Neurotoxicity with polyneuropathy as its predominant clinical sign is the major dose limiting side effect of chemotherapy with DNA reactive drugs such as cisplatin. The underlying mechanisms, however, are still not clear. Platinum compounds are known to exert their antineoplastic activity by forming distinct Pt-DNA adducts. Both the DNA repair rates and the extent of tolerance to persisting lesions determine the cell type specific sensitivity to cisplatin. In this report, DNA damage and repair are investigated with regard to their contribution to cisplatin-induced neurotoxicity in a mouse model. As the nucleotide excision repair (NER) pathway is considered to be an important mechanism for the processing of Pt-DNA lesions, its role was studied in an experimental setting with NER-deficient mice lacking functional XPA or XPC proteins. Employing monoclonal antibodies, structurally defined DNA lesions were quantified in various cells of the central and peripheral nervous system in wild type (WT) and repair deficient mice at different time points after a single exposure to cisplatin. In order to clarify, if accumulation of DNA lesions correlates with functional impairment, adduct quantification and electrophysiological studies were assessed in WT and XPA-/- mice under chronic cisplatin administration. The accumulation of Pt-DNA-adducts was dose dependent and two-fold higher in dorsal root ganglion (DRG) neurons and satellite cells (outside the blood-brain barrier) than in the same cell types of the spinal cord. Furthermore satellite and glial cells showed a higher adduct burden than corresponding neurons. Pt-DNA lesions peaked in all cell types of nervous tissue 24 hours after treatment. In the cells of DNA repair proficient WT mice adducts were removed by DNA repair within 7 days to a level, which represented 30% of the initial burden. Both NER deficient strains showed increased Pt-DNA adduct accumulation due to the insufficient repair of early monovalent adducts, which are known to occur as intermediate structures during cisplatin-DNA interaction. As a consequence, a massive loss of DRG cells was detected 48h after cisplatin application. While in the absence of active XPA protein some reduction of DNA lesions still could be observed, the loss of XPC function resulted in a complete lack of repair and in long term persistence of high adduct levels. Under chronic treatment, XPA-/- mice showed a significant increase in persisting DNA lesions compared to the WT counter points due to an impaired adduct removal in these cells. Furthermore, the enhanced persistence of Pt-DNA adducts in DRG cells of DNA repair deficient mice was strongly associated with an early functional impairment of sensory nerves: both, the amplitude of H-reflex and sensory nerve conduction velocity decreased significantly, whereas NER competent WT mice, bearing 35% less platinum-DNA lesions in their DRG cells, remained electrophysiologically intact under the same cumulative doses. Thus, there was a strong correlation of Pt-DNA adduct levels in DRG neurons with the extent of their morphological damage and with electrophysiological changes in sensory nerve fibers. The obtained data suggest that accumulation of DNA adducts is the crucial factor in the development of cisplatin-induced sensory polyneuropathy. Elucidation of these pharmacokinetic and pharmacodynamic processes is the basis for the development of neuroprotective substances which should prevent patients from the therapy-induced neuronal damage without blocking anti-neoplastic effects of cisplatin in malignant cells.