Identification of a collagen marker in urine improves the detection of colorectal liver metastases

Previously, we reported that a combination of a collagen alpha-1(I) natural occurring peptide (NOP) AGPP(-OH)GEAGKP(-OH)GEQGVP(-OH)GDLGAP(-OH)GP (AGP) and serum carcinoembryonic antigen (CEA) has the potential to detect colorectal liver metastases (CRLM). The combined method needs to be adapted to increase the sensitivity and specificity before it can be implemented in the clinic. This mass spectrometry study aimed to identify additional collagen NOPs in urine to further increase the sensitivity and specificity of our previously published method and search for the most discriminating NOP panel. The new assay was developed based on urine samples from 100 healthy controls and 100 patients suffering from CRLM. Two additional NOPs were identified: GPPGEAGK(-OH)P(-OH)GEQGVP(-OH)GDLGAP(-OH)GP (GPP) originating from collagen alpha-1(I) and GNDGARGSDGQPGPP(-OH)GP(-OH)P(-OH)GTAGFP(-OH)GSP(-OH)GAK(-OH)GEVGP (GND) originating from collagen alpha-1(III). A molecular model combining NOPs (AGP, GPP, and GND) and CEA was generated. Molecules that did not contribute to the diagnostic power of the model were removed, resulting in a model only consisting of GND and CEA. In this model, 86% sensitivity and 84% specificity in the discovery set, and 92% sensitivity and 90% specificity in the validation set were reached. These percentages are significantly higher than the current model based on AGP and CEA (p=0.032). The performance of the new model is better than the currently used techniques in the clinic (e.g. CT-scan, ultrasound, serum CEA), which have a sensitivity between 57-70% and a specificity between 90-96%.