Low Serum Levels of miR-101 Are Associated with Poor Prognosis of Colorectal Cancer Patients After Curative Resection.

BACKGROUND Recent studies showed low expression of microRNA (miR)-101 in various malignancies. However, the association of serum miR-101 and colorectal cancer (CRC) remains unknown. We investigated diagnostic and prognostic significance of serum miR-101 in CRC. MATERIAL AND METHODS A total of 263 consecutive CRC patients and 126 healthy controls were enrolled in this study. Serum miR-101 levels were measured using real-time quantitative reverse transcription polymerase chain reactions. The association between serum miR-101 level and survival outcome was analyzed. RESULTS Serum miR-101 in CRC patients was significantly lower than in healthy volunteers (P<0.001). Low serum miR-101 level was significantly associated with advanced cancer stage. Moreover, survival analysis demonstrated that patients with a low serum miR-101 had poorer 5-year overall survival than patients with a high serum miR-101 level (p=0.041). Serum miR-101 level also were confirmed as an independent risk factor for CRC in multivariate analysis (hazard ratio, 1.468; 95%CI, 0.981-1.976; p<0.001). CONCLUSIONS Serum miR-101 level was significantly downregulated in CRC patients and was closely correlated with poor clinical outcome, suggesting that serum miR-101 might be a useful diagnostic and prognostic marker for CRC.

Background

Colorectal cancer (CRC) is one of the most common and lethal gastrointestinal cancer in China and worldwide [1,2]. Although recent advances in diagnostic and therapeutic techniques have improved early detection and decreased mortality over the past decades, the prognosis of CRC patients remains poor, especially for advanced-stage cancer [3-5]. It has been revealed that metastasis and local relapse are the main causes of unsatisfactory long-term prognosis for CRC patients [6,7]. Furthermore, in clinical settings, CRC patients have quiet heterogeneous prognoses and chemotherapy responses, and an effective method is needed for the clinical risk stratification of patients [8]. TNM and histological stage are well established methods for the risk stratification of CRC patients and the administration of adjuvant treatments, but these biomarkers cannot be obtained until completion of postoperative carcinoma tissue histological evaluation, which are not conducive to preoperative neo-adjuvant treatments [9-11]. Therefore, more feasible and effective biomarkers that can be obtained before treatments are needed for predicting CRC prognosis and risk stratification.

MicroRNAs (miRNAs) are a subset of small, non-coding RNAs consisting of approximately 18–22 nucleotides, which inhibit gene expression by specifically binding the 3′untranslated region (UTR) of their target messenger RNAs (mRNAs), thus resulting in translation suppression of specific protein-coding genes [12]. It has been demonstrated that dysregulation of miRNAs plays crucial roles in the development of various malignancies [13-18]. Emerging evidence suggests that remarkably stable miRNAs can be detected in plasma [19-22]. Serum miRNAs can bind to specific proteins or be packaged into apoptotic bodies or exosomes, and thus are resistant to endogenous ribonuclease activity [23,24] and various serum miRNAs expressions have been confirmed as valuable biomarkers for cancer [25-27].

MiRNA-101 is a type of tumor-suppressor miRNA that targets several oncogenes and is strongly downregulated in various cancers [28-30]. The abnormal expression of miR-101 not only has diagnostic significance but also can serve as a prognostic predictor for cancer patients [31]. miR-101 in cancer tissues has previously been associated with cancer-relevant biological processes and clinical outcome in CRC [32,33]. However, there are no reports concerning the significance of circulating miRNA-101 for CRC patients. Therefore, we conducted this study to evaluate serum level of miRNA-101 in CRC patients and to analyze its prognostic significance.

Material and Methods

Patients

Our study was conducted according to the relevant global and local guidelines and regulations. Written informed consent was obtained from all subjects. Our study was also approved by the Institutional Review Board of the First Affiliated Hospital of Xin-Xiang Medical University. A total of 263 primary colorectal cancer patients who underwent surgery at the Department of General Surgery in the First Affiliated Hospital of Xin-Xiang Medical University between June 1, 2012 and June 1, 2017 were enrolled in this retrospective study. A number of cases were excluded from study for the following reasons: histology other than adenocarcinoma, preoperative acute and severe comorbidity, distant metastasis at the time of surgery, preoperative neo-adjuvant chemotherapy, clinical and histopathological data not obtainable, and life expectancy less than 24 weeks. All patients underwent R0 resection and postoperative adjuvant radiotherapy and/or chemotherapy, and all of the tumor specimens were pathologically evaluated as colorectal adenocarcinoma. A control group consisted of 126 age-matched healthy volunteers with no history of cancer and in good health based on self-report.

Clinical and pathological data of all patients were collected from the hospital records by one surgeon and check by another surgeon, including gender, age, tumor site, tumor size, tumor invasion depth, lymph node involvement, TNM stage, pathological differentiation, and operation records. Tumor stages were evaluated based on the Union for International Cancer Control (UICC) classification system. All patients were followed up by clinical visiting or telephone calls at regular intervals. Clinical follow-up lasted from the date of surgery to either death or January 2018. Outcome was assessed as 5-year overall survival (OS) rate. OS was accurately defined as the duration from operation day to death.

