Table 3.

Evaluation of RTCA predictivity to detect genotoxicity with a set of 81 proprietary UCB compounds. HepG2 cells were exposed for at least 72 h to 81 proprietary UCB compounds belonging to two central nervous system projects (CNS1 and CNS2). The concentrations tested were 125, 250, 500 and 1,000 µM (unless solubility problems were encountered). The objective was to determine if the compounds that were identified as genotoxic by traditional genotoxicity in vitro assays generated RTCA genotoxic profiles. A compound was classified as genotoxic according to the impedance measurements when after the compound addition, the cell index generated by the compound was higher than the control curve followed by a decrease in cell index that reached at least 50% mortality within 48 h exposure. A compound was classified as negative when both conditions were not met. Examples of typical genotoxicity profiles are presented in Figure 7(A,B) (i.e., HepG2 and A549 cells exposed to etoposide, respectively). For more details about the definition of the terms (sensitivity, specificity and concordance/accuracy), please refer to the Materials and Methods section.

	CNS project 1	CNS project 2	CNS project 1 & 2
Number of compounds tested	35	46	81
Genotoxic compounds	22	7	29
Non-genotoxic compounds	13	39	52
Genotoxic signature (RTCA)	14	0	14
Non-genotoxic signature (RTCA)	12	38	50
Sensitivity	63.6% (14/22)	0% (0/7)	48.3% (14/29)
Specificity	92.3% (12/13)	97.4% (38/39)	96.2% (50/52)
Concordance (accuracy)	74.3% (14 + 12/35)	82.6% (0 + 38/46)	79.0% (14 + 50/81)