Colon cancer ranks among the most frequent malignancies and is the fourth leading cause of cancer-related death worldwide [1, 2]. Detection of colon cancer at early stages is critical for curative treatment interventions: although the 5-year disease-free survival for International Union Against Cancer (UICC) stage I tumors exceeds 90%, this rate is reduced to 63% in UICC III and < 5% in UICC IV carcinomas . Yet, despite the implementation of current screening programs about 50% of these malignancies are detected at advanced tumor stages. Therefore, tools and methodologies that allow early colon cancer detection directly impact on survival times. In present clinical practice, screening for cancer and premalignant polyps of the colon is based on clinical examination, the detection of fecal occult blood, and on sigmoidoscopy or colonoscopy [4, 5]. The successful implementation of these screening procedures has contributed to a reduction of disease-associated mortality of colon carcinomas . The persistent delay in diagnosis and the associated high mortality rate are attributable to a low compliance to some screening tests and to the low sensitivity of other tests . An optimal, alternative screening test would be relatively non-invasive and achieve high patient compliance, fulfil excellent analytical performance regarding sensitivity and specificity and still be robust and cost-effective. Such a test can be envisioned if changes in the composition of serum proteins could indicate specific diseases and/or disease stages. Comprehensive serum proteome profiling for tumor-specific markers has therefore become a field of intensive research. For colon cancer screening the application of serological testing has not been established so far, even though very promising candidate markers have been reported. Based on a thorough literature review, the most promising markers were identified. By using preoperative serum levels for CRC diagnosis with sensitivity and specificity of 90% was reported for protein 26 (CD26) . Other diagnostic approaches based on the detection of carcinoembryonic antigen (CEA), vascular endothelial growth factor (VEGF) , macrophage colony-stimulating factor (M-CSF) , Nicotinamide N-methyltransferase (NNMT) , or C-reactive protein (CRP) concentrations  reached either a high sensitivity or specificity. Furthermore, within our consortium, Interleukin 8 (IL-8) , Calgizzarin (S100A11)  and complement component 3a (C3adesArg)  serum levels were determined as potentially promising biomarkers for CRC . However, most of the markers are yet to be validated in well defined, large screening studies. Hereby, the measurement of a disease specific panel of markers could outperform the measurement of individual markers regarding sensitivity and specificity. In addition, this approach could provide a more comprehensive reflection of molecular networks and pathophysiological conditions of diseases.
Biochip array technology allows multiplex determination of multiple biomarkers from a single sample [16, 17]. This is also relevant when volumes of clinical samples are limited. Implementations of this technology in clinical settings have been reported for different biochip arrays including cytokines , cerebral and cardiac arrays [18, 19], adhesion molecules , and also detection of drug abuse .
The aim of this study was to apply biochip array technology to colon cancer screening. For this purpose, a biochip array was designed and developed for the multiplex determination of nine serum markers allowing for low inter-analysis variability, decreased workload and faster processing time as well as lower costs due to high-throughput automation. The performance of the two biochip arrays for colon cancer screening was then evaluated in a training and a validation set consisting of 317 highly standardized, liquid nitrogen preserved serum samples.