Sample preparation and RNA isolation

Sterile peripheral venous blood (5 ml) was collected from each patient on the day before surgery, as well as from the healthy controls. Serum was extracted by centrifugation from blood samples, then transferred to RNase/DNase-free tubes and immediately stored at −80°C for further processing. Total RNA was extracted from serum by use of the mirVana™ miRNA Isolation Kit (Thermo-Fisher Scientific; Waltham, MA, USA) according to the manufacturer’s instructions. The RNA concentration was determined using a NanoDrop ND-1000 spectrophotometer (NanoDrop 85 ND-1000, Nanodrop Technologies, DE, USA).

Quantification of miRNA by qRT-PCR

Total RNA from study participants was used to reversely transcribe miRNAs to a strand cDNA using TaqMan MicroRNA assays (Applied Biosystems, Foster City, CA, USA). Amplifications were performed using a miScript SYBR Green PCR kit (Qiagen, Valencia, CA, USA) and qRT-PCR was run on an Applied Biosystems 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Exogenous cel-miR-39 (Qiagen, Valencia, CA, USA) was used as a control. Expression of serum miR-101 was quantitatively analyzed by the 2−ΔΔCT method relative to cel-miR-39. Each sample was analyzed in triplicate.

Statistical analysis

All statistical analyses in this study were performed using SPSS 20.0 (IBM, USA). The results were considered as statistically significant when P<0.05 (2-sided). Continuous variables are expressed as mean ±SD and categorical variable are represented by frequencies. The χ2 test or Fisher’s exact test was used to compare different of categorical variables between the 2 groups, while continuous variables were analyzed with the independent-samples t test or one-way ANOVA. Receiver operating characteristic curve (ROC) analysis was conducted to evaluate the feasibility of serum miRNA-101 levels as a diagnostic indictor for CRC detection. The cut-off value for the serum miRNA-101 levels was also evaluated by ROC analysis. Kaplan-Meier survival curves were constructed by log-rank test in univariate analysis. A multivariate Cox hazard regression model was used to confirm the independent prognostic factors for CRC.

Results

Serum miR-101 was elevated in patients with CRC

Serum miR-101 was assessed in all samples from 263 CRC patients and 126 healthy controls. Serum miR-101 levels in CRC patients were significantly lower than in the healthy volunteers (P<0.01) (Figure 1). Furthermore, ROC curve analysis showed that the optimal cut-off level that could detect cancer was 8.32 using a sensitivity of 68.0% and a specificity of 71.7% as optimal conditions. The area under the curve was 0.732 with a 95% confidence interval between 0.658 and 0.806, P<0.001 (Figure 2).

Figure 1

Serum miR-101 level in colorectal cancer patients and healthy controls. The serum miR-101 level of 128 colorectal cancer patients was significant lower than that of 126 age-matched healthy volunteers (P<0.001).

Figure 2

Receiver-operator characteristic curve for CRC detection. ROC analysis showed a maximum AUC of 0.73 for miR-101.

Serum miR-101 was correlated with clinicopathological characteristics of CRC patients

We analyzed the association between serum miR-101 levels and clinicopathological characteristics of CRC patients using the chi-squared test. The results showed that low serum miR-101 levels were significantly associated with advanced T stages (P<0.001) and TNM stage (P<0.001), but was not associated with sex, age, tumor site and size, and pathological differentiation, or N stage (Table 1).

Table 1

Correlation between serum miR-101 level and clinicpathologic characteristics of CRC patients.

Characteristic	Serum miR-101		p
	High (n=77)	Low (n=186)
Age (years)			0.425
≥60	36	77
<60	41	109
Gender			0.263
Male	54	117
Female	23	69
Tumor site			0.111
Colon	31	56
Rectum	46	130
Tumor size (cm)			0.529
≥5	23	63
<5	54	123
pT (TNM)			<0.001
T1+T2	59	82
T3+T4	18	104
pN (TNM)			0.404
N0	26	73
N1	51	113
Clinical stage (TNM)			<0.001
III	54	56
I+II	23	130
Pathological differentiation			0.369
Well/moderate	46	122
Poor	31	64

Prognostic significance of serum miR-101 level for CRC patients

The median follow-up period was 33.5 months (6.1–85.1 months). During the follow-up period, 55 (20.9%) patients died due to cancer-related causes and 208 patients survived. In univariate survival analysis, we found that patients with low expression of serum miR-101 had a significantly worse 5-year overall survival compared to those with high expression of this small molecule (p=0.041) (Table 2, Figure 3).

Table 2

The prognostic characteristics of CRC patients.

