Serum YKL-40 in risk assessment for colorectal cancer. A population based, prospective study of 4,987 subjects at risk of colorectal cancer

Abstract

4136 Background: Early detection of colorectal cancer (CRC) translates into improved survival. YKL-40 is produced by CRC cells and tumor associated macrophages. YKL-40 may play a role in cancer cell proliferation and differention, metastasis potential and angiogenesis. High serum YKL-40 is a potential biomarker of poor prognosis in patients with cancer, including CRC, and is independent of CEA and LDH. Highest serum YKL-40 is found in patients with metastatic cancer. We tested the hypothesis that serum YKL-40 would improve detection of CRC. Methods: Serum YKL-40 was determined by ELISA (Quidel) in a prospective, population based study of 4,987 subjects (age 18–97 years, 2,284 males and 2,704 females) referred to endoscopy (sigmoideoscopy n=1,720, colonoscopy n=3,206, rectoscopy n=53, unknown n=9) due to symptoms or risk factors for CRC. The subjects had a blood sample collected at the time of endoscopy (Jan 2004 - Dec 2005). Baseline variables including life style variables, current medication and co-morbidity including malignant disease (registered using ICD-10 codes) within the last 5 years. The measurements of serum YKL-40 were performed blinded in a randomized order at the completion of the study. Results: 189 colon cancers (3.8%), 114 rectal cancers (2.3%), 921 adenomas (18.5%), 10 other cancers (0.2%), and 1,217 other findings (24.4%) were found. Co-morbidity was observed in 2,624 subjects. Univariate analysis showed that high serum YKL-40 predicted CRC (OR=1.6, 95% CI: 1.4–1.7, p<0.0001, AUC=0.69 (AUC is the area under the ROC curve)). Restricting the analysis to subjects without other findings or co-morbidity demonstrated increased discrimination of serum YKL-40 (OR=2.5, 95% CI: 2.2–2.9, p<0.0001, AUC=0.79) Multivariate logistic regression analysis including serum YKL-40, age, sex, BMI, smoking, alcohol intake and co-morbidity demonstrated that serum YKL-40 independently predicted CRC (OR=1.56, 95% CI: 1.34–1.81, p<0.0001, AUC=0.75). Restricting the analysis to subjects with no co-morbidity showed that serum YKL-40 was significant (OR=1.95, 95% CI:1.53–2.48, p<0.0001, AUC=0.80). Conclusion: These results suggest that serum YKL-40 could be useful in the assessment of risk for colorectal cancer. No significant financial relationships to disclose.