WP9: Pre-validation of the testing strategy

 

WP9 results, download the entire summary: Results WP9.pdf

 

During the last half-year of the project, candidate assays for the in vitro testing strategy will be challenged with 33 additional compounds under blind conditions. The results obtained will be used retrospectively to validate the preliminary testing strategy.

 

With the aim to select the best combination of tests for the testing strategy, PLS analysis on all data generated in the different in vitro models were performed (see Results WP3). This analysis identified the best combinations of in vitro tests that give a relatively good correlation with in vivo (rat and human) data, however it showed also that these combinations allow only a slight improvement of these correlations as compared to the validated in vitro cytotoxicity Neutral Red Uptake Assay alone. Moreover, the results obtained and the analysis performed did not enable to select the most promising methods that will be part of the testing strategy.

 

Based on these results it was concluded that further data analysis and data mining was needed before the construction of the testing strategy could be initiated. An independent consultant, Annette Kopp-Schneider at The German Cancer Research Institute was subcontracted to carry out the statistical analysis.

 

As a result of these statistical analysis, candidate assays for the testing strategy were chosen according to their reproducibility and reliability and most importantly, according to their potential to classify chemicals into the five GHS (The Globally Harmonized System) classes.

 

The following assays have been selected as candidate assays for a testing strategy on the basis of the statistical analysis:

• Neutral Red Uptake in 3T3 mouse fibroblasts (general cytotoxicity)
• Cytokine release (IL-1, TNFa, IL-6) in human whole blood (immunotoxicity
• Gene expression (GFAP, NF, Hsp-32, MBP) in rat brain aggregates (neurotoxcity)
• Uridine and methionine uptake in rat brain aggregates (neurotoxcity) 
• CFU-GM assay (hematotoxicity)
• Cytomic panel (incl. endpoints for oxidative stress, Ca+ uptake, mitochondrial and plasma membrane potential) in A704, HepG2, SH-SY5Y cells
• MTT assay in rat hepatocytes (metabolism)

In addition, the inclusion of the following algorithms in the testing strategy will be considered:

• The estimation of the oral dose from the effective concentration observed in vitro (by including kinetic parameters such as Vd, protein binding, clearance, oral absorption).
• The estimation of compound passage through the BBB using neuronal networks (for neurotoxicity assays).
  

Probably not all of the tests listed above will be included in the final testing strategy. During the first part of 2010 thirty three compounds will be testing in the candidate assays (listed below) under blind Conditions and the results will be used retrospectively to validate the preliminary testing strategy.