Colorectal cancer (CRC) is the most common neoplastic pathology in the developed world. It is the second in frequency in men after prostate cancer and the second in women after breast cancer. Survival at five years, if all stages are taken into account, is approximately 50%. That represents a mortality of 20 cases per 100,000 inhabitants per year. The five-year survival of early stages, a situation almost equivalent to cure, is approximately 90%, whereas in advanced or metastatic stages it is less than 10%. Therefore, early diagnosis plays a central role in the improvement of survival rates. The main goal of this project is to find new ways to detect the cancer in the very early stages. To do so, the sample collection must be as simple, economic and convenient as possible. Only in this manner, regular screening approaching 100% of adult population will take place. The approach we propose is the identification and validation of new biomarkers in urine that can be used to diagnose colorectal cancer (CRC) at a very early stage. The central hypothesis is that CRC associated metabolites in the urine of patients can be segregated from patients with polyps (some of them precursors of CRC) and control subjects, and that their levels are correlated with clinical diagnostics of a CRC stage. Urine is an easy collectible biofluid for its non-invasiveness, and it can provide early detection of cancers. New sample measurement protocols and techniques are enhancing both selectivity and sensitivity by orders of magnitude, such as SPME and Twisters or GC/LC-MS systems (GCxGX/LC-QTOF-MS). The more sensitivity and selectivity, the more confindent diagnosis could be made at earlier stages of the cancerous process leading to a better survivability rate. Moreover, our proposal includes a multi-cohort, multi-laboratory validation of the findings, therefore ensuring that any discovery can readily be applied to any population and instrumentation around de world.