Modified nucleosides are formed post-transcriptionally in RNA. In cancer disease, the cell turnover and thus RNA metabolism is increased, yielding higher concentrations of excreted modified nucleosides. In the presented study, urinary ribonucleosides were used to differentiate between breast cancer patients and healthy volunteers. The nucleosides were extracted from urine samples using affinity chromatography and subsequently analyzed via liquid chromatography ion trap mass spectrometry (LC-IT-MS). The peak areas were related to the internal standard isoguanosine and to the urinary creatinine level. For bioinformatic pattern recognition we used the support vector machine. We examined 113 urine samples from breast cancer patients (stage Tis-T4) and 99 control samples from healthy volunteers. We achieved a sensitivity of 87.67% and a specificity of 89.90% when including 31 nucleosides. The medical metabonomics concept based on the urinary nucleoside profile reveals a significantly improved classification compared with currently applied breast cancer biomarkers such as CA15-3.