Alteration of miR-21 and miR-24 expression: biomarker for early detection of synchronous metastases in colorectal cancer: a cross-sectional study in Indonesia
Article information
Abstract
Background/Aims
Timely detection with highly accurate biomarkers would be helpful in effectively managing colorectal cancer (CRC). We aim to investigate the accuracy of 3 emerging biomarkers—miR-21, miR-24, and miR-145—in detecting synchronous metastases in CRC.
Methods
We recruited newly diagnosed CRC patients with extensive investigations to determine cancer staging and metastatic status. The expression levels of miR-21, miR-24, and miR-145 in tumor biopsy were measured using reverse transcription quantitative polymerase chain reaction. Multivariate and receiver operating characteristic analyses were conducted to evaluate the association and performance of these miRNAs in identifying various metastases.
Results
Out of the 63 Indonesian patients involved, 37 (58.7%) were diagnosed with localized CRC, whereas the remaining 26 (41.3%) were identified as having metastases: 31.7% liver, 14.3% lung, 3.2% bone, and 4.8% other metastases. There was a significant downregulation of miR-24 expression in metastatic CRC patients compared to those without metastases (0.024 [4.680] vs. 12.900 [42.376], P< 0.01). Overexpression of miR-21 was identified as an independent risk factor of synchronous metastasis (odds ratio [OR], 1.016; 95% confidence interval [CI], 1.003–1.030; P< 0.05), particularly lung (OR, 1.011; 95% CI, 1.002–1.020; P< 0.05) and bone (OR, 1.022; 95% CI, 1.001–1.043; P< 0.05) metastases. No association was found between miR-145 expression and metastatic status. The miR-21/24 ratio accurately identified synchronous metastases irrespective of organ site, with an area under the curve (95% CI) of 0.833 (0.722–0.944) and positive predictive value of 94.4%.
Conclusions
Alteration of miR-21 and miR-24 expression levels was associated with a high incidence of synchronous metastases in Indonesian CRC. The mir-21/24 ratio demonstrated significant potential as a biomarker for detecting synchronous metastases in CRC.
INTRODUCTION
Colorectal cancer (CRC) has emerged as the third most prevalent cancer worldwide, with an alarming incidence rate of 10%. Concurrently, CRC is responsible for 9.4% of all cancer-related deaths, placing it at the second-highest mortality rate among all cancers [1]. As the world’s fourth most populous nation, Indonesia is confronted with unparalleled challenges in cancer treatment and prevention. With an incidence rate of 8.6% and a mortality rate of 7.9%, CRC remains a significant burden of disease in Indonesia [2]. About 20% of newly diagnosed CRC cases are metastatic, and an additional 20% of cases advance to metastatic disease [3,4]. Metastatic CRC patients face a limited life expectancy and are at higher risk for life-threatening tumor-related obstruction, perforation, and bleeding complications [5]. The term “synchronous” in metastatic CRC refers to metastatic evidence identified concurrently with, before, or within a generally shorter timeframe following the diagnosis of primary CRC [6].
The absence of reliable cancer-specific biomarkers has rendered early diagnosis of CRC very challenging. Initial cancer development and progression drivers include significant cellular and subcellular alterations involving DNA, RNA, protein structure, and function [7]. MicroRNAs (miRNAs) are small non-coding RNAs that interact with target messenger RNAs (mRNAs) in the 3´-untranslated region (3´-UTR), causing posttranscriptional inhibition and mRNA degradation [8]. Evidence suggests that miRNAs modulate angiogenesis, tumor invasion, epithelial-mesenchymal transition (EMT), and cancer cell stemness, among various aspects of CRC metastasis [9]. The investigation of miRNAs in diagnosing and prognosis cancer and other disorders has become increasingly popular in the past few years [10]. Predicting metastasis of malignancies, such as CRC, continues to be challenging. While cures are rare, an increasing number of patients would anticipate prolonged survival through early diagnosis and appropriate treatment of metastatic CRC. Detecting metastases, particularly in the abdomen, often relies on advanced imaging techniques, such as positron emission tomography scans, computed tomography (CT), and magnetic resonance imaging (MRI). In low-resource settings such as Indonesia, the high costs and low availability of specialized infrastructure limit the widespread implementation of these diagnostic tools. Consequently, this can delay the detection and treatment of metastatic CRC, impacting patient outcomes.
A review of 23 miRNA expression studies revealed that out of the 164 miRNAs discovered to be significantly altered in CRC, about two-thirds were upregulated, and one-third were downregulated [11]. It is evident that specific miRNAs demonstrate crucial oncogenic roles while others demonstrate crucial tumor suppressor activities. These functions should be assessed for each miRNA individually in CRC. Previous investigations indicated that miR-24 and miR-145 were tumor suppressors due to lower expression in CRC [12,13]. In contrast, miRNA-21 was demonstrated to be overexpressed in CRC, suggesting its potential as an oncogenic miRNA [11]. Despite mounting evidence of miRNAs in cancer, studies investigating their role in the metastatic process of CRC remain scarce. This study aims to evaluate the accuracy of miR-21, miR-24, and miR-145 in predicting synchronous metastases in CRC.
