Collagen Peptides in Urine: A New Promising Biomarker for the Detection of Colorectal Liver Metastases

Introduction: For both patients and the outpatient clinic the frequent follow-up visits after a resection of colorectal cancer (CRC) are time consuming and due to large patient numbers expensive. Therefore it is important to develop an effective non-invasive test for the detection of colorectal liver metastasis (CRLM) which could be used outside the hospital. The urine proteome is known to provide detailed information for monitoring changes in the physiology of humans. Urine collection is non-invasive and urine naturally occurring peptides (NOPs) have the advantage of being easily accessible without labour-intensive sample preparation. These advantages make it potentially useful for a quick and reliable application in clinical settings. In this study, we will focus on the identification and validation of urine NOPs to discriminate patients with CRLM from healthy controls. Materials and Methods: Urine samples were collected from 24 patients with CRLM and 25 healthy controls. In the first part of the study, samples were measured with a nano liquid chromatography (LC) system (Thermo Fisher Scientific, Germaring, Germany) coupled on-line to a hybrid linear ion trap/Orbitrap mass spectrometer (LTQ-Orbitrap-XL, Thermo Fisher Scientific, Bremen, Germany). A discovery set was used to construct the model and consecutively the validation set, being independent from the dis Results: Seven peptides were selected and applied in a discriminant analysis a sensitivity of 84.6% and a specificity of 92.3% were established (Canonical correlation: 0.797, Eigenvalue: 1.744, F: 4.49, p: 0.005). The peptides AGPP(-OH)GEAGKP(-OH)GEQGVP(-OH)GDLGA P(-OH)GP and KGNSGEP(-OH)GAPGSKGDTGAKGEP(-OH)GPVG were selected for further quantitative analysis which showed a sensitivity of 88% and a specificity of 88%. Conclusion: Urine proteomic analysis revealed two very promising peptides, both part from collagen type 1, AGPP(OH) GEAGKP(-OH) GEQGVP(-OH) GDLGAP(-OH)GP and KGNSGEP(-OH)GAPGSKGDTGAKGEP(-OH)GPVG which could detect CRLM in a non-invasive manner.