Predicting tumour response to Mitomycin C (MMC) in experimental tumour models and human superficial bladder cancer patients.
aCancer Research Unit, Tom Connors Cancer Research Centre, Bradford, United Kingdom, bSchool of Health Studies, University of Bradford, Bradford, United Kingdom, cDepartment of Urology, St Lukes Hospital, Bradford, United Kingdom
AIMS: The bioreductive drug Mitomycin C (MMC) is clinically used to treat a variety of human cancers although response to the drug is highly heterogenous. Assays that can forecast the response of tumours to MMC would enable clinicians to tailor chemotherapy to individual patients. MMC is enzymatically reduced to cytotoxic species and the main aim of this study was to determine whether analysis of tumour enzymology (focusing on the enzyme NQO1, NAD(P)H:quinone oxidoreductase) could predict the response of superficial bladder cancers to MMC. METHODS: The study design involved a retrospective analysis of NQO1 using paraffin embedded formalin fixed tissues from patients with superficial bladder cancers that had been treated with intravesical MMC. NQO1 protein levels were determined by immunohistochemistry (IHC) and NQO1 genotype (a polymorphic C609T variant exists which lacks enzyme activity) was determined by PCR-RFLP. Time to first recurrence was used as the clinical endpoint for assessing response to MMC treatment. In addition, MMC metabolism by tumour and cell line homogenates was determined by HPLC and analysis of DNA damage induced in cell lines (with a broad spectrum of sensitivity to MMC) was assessed using the comet assay. RESULTS: A total of 160 patients with superficial bladder cancers (pTa and pT1) were entered into the study of which so far 135 patients have been genotyped. Times to first recurrence following intravesical MMC therapy ranged from 2 months to > 96 months. NQO1 protein was predominantly confined to tumour cells and wide inter-patient heterogeneity in NQO1 protein levels exists. Genotyping of NQO1 produced the following results: wild type (n = 81), heterozygotes (n = 51) and homozygous mutant variants (n = 3). Kaplan-Meier analysis demonstrated that a poor correlation between clinical outcome and both NQO1 protein levels and genotype status existed. In tumour homogenates derived from fresh, surgically resected superficial bladder cancers, rates of MMC metabolism varied considerably between individual patients (T½ ranging from 20 min to > 250 min). MMC metabolism by cell homogenates also varied considerably although no correlation between metabolic rates and chemosensitivity was observed. Analysis of DNA damage in vitro suggests that the extent of DNA damage and kinetics provides a good correlation with chemosensitivity. CONCLUSIONS: In superficial bladder cancers treated with intravesical MMC, clinical response cannot be forecast on the basis of NQO1 protein levels or genotype status. Inherent differences in the rate at which tumours metabolise MMC exist although the implications of these results in terms of predicting clinical outcome need further investigation. In vitro studies suggest that MMC metabolism may not only reflect drug activation pathways but also include detoxification pathways and that the extent of DNA damage significantly influences chemosensitivity in vitro.
Paper presented at the International Symposium on Predictive Oncology and Intervention Strategies; Nice, France; February 7 - 10, 2004; in oral session 895 (Chemotherapy response).