Background: Tumors are addicted to pathways neutralizing oxidative stress as a result of their high levels of reactive oxygen species (ROS). These ROS can oxidize nucleotides (NTs) within the precursor pool, which can cause cell death when incorporated into DNA. This tumor specific vulnerability might be exploited therapeutically. The mutT-homologue (MTH1) enzyme is recently identified as a non-oncogene target, which prevents incorporation of oxidized NTs into DNA (Gad et al, 2014; Huber et al, 2014). MTH1 inhibition by TH588 and (S)-Crizotinib caused tumor-specific oxidative DNA damage and cell death (Gad et al, 2014; Huber et al, 2014). Interestingly, hypoxia might enhance sensitivity to MTH1 inhibition (Qiu et al, 2015; Bräutigam et al, 2016). Thus, exploiting tumor oxidative stress via inhibition of MTH1 might be an exciting new approach to fight colorectal cancer (CRC) even in the presence of hypoxia.

Aims and Hypotheses: In the present study we aim to compare the ability of TH588 HCl and (S)-Crizotinib to inhibit CRC growth in vitro and in vivo and to assess their mechanism(s) of action, including MTH1-dependent and -independent effects. We hypothesize that both compounds have the potency to inhibit CRC growth, conceivably via different mechanisms of action, which might not dependent on MTH1 inhibition (Kettle et al, 2016).

Methods: All data were obtained using a human CRC liver metastasis-derived 3D spheroid line untreated or treated for 3 days with TH588 HCl or (S)-Crizotinib. Spheroids were pre-cultured and treated under normoxia (21% O2), hypoxia (0.1% O2) or only pre-cultured under hypoxia to mimic hypoxia and reoxygenation respectively.

Results: Using a CellTiter-Glo cell viability assay we showed that increasing concentrations of (S)-Crizotinib under normoxia reduced cell viability with 23%, 64%, and 98% for 5, 10 and 20 μM respectively. By contrast, the maximal reduction of cell viability after treatment with TH588 HCl was only 32%. Interestingly, the dose response curve of (S)-Crizotinib was steeper when spheroids were reoxygenated just before treatment compared to normoxia and hypoxia, which implies that (S)-Crizotinib is more efficient. Importantly, hypoxia completely abolished the TH588-induced reduction in cell viability seen during normoxia. While the immediate effects on cell viability were different, both drugs (10 μM) severely reduced clone-forming efficiency both under normoxia and after hypoxia-reoxygenation prior to treatment. These results suggest different mechanisms of action of both drugs. As expected, both drugs caused an increase in DNA double-strand breaks, as measured by γH2AX staining. While chemotherapy treatment usually enriches the content of cancer stem cells, so far we did not observe this with either MTH1 inhibitor.

Discussion and Conclusions: Until now, most research involving MTH1 inhibitors has been performed using cancer cell lines. In the present study we use patient-derived CRC spheroids and organoids. We previously found that organoids derived from patients with a poor-prognosis type of CRC are generally characterized by high expression of a H2O2 stress signature (Emmink et al, 2014). Exploiting this oxidative stress phenotype via inhibition of MTH1 might be an attractive new therapeutic avenue. Our results suggest that TH588 HCl and (S)-Crizotinib induce DNA damage and effectively reduce clone-forming potential of a CRC spheroid line, which was not limited by hypoxia pre-culturing. Interestingly, both compounds appear to act via different mechanisms. Further research should unravel these mechanisms of action.

Citation Format: Lizet M. van der Waals, Danielle A.E. Raats, Jennifer M.J. Jongen, Tobias B. Dansen, Inne H.M. Borel Rinkes, Onno W. Kranenburg. Exploiting oxidative stress using MTH1 inhibitors in colorectal cancer. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Targeting the Vulnerabilities of Cancer; May 16-19, 2016; Miami, FL. Philadelphia (PA): AACR; Clin Cancer Res 2017;23(1_Suppl):Abstract nr A29.