Systemic chemotherapy is extensively used in cancer therapy, however, for many treatments' response rates are limited. Furthermore, certain regimens are frequently associated with significant morbidity and occasional mortality. Consequently, when alternative options exist, it is desirable to reserve a particular chemotherapy for those patients whose tumours will respond. Therefore, attention is turning to the development of techniques that could provide predictive information regarding a tumour's particular chemosensitivity, as a means of enhancing patient selection for that specific treatment. One approach has been to focus on measures of DNA damage formation and repair as being potentially predictive of cancer cell chemosensitivity, the premise being that higher levels of induced DNA damage (resulting from the chemotherapeutic agents) and/or deficiencies in DNA damage repair are indicative of greater sensitivity. In the present study we have investigated the Comet assay response of a panel of non-small cell lung cancer cell lines towards cisplatin and found an inverse correlation between sensitivity and damage formation resulting from this agent. Moreover, an inverse correlation was found between resistance and extent of damage repair. Further analysis of multiple alternate cellular end-points (including cell cycle analysis, apoptosis and gene expression changes) revealed cisplatin damage tolerance to be a chemoresistance mechanism in this model system. This study highlights damage tolerance mechanisms as potentially confounding factors in attempts to develop predictive tests based on measures of genotoxicity. To address this we would argue that a range of multiple end-points should be analysed to ascertain the "complete predictive picture".