.6% and 87.1% when the S-9 fraction and the microsome were used, respectively. In addition, the sensitivity of our current method with microsomes would be the highest among other methods compared. This confirmed that our newly developed method incorporated with induced liver microsomes was as comparable as to the other developed methods. Moreover, the current method would be very advantageous, not only because no mammalian cell or tissue cultures are required for the test, but also because the concentration tested might be lower than the existing methods (Clouet et al., 2017; Urbisch et al., 2016). Although reasonably high predictivities were achieved in the previous study without a metabolic activation system, the determination of some pre- or pro-haptens were challenging. To overcome the problem, in the present study, we developed a more advanced method in which the chemicals were activated by induced liver microsomes for conversion into their reactive forms to cause skin sensitization. As shown in Table 4, high levels of sensitivity, specificity, and accuracy of 95%, 72.7%, and 87.1%, respectively, were achieved in the combination model with microsomes. The currently developed method can be considered as one of the best methods to determine chemicals as skin sensitizers and non-sensitizers, even though neither mammalian cell cultures nor animals are required in the classification (Fig. 6). Because of the existence of complex molecular events of skin sensitization in animal models, it would be very difficult to mimic the complete in vivo scenario of skin sensitization. However, the current new in vitro method provided a high prediction capacity in cost-effective and robust manners. Further studies using other strains of bacteria are in progress to verify to what extent the discrepancies in bacterial strain would cause the differences in the results and to further optimize the method to be simpler. In addition, the applicability of current method to detect metallic skin sensitizers are also currently under investigation.