Comet assay measures of cisplatin and mitomycin C-induced DNA crosslink formation correlate bladder cancer cell chemosensitivity in vitro

1545 Systemic chemotherapy is used extensively in cancer therapy, but suffers from limited response rates in many tumor types. Moreover, it is frequently associated with significant morbidity and mortality. Therefore, where alternative treatment options exist (for example, radiotherapy or surgery), it is desirable to reserve chemotherapy for patients whose tumours will respond. To facilitate this, attention is turning to the development of techniques that provide predictive information about tumor chemosensitivity, as a means of enhancing patient selection for treatment with chemotherapy. The alkaline comet assay (ACA) is a technique for assessing DNA damage formation and repair, at the level of individual cells. Recently, we have demonstrated ACA to be predictive of cell radiosensitivity for a panel of human bladder cancer cell lines, with a greater comet response and a poorer extent of damage repair being noted in the radiosensitive cell lines (AL Moneef et al. British J. Cancer (2003) in press ). ACA is ideal as a clinical test; it is simple, rapid, inexpensive and requires low cell numbers. In the present study we demonstrate that ACA also predicts chemosensitivity in bladder cancer cell lines exposed to two commonly used DNA crosslinking chemotherapeutic agents. Using the same panel of bladder cancer cell lines, levels of cisplatin and mitomycin C (MMC)-induced DNA crosslink damage, as assessed using a modified version of ACA, were compared to measures of cell survival (clonogenic) and cell viability (MTT). For both agents, clear dose response curves were obtained, with the more chemosensitive cell lines displaying higher levels of crosslink formation and the most chemoresistant cell lines lower levels. In particular, there were high degrees of correlation between measures of cell viability and crosslink formation (R 2 >0.91), including an anomalous differential behaviour noted for T24 cells treated with MMC. Interestingly, the most chemosensitive cells tended to be the most radiosensitive and vice versa . This suggests that for these cell lines the same factor(s) may be responsible for radiosensitivity and chemosensitivity, both of which are predicted by ACA. These preliminary studies demonstrate that ACA can predict cancer cell chemosensitivity in vitro . To determine whether this technique can be applied to identify patients with chemosensitive tumors, clinical studies using cells derived from tumour biopsies need to be undertaken. If ACA is demonstrated to be predictive of chemosensitivity/outcome in vivo , then patients with sensitive tumors could be considered suitable candidates for treatment with chemotherapy. Alternatively, patients with chemo-resistant disease could be offered alternative intervention, for example, surgery.