Pre-Clinical Testing of Chemotherapy Combinations in Retinoblastoma Cell Lines Using the Dimscan Cytotoxicity Assay

Abstract

Abstract: : Purpose: The CEV regimen, carboplatin (C), etoposide (E), and vincristine (V), is commonly used for treating retinoblastoma (RB). Cyclosporin A (CsA) has been added as an MDR-1 modulator to potentiate the clinical activity of CEV. We tested these agents, and also melphalan (M) and topotecan (T) on a panel of 10 RB cell lines in 20% and 2% O₂. Methods: The DIMSCAN cytotoxicity assay uses digital imaging microscopy and fluorescein diacetate/eosin Y to quantify viable cells in microplates. For pre-clinical testing of 2 to 4 anticancer drugs in various combinations, we have developed a high-throughput (3-4 log dynamic range) assay using 384-well microplates (DIMSCAN-384). 5/10 cell lines were established from untreated patients (CHLA-192, 215, 222, 223, 224) and 5 from patients at relapse after chemotherapy (chemo) or radiotherapy (rad) (CHLA-194, 203, 209, 210, 246). Cell lines with an LC₉₀ higher than the clinically achievable plasma concentration (C= 3 µg/ml, E = 5 µg/ml, V = 0.5 µg/ml, M = 10 µg/ml, T = 0.1 µg/ml) were considered resistant; drug levels in the eye, and especially in vitreous, are unknown. Results: Resistance to single agents was seen in 30% (C), 30% (E), 20% (V), 10% (M), 30% (T) of tested cell lines, with 6 of 10 cell lines sensitive to all drugs (3 post-chemo, 1 post-rad), and 2 (1 pre-therapy, 1 post-rad) sensitive to only 1 (M or V). Median LC₉₀ values were C = 1.2 µg/ml, E = 0.8 µg/ml, V = 10 ng/ml, M = 0.6 µg/ml, T = 6.9 ng/ml. Using DIMSCAN-384, we tested CEV + CsA, and M + T as single agents and in various combinations in 5 RB cell lines. In 4 of 5 lines M+T did not achieve synergistic cytotoxicity (P > 0.2). CEV was no more active than single agents in 4 of 5 lines (P > 0.15). Single agent CsA was inactive in 5 of 5 tested cell lines (median LC₉₀ = 59 µg/ml). CsA enhanced (P < 0.05) the cytotoxicity of E (in 2 lines) and V (in 3 lines), but not C (in 4/5 lines; P > 0.05). CsA enhanced CEV cytotoxicity (P < 0.05) in 3 of 5 cell lines, achieving an additional 1-2 logs of cell kill over CEV alone. CsA also enhanced the cytotoxicity of most 2 drug combinations (including those with C), achieving 0.5-2 logs of additional cell kill over combinations without CsA, and EV + CsA (in 5 of 5 lines), CV + CsA (4/5 lines) or CE + CsA (3/5 lines) were as effective as CEV + CsA (P > 0.05). Hypoxia (2% O₂) did not antagonize cytotoxicity of C, E, V, CsA, or any combination of these agents (P > 0.5). Conclusions: These data suggest that CEV may not be more active than single agents in many RB, and suggest that at drug levels achievable in plasma, CsA can significantly enhance the activity of CEV against some RB.

Keywords: retinoblastoma • clinical laboratory testing