Abstract 3777: In silico rendering of cell cycle progression of erlotinib and gemcitabine treatment in pancreatic cancer cells

The association of erlotinib (ERL) with gemcitabine (GEM) is a promising combined therapy for advanced pancreatic cancer, but current protocols have only a modest impact on the overall survival of patients respect to GEM alone. Sequential treatment can improve the efficacy of this drug combination. Towards optimization of the drug combination, we simulated cell cycle progression during/after treatment, based on the dynamics of checkpoint activation, with the probabilities of cell arrest or death in each cell cycle phase, applying a previously established experimental/computational approach (Lupi et al. Cancer Res. 64: 2825; Ubezio et al Cancer Res. 69: 5234; Falcetta et al. PLoS Comput Biol. e1003293). The simulation gives the temporal evolution of cell flow through G1, S and G2M over subsequent division cycles, from which the time courses of cell cycle percentages and the number of cells in each cell generation were derived enabling a joint validation of the model with independent flow cytometric and time-lapse live-imaging experimental data, collected in time course experiments up to 96h with different concentrations of GEM (6h treatment) and ERL (48h treatment) with two pancreatic cell lines (BxPC-3 and capan-1). Analysis of the response to ERL treatment reveals that at cytostatic concentrations a partial G1 arrest mainly occurs after the first division following the start of treatment (gen 1 cells) while a stronger delay of the progression in S phase is active also before division (gen 0). The activity of G1 and S phase checkpoints were dose dependent and at higher cytotoxic concentrations we observed a near complete arrest in both G1 and S phases while most cell deaths were in phase S in gen 0. After treatment discontinuation, while G1 blocked cells re-enter in cycle, S phase progression in gen 0 and gen 1 remained slow but without additional death events. The model of GEM demonstrated the activation of a strong delay in phase S shortly after the start of treatment, both in gen 0 and gen 1, already at non-cytotoxic concentrations, lasting several hours after treatment discontinuation, and thereafter progressively reduced, resulting in a wave of broadly semisynchronised cells traversing S and G2M phases in gen 0. At cytotoxic concentrations, part of the cells remained arrested in S phase and eventually died. A temporary delay in G1 was also demonstrated already at fairly low concentrations, acting only during exposure to the drug. Based on these models of GEM and ERL treatment, studies are ongoing to simulate the expected outcomes of sequential treatments. Preliminary analysis suggests that the GEM = > ERL sequence is particularly favourable in both cell lines when a cytostatic ERL concentration was given after a fairly cytototoxic GEM concentration when S phase progression resumed, by strengthening the long term cytostatic effects in S phase. Citation Format: Monica Lupi, Francesca Falcetta, Maurizio D'Incalci, Paolo Ubezio. In silico rendering of cell cycle progression of erlotinib and gemcitabine treatment in pancreatic cancer cells. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 3777. doi:10.1158/1538-7445.AM2015-3777