In vitro activity profile of the novel hypoxia-activated cytotoxic prodrug TH-302

1441 TH-302 is a hypoxia-activated cytotoxic prodrug comprised of a 2-nitroimidazole hypoxia trigger linked to a dibromo isophosphoramide mustard toxin. TH-302 exhibits potent in vitro cytotoxicity under hypoxia, is efficacious in both ectopic and orthotopic xenograft cancer models, and demonstrates a favorable PK and tox profile. TH-302 is currently in a Phase 1 clinical trial for the treatment of cancer. Here we present details of TH-302’s in vitro activity profile.
 Examination of TH-302’s cytotoxic potency and hypoxia selectivity in different cancer cell lines (PC-3, MIA PaCa-2, ACHN, HCT 116) yielded results comparable to those observed in H460 cells (in a 2h exposure 3d proliferation assay 300). The cytoxicity of TH-302 under normoxia across the entire panel of 60 human cancer cell lines comprising the NCI-60 showed none to be more sensitive than H460 under normoxia (2d proliferation assay). This latter result indicates that, at least within this panel of cell lines, none are able to significantly activate TH-302 in a hypoxia-independent manner. When TH-302 was tested in pairs of cell lines that allow the profiling of susceptibility to efflux by the major drug resistance pumps (i.e. MDR1, MRP1, and BCRP), TH-302 showed no difference in potency or selectivity in lines over-expressing the specific pump, versus matched control lines (MESSA vs. MESSA-DX5; H69 vs. H69AR; HEK293-PCDNA vs. HEK293-BCRP, respectively). This result indicates that TH-302 is not a substrate for these clinically relevant drug resistance pumps.
 The oxygen concentration dependence of TH-302 activation cytotoxicity was examined employing a set of defined gas mixtures (0%, 0.1%, 0.3%, 0.6%, 1%, 10%, 21% O2) in the 2h exposure 3d proliferation assay with H460 cells. Results demonstrated an O2 concentration dependence of prodrug activation (IC 50’s were 0.1, 1, 3, 5, 7, 30, and 30µM, respectively).
 Examining downstream effects of TH-302 activity, we demonstrate both γH2AX phosphorylation and G2/M phase cell cycle arrest in a TH-302 concentration- and time-dependent manner in H460 cells. These activities mirrored the cytotoxicity profile in terms of potency and hypoxia-selectivity. Utilization of CHO cell-based DNA repair mutant cell lines showed that lines deficient in homologous recombination (irs1SF and UV41) exhibited greater sensitivity to TH-302 under hypoxia than the parental control line (AA8) or in lines deficient in base or nucleotide excision repair (EM9, UV35). These findings are consistent with a DNA cross-linking mechanism of action underlying TH-302’s cytotoxicity in vitro.
 Taken together, these results demonstrate that TH-302 exhibits a favorable in vitro activity profile for a hypoxia-activated prodrug in clinical development.