Urinary Cysteine-Rich Protein 61 and Trefoil Factor 3 as Diagnostic Biomarkers for Colorectal Cancer

Takaya Shimura; Hiroyasu Iwasaki; Mika Kitagawa; Masahide Ebi; Tamaki Yamada; Tomonori Yamada; Takahito Katano; Hirotada Nisie; Yasuyuki Okamoto; Keiji Ozeki; Tsutomu Mizoshita; Hiromi Kataoka

doi:10.1016/j.tranon.2018.12.006

Urinary Cysteine-Rich Protein 61 and Trefoil Factor 3 as Diagnostic Biomarkers for Colorectal Cancer

Transl Oncol. 2019 Mar;12(3):539-544. doi: 10.1016/j.tranon.2018.12.006. Epub 2019 Jan 3.

Affiliations

¹ Department of Gastroenterology and Metabolism, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya 467-8601, Japan. Electronic address: tshimura@med.nagoya-cu.ac.jp.
² Department of Gastroenterology and Metabolism, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya 467-8601, Japan.
³ Department of Gastroenterology and Metabolism, Nagoya City University Graduate School of Medical Sciences, 1 Kawasumi, Mizuho-cho, Mizuho-ku, Nagoya 467-8601, Japan; Department of Gastroenterology, Aichi Medical University, 1-1 Karimata, Iwasaku, Nagakute 480-1195, Japan.
⁴ Okazaki Public Health Center, 1-3 Harusaki, Harisaki-cho, Okazaki 444-0827, Japan.
⁵ Department of Gastroenterology, Japanese Red Cross Nagoya Daini Hospital, 2-9 Myoken-cho, Showa-ku, Nagoya 466-0814, Japan.

Abstract

Since a fecal occult blood test for colorectal cancer (CRC) does not offer sufficient diagnostic power for CRC, novel non-invasive biomarkers are hopeful for CRC screening. We conducted the current study to discover non-invasive urinary biomarkers for diagnosing CRC. Among urine samples from 258 patients (CRC, n = 148; healthy controls, n = 110), a cohort of 176 patients composed of 88 patients with GC and 88 healthy controls was selected after age- and sex-matching using propensity score. This cohort was then randomly divided into 2 groups: 53 pairs (106 patients) in the training cohort, and 35 pairs (70 patients) in the validation cohort. No significant differences were found for baseline characteristics between the CRC and healthy control groups in both training and validation cohorts. On multivariate analysis in the training cohort, urinary levels of cysteine-rich protein 61 (uCyr61) and trefoil factor 3 (uTFF3) were identified as independent significant diagnostic markers for CRC. Moreover, uCyr61 alone and the combination of uCyr61 and uTFF3 allowed significant differentiation between healthy controls and CRC groups in the training set (uCyr61: area under the curve (AUC) = 0.745 [95% CI, 0.653-0.838]; uCyr61 + uTFF3: AUC = 0.753 [95% CI, 0.659-0.847]). In the validation cohort, uCyr61 and uTFF3 were significantly higher in the CRC group than in the healthy control group, and they also allowed significant differentiation between healthy control and CRC groups (uCyr61: AUC = 0.696 [95% CI, 0.571-0.822]; uTFF3: AUC = 0.639 [95% CI, 0.508-0.770]; uCyr61 + uTFF3: AUC = 0.720 [95% CI, 0.599-0.841]), as in the training cohort. A panel combining uCyr61 and uTFF3 offers a promising non-invasive biomarker for diagnosing CRC.