Establishing a genotoxic predictive test of toxicities and response to irinotecan chemotherapy in metastatic colorectal cancer

Irinotecan is a topoisomerase I inhibitor prescribed to treat metastatic colorectal cancer. Its use is limited by the heterogeneity in its toxicities and clinical response which are currently unpredictable for the most commonly prescribed doses. Presented in this thesis are the design, conduct and mechanistic analysis of the first prospective clinical study performed to assess whether DNA damage induced in peripheral blood lymphocytes (PBLs), following irinotecan exposure, is a predictive biomarker of this drug’s effect. PBLs were isolated from patients before and after receiving irinotecan based chemotherapy. A predictive test of irinotecan effect may have improved the management in 11 of the 42 patients recruited who experienced severe toxicities and 7 who gained no clinical benefit. Irinotecan did not induce DNA damage that could be measured using the alkaline comet assay in PBLs in vivo. An ex vivo method was subsequently developed using mitogenic stimulation of PBLs prior to treatment with SN-38 (the active metabolite of irinotecan) to induce DNA damage that could then be detected. Results demonstrated the presence of a wide inter-individual variation in the DNA damage levels. Several factors including intrinsic assay variability, cell cycle disturbance, apoptosis and DNA repair were investigated and purported to account for these variations. DNA damage did not correlate with the presence of the UGT1A1*28 polymorphism (associated with slow glucuronidation of SN-38) nor with toxicities to treatment. Liquid chromatography-mass spectrometry analysis of metabolites extracted from PBLs, demonstrated that glucuronidation of SN-38 was not occurring ex vivo, thus providing a potential explanation for these absent associations. However, DNA damage was weakly associated with tumour response and significantly correlated with progression free survival (PFS). In conclusion, there was no evidence that this model predicted normal tissue toxicities to irinotecan treatment; but associations, warranting further investigation, of DNA damage with response and survival were demonstrated.