	Univariate			Multivariate
	n	5-year OS rate	p	HR	95% CI	p
Age (years)			0.881
≥60	113	68.3%
<60	150	73.1%
Gender			0.912
Male	171	72.3%
Female	92	70.7%
Tumor site			0.837
Colon	87	71.9%
Rectum	176	69.5%
Tumor size (cm)			0.182
≥5	86	68.5%
<5	177	72.8%
Tumor invasion depth			0.021
T1+T2	141	75.2%
T3+T4	122	68.5%
Lymph node involvement			0.001
N0	99	74.3%
N1	164	67.9%
Clinical stage			0.036	1.312	0.928–1.631	<0.001
I+II	163	75.6%
III	110	64.2%
Pathological differentiation			0.012	1.257	0.921–2.127	0.010
Well/moderate	168	74.6%
Poor	95	64.7%
Serum miR-101			0.025	1.468	0.981–1.976	<0.001
Low	186	67.8%
High	77	76.6%

OS – overall survival; CI – confidence interval; HR – hazard ratio; CRC – colorectal cancer; miR-101 – micro RNA-101.

Figure 3

Lower plasma miR-101 level was associated with worse prognosis for colorectal cancer. The prognostic analysis revealed that a low serum miR-101 level was significantly associated with a worse overall survival rate (P=0.041).

We performed multivariate analysis of age and sex of patients, tumor size, lymph node involvement, clinical stage, histological differentiation type, and preoperative miR-101 level in a Cox regression model to determine independent prognostic biomarkers for colorectal cancer patients. The results showed that a low serum miR-101 level (p<0.001; hazard ratio, 1.468; 95% confidence interval [CI]: 0.981–1.976), clinical stage (p<0.001; hazard ratio, 1.312; 95% confidence interval [CI]: 0.928–1.631), and pathological differentiation (p=0.01; hazard ratio, 1.257; 95% confidence interval [CI]: 0.921–2.127) predict poor outcome in CRC patients independent of TNM stage and pathological differentiation (Table 2).

Discussion

In the present study, we found that the serum level of the tumor-suppressor miR-101 was significantly downregulated in CRC patients compared with healthy controls. In addition, serum miR-101 was confirmed as a good indictor to discriminate CRC patients from healthy subjects. We also analyzed the potential role of serum miR-101 obtained prior to surgery as a candidate biomarker to predict postoperative prognosis of CRC, showing that low serum miR-101 level was significantly associated to poor survival of CRC patients. We also found that there was a significant correlation between low serum miR-101 levels and unfavorable clinicopathological characteristics in CRC patients. According to these results, we suggest that miR-101 can be used for diagnosis and optimal risk stratification of individual CRC patients, and serum miR-101 also can serve as a promising serum biomarker for postoperative survival of CRC patients.

Recently, several studies showed that miR-101 is widely expressed in various tissues and organs and is frequently downregulated in various cancers [34-36]. miR-101 has been confirmed as a tumor suppressor that can inhibit many critical oncogenes [37,38]. In CRC, deregulation of miR-101 expression promotes Wnt/β-catenin signaling pathway activation and increases malignancy in colon cancer cells [39], and miR-101 also inhibits CRC cells growth through down-regulating sphingosine kinase 1 (SphK1) [40]. In the present study, our results showed that miR-101 is highly downregulated and correlated with poor prognosis in CRC patients, and these results are consistent with the previous reports mentioned above.

However, the function, mechanism, and origin of circulating miR-101 in cancer patients have not yet been fully elucidated. Several mechanisms for the release of circulating miRNAs have been reported, including passive leakage from cells due to injury, chronic inflammation or necrosis, active secretion via membrane vesicles such as exosomes, and active secretion by complex formation with lipoproteins or RNA-binding proteins. Furthermore, circulating miRNAs secreted from cancer cells can induce tumorigenesis in recipient cells [21,41,42]. Kosaka et al. reported that tumor-suppressor miRNAs from normal epithelial cells can inhibit growth of cancer cells [43]. Zheng et al. reported that systemic delivery of lentivirus-mediated miR-101 can dramatically suppress the development and metastasis of HCC in animal experiments [44]. Imamura et al. reported that depletion of plasma miRNA-101 was related to tumor progression and poor outcomes of gastric cancer patients [28]. It was revealed that high levels of certain miRNAs were significantly correlated with lymph node and distant metastasis, and thus are associated with advanced clinical stage of cancer [45]. Our results are consistent with these previous studies. Therefore, serum miR-101 level could be a novel treatment target for CRC patients.

This study is the first to report that the miR-101, which is depleted in the serum of CRC patients, can serve as both a serum biomarker and a novel therapeutic target for CRC. However, there are several limitations in the present study, including its relatively small sample and its single-center, retrospective design. Therefore, larger-scale prospective studies with longer follow-up are required to validate these results. Furthermore, the underlying molecular mechanisms of miR-101 in colorectal cancer have not yet been fully defined. Further experiments are needed to elucidate the mechanisms of miR-101 in carcinogenesis.

Conclusions

Our study demonstrates that serum miR-101 levels were downregulated in CRC patients. Moreover, low serum miR-101 levels were positively correlated with poor prognosis of CRC, suggesting that miR-101 acts as a tumor-suppressor gene in CRC. Serum miR-101 might not only serve as a diagnostic and prognostic biomarker for operable CRC, but also as a potential novel treatment target.