To evaluate the sensitivity and specificity of the test kit "COLOSAFE' in Hong Kong

Colorectal cancer (CRC) remains to be a leading cause of cancer mortality [1]. It induces a substantial financial burden in terms of healthcare utilization and quality-adjusted life years (QALY) lost [2]. Fecal Occult Blood Tests (FOBT) and colonoscopy were proven effective in reducing CRC mortality by 33% and 68%, respectively [3-5]. Both tests have been proposed as primary screening modalities for asymptomatic subjects by international guidelines and Asia Pacific consensus statements [6-8]. However, colonoscopy has a low compliance rate at approximately 20% in Chinese populations, and has been perceived by screening participants as invasive, expensive and inconvenient due to the need for bowel preparation [9, 10]. In addition, the accuracy of FOBT is limited and its adherence declined sharply over time [11, 12]. Hence, the recent decade witnessed a rapid development of non-invasive biomarkers to detect CRC.

Faecal-DNA test is a novel screening test for CRC, using molecular techniques to identify CRC-relevant biomarkers in stool. One of its toolkits, branded 'ColoGuard' (sDNA, Exact Sciences), was first approved by the FDA in 2014 for its application in clinical practice [13]. It consists of quantitative molecular assays for KRAS mutations, aberrant NDRG4 and BMP3 methylation, and β-actin, and a hemoglobin immunoassay [13], which has been widely promoted in US. In 2016, it was endorsed by the United States Preventive Service Task Force (USPSTF) as one of the recommended screening tests [14].

Recently, a test kit named 'COLOSAFE' (or namedChang An Xin) has been developed as a faecal-DNA product targeted to improve diagnostic accuracy of CRC screening [15]. It is a stool test of methylated SDC2 for detection of CRC. This stool test of methylated SDC2 detected 81.1% of CRC and 58.2% of adenomas at a specificity of 93.3%. SDC2 is also named fibroglycan, encoding a type I trans-membrane heparansulfate proteoglycan. In certain cancers, hypermethylation of SDC2 had been reported. Recent studies showed that methylated SDC2 was detected at high frequency in blood from CRC patients [16, 17]. As tumor cells are exfoliated into the gut lumen earlier than vascular invasion when CRC develops [18], faecal sample could represent a more suitable specimen than plasma for detection of early cancer.

A recent study involving 1,200 subjects from 2015 to 2017 by the Sixth Affiliated Hospital of Sun Yat-sen University, Nanfang Hospital of Southern Medical University and Shandong Cancer Hospital showed that the 'COLOSAFE' test kit attained a sensitivity of 85% while maintaining a specificity of 98% (unpublished data). Because of its novelty, the 'COLOSAFEtest kit was recognized as a 'creative medical apparatus' by the National CFDA in March 2017. In June 2017, the project 'Screening and intervention research of Chinese CRC' (2017YFC 1308800) was launched with national key project for non-communicable disease control. Nevertheless, there are limitations with the design of the original study [15]. For instance, it involved a small, single-centre that recruited a homogeneous sample of individuals only. An important knowledge gap exists and additional clinical trials are required to further validate its diagnostic accuracy in other populations.

Inclusion Criteria:

1. CRC patients, individuals with normal colonoscopy findings, or patients with other colorectal pathologies;
2. Chinese subjects aged 45 to 79 years living in Hong Kong for most of the time in the past 5 years;
3. Having received colonoscopy within 8 days to 3 months before enrolment in the study);
4. Absence of medical conditions that render the subject unable to comprehend the study;
5. Willing to provide informed consent;
6. Willing to provide faecal sample for laboratory investigation in the present study.

Exclusion Criteria:

1. History of major chronic illnesses which are serious (including those with respiratory, cardiac, renal, liver or immune function impairment or requiring long-term immunosuppressive therapy);
2. Previous history of inflammatory bowel diseases or digestive cancers other than CRC;
3. History of colorectal resection for any reason other than sigmoid diverticula;
4. Had a surgery or an invasive procedure within the past 3 months;
5. Received chemotherapy or radiotherapy with the past 3 months;
6. Had participated in any interventional clinical study within the previous 30 days.

A total of 400 individuals aged 45-79 years will be consecutively recruited for this study. Among them, we aim to recruit 150 subjects with known diagnosis of CRC; 150 subjects with normal colonoscopy findings; and 100 subjects whose colonoscopies showed other pathologies (e.g. inflammatory polyps, adenomas, diverticular diseases). To enhance comparability of the cases and the controls, we aim to recruit these subjects in the same hospital (medical or surgical clinics of the Prince of Wales Hospital). Stool samples will be collected within 8 days to 3 months after the individuals received colonoscopy, but before any treatments, including surgery, chemotherapy, and radiotherapy. We aim to match cases and controls by their age and gender.