A NEW SERUM BIOMARKER FOR EARLY DETECTION OF COLORECTAL CANCER 

The finding of a new serum biomarker associated with colorectal cancer and colonic polyps opens up the possibility that blood tests will be able to screen for these conditions in the future.

Serum miR-21, an oncogenic micro (mi)RNA, accurately identified up to 92% of patients with colorectal cancer (CRC) and up to 82% of patients with advanced colonic polyps, according to a study published in the June 19 issue of theJournal of the National Cancer Institute.

The identification of this biomarker is "highly encouraging." The high mortality rates associated with CRC are related to the late detection of the disease, which underscores "the need for improved early detection, prevention, risk assessment, and intervention," senior author Ajay Goel, PhD, director of epigenetics and cancer prevention at the Baylor Research Institute in Houston, Texas, said in a news release.

"This blood-based test could be transformative in how we screen patients for [CRC]; it would save lives and could result in major savings of healthcare dollars," said Michael Ramsay, MD, president of the Baylor Research Institute, in the release.

MicroRNAs are small RNA molecules that are easily detected in blood and many other bodily fluids. miR-21 and miR-31 are oncogenic miRNAs that negatively regulate tumor-suppressor genes, but no previous studies have assessed their role in the detection of CRC or advanced colonic polyps, which are associated with an elevated risk of developing CRC later in life.

Dr. Goel and colleagues screened for the expression of miR-21 and miR-31 in media from 2 CRC cell lines to examine whether they are secretory miRNAs, and then analyzed serum from 12 patients with CRC and 12 healthy control subjects.

They validated the expression of these candidate miRNAs in serum samples from an independent cohort of 186 CRC patients, 60 postoperative patients, 43 advanced adenoma patients, and 53 control subjects, and in tissue samples from 166 matched primary CRC tissues.

Kaplan–Meier analyses and Cox regression models were used to evaluate patient survival. All statistical tests were 2-sided.

Study Findings

Levels of miR-21 were significantly elevated in preoperative serum from patients with adenomas (P < .001) and CRCs (P < .001). In patients who had undergone curative surgery, levels of miR-21 in postoperative serum fell significantly (P < .001).

Receiver-operating characteristic analysis showed that serum miR-21 levels clearly differentiated patients with adenoma or CRC from control subjects. Area under the curve for adenoma was 0.813 (95% confidence interval [CI], 0.691 - 0.910) and for CRC was 0.919 (95% CI, 0.867 - 0.958).

For diagnosis of CRC, the sensitivity of miR-21 was 91.9% and the specificity was 81.1%. For differentiation of adenomas from healthy control subjects, sensitivity was 81.1% and specificity was 76.7%. Serum miR-21 levels could not be used to differentiate advanced adenomas from CRC.

There was a statistically significant association between elevated miR-21 expression in serum and tissue samples and tumor size, distant metastasis, and worse survival.

In addition, miR-21 was found to be an independent prognostic marker for CRC (hazard ratio, 4.12; 95% CI, 1.10 - 15.4; P = .03). For subjects with high levels of miR-21 expression, the odds ratio for CRC was 43.3 (95% CI, 17.53 - 107.13), which the researchers describe as "an extraordinary performance for a noninvasive biomarker."

The observed associations persisted even after multivariate analysis adjusted for age, sex, TNM stage, lymph node metastasis, carcinoembryonic antigen level, distant metastasis, and pathologic differentiation.

Study Limitations and Clinical Implications

The value of miR-21 as a diagnostic biomarker make this study unique, the researchers note. Previous research has highlighted miRNAs that are frequently overexpressed in cancer cells, but relatively few are detectable in the circulation. Furthermore, nearly 30% of the released miRNAs do not reflect cellular expression profiles that originate in specific cancer cells, which reduces their specificity and, hence, their diagnostic utility.

This study shows that miR-21 expression is a promising noninvasive biomarker for the early detection of not only CRC, but also for clinically meaningful adenomas, which are "a critical target lesion for any CRC screening strategy," the researchers note. "Another interesting finding of our study is that miR-21 expression in serum also serves as a prognostic biomarker for CRC."

The value of miR-21 as a biomarker is limited because it can be detected in the blood of patients with solid cancers other than CRC, including glioblastoma, pancreatic cancer, and breast cancer. However, this study showed a statistically significant correlation between tissue and serum levels of miR-21 across all stages of colorectal neoplasia. In addition, a drop in miR-21 levels after surgical removal of the colorectal tumors in some patients suggests that miR-21 has utility as a specific biomarker for CRC.

The researchers caution that the clinical materials examined in this study were exclusively from patients of Japanese origin. They therefore recommend that serum miR-21 expression levels be validated in diverse ethnic populations.

However, they hope this noninvasive screening and prognostic tool "can be incorporated into routine clinical practice in the not-so-distant future."

May Be a Keystone

MiR-21 might not be "just another brick in the wall but rather may be the keystone leading to a molecularly justified, miRNA-based biomarker era in [CRC]," write Heinz-Josef Lenz, MD, from the Norris Comprehensive Cancer Center, University of Southern California, in Los Angeles, and Sebastian Stintzing, MD, from the Comprehensive Cancer Center at Klinikum Grosshadern, Ludwig-Maximilians-University of Munich, Germany, in an accompanying editorial.

The editorialists declare miR-21 "an almost ideal candidate for a diagnostic tool in the diagnosis and treatment monitoring of CRC," and state that higher miR-21 tissue levels could be associated with reduced pathologic tumor response in patients treated with FOLFOX (5-fluorouracil, leucovorin, oxaliplatin) in a neoadjuvant setting for locally advanced CRC.

However, they note that all published miRNA data have been generated from retrospective analyses, and reaffirm the need for validation with prospective testing in independent control studies.

They also suggest the need to compare the value of miR-21 in the diagnosis and treatment monitoring in CRC patients with that of protein-based markers such as carcinoembryonic antigen, and to determine the prognostic or diagnostic value of other miRNA serum markers, such as miR-141, miR-601, and miR760.

"The exact mechanism of how miR-21 is involved in carcinogenesis and tumor progression remains unknown," the editorialists write. It is possible that "miR-21 serum levels are elevated in colon cancer as a sign of the inflammatory process of the adjacent healthy tissue." The mechanism of action in tumorigenesis "might even be a druggable target," they note.

This study was supported by the National Cancer Institute, National Institutes of Health, the Baylor Research Institute, and a pilot project grant from the Charles A. Sammons Cancer Center at the Baylor University Medical Center. The editorialists have disclosed no relevant financial relationships.

J Natl Cancer Inst. 2013;105: 840-841, 849-859. Abstract, Editorial