METHODS
1. Population and Sample
From 2021 to 2024, we included patients who underwent a colonoscopy and received their initial diagnosis of CRC at Dr. Soetomo General Hospital, Surabaya. Subjects eligible for inclusion criteria are as follows: (1) age over 18 years; (2) diagnosis of CRC confirmed by histopathological results. All subjects performed extensive examinations, including abdominal CT/MRI and additional imaging as indicated by clinical suspicion of metastases. Patients were excluded if they had 1 or more of the following criteria: (1) incomplete clinical staging and metastatic status information; (2) a prior history of other malignancies; (3) receiving chemotherapy and/or radiotherapy; or (4) tissue specimens that did not meet the qualifications for histological, immunohistochemical, and polymerase chain reaction (PCR) analysis.
Demographic and clinical characteristic data were collected. The clinical stage of CRC was determined according to the 8th edition of the American Joint Committee on Cancer (AJCC) [14]. Forceps radial jaw 4 (Boston Scientific, Marlborough, MA, USA) were used to collect mucosal biopsies of the tumor. Biopsy samples were fixed using liquid nitrogen and formaldehyde before analysis.
We conducted this study in compliance with the principles of the Declaration of Helsinki. The study’s protocol was reviewed and approved by the Dr. Soetomo Hospital Health Research Ethical Committee Board (No. 0179/KEPK/IV/2021). Written informed consent was obtained from research subjects.
2. RT-qPCR to Detect the Expression of miR-21, -24, and -145
Total RNA was extracted using the Maxwell RSC miRNA Tissue Kit (Promega, Madison, WI, USA; AS1460) and a Maxwell RSC instrument (Promega, AS4500), following the manufacturer’s protocol. The RNA quantity was measured, and the samples were stored at –80°C. To prepare for PCR, reverse transcription was conducted to produce cDNA using reverse transcription (GoScript Transcription System; Promega) following the manufacturer’s instruction, which was used as the template. The PCR was carried out according to the guidelines provided with the reverse transcription quantitative PCR (RT-qPCR) kit using GoTaq qPCR Master Mix (Promega). Details of the primers used are provided in Supplementary Table 1, with U6 RNA as the endogenous control [15,16].
3. Statistical Analysis
Descriptive statistics are presented using the mean± standard deviation, median (interquartile range), and count (percentage). We compared the clinical characteristics between localized and metastatic CRC patients using the independent t test or Mann-Whitney test for numerical data and the chi-square test or Fischer exact test for categorical data.
We analyzed the expression levels of miRNAs about metastatic status and conducted further subanalyses by metastatic sites utilizing the Mann-Whitney test. Backward stepwise logistic regression was used for multivariate analysis. Receiver operating characteristic (ROC) analysis was performed to assess the ability of miRNAs to discriminate against CRC patients with and without synchronous metastases. Optimal cutoff points were determined by employing the Youden index. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were used to express the performance of miRNAs in predicting synchronous metastases. All statistical analyses were performed using IBM SPSS Statistics version 24.0 (IBM Co., Armonk, NY, USA).
RESULTS
1. Clinical Characteristics of Subjects
Out of 84 CRC patients confirmed by histopathological results, 21 were excluded due to incomplete clinical staging information. Of the 63 CRC patients we included, 37 (58.7%) had localized CRC, and 26 (41.3%) had metastatic CRC. The study subjects comprised 46.0% males, 74.6% Javanese, and a mean age of 56.3 years. A first-degree family history of CRC was reported in only 10 (15.9%) subjects. Age, sex, ethnicity, body mass index, smoking history, and a history of CRC in first-degree relatives were not statistically different between localized and metastatic CRC (P>0.05). Despite the prevalence of left-sided CRC in the overall subjects (81.0%), no significant differences were observed between the groups based on tumor location. Most subjects were classified as CRC stage III (47.6%) and IV (41.3%) according to the AJCC 8th edition staging system. The liver was found to be the most common site for metastasis (31.7%), followed by the lungs (14.3%) and bones (3.2%). Two subjects with distant lymph node metastases and one ovarian metastasis accounted for the other (4.8%) metastatic sites. Furthermore, histological analysis revealed that 77.8% of the biopsy samples collected were identified as well-differentiated adenocarcinoma (Table 1).
Clinical Characteristics of Subjects
2. miRNAs Expression in Localized and Metastatic CRC
Table 2 displayed a comparison of miRNA expression levels based on metastatic status. We observed that localized CRC had significantly lower expression of miR-21 as compared with metastatic CRC, regardless of the metastatic site (3.404 [34.230] vs. 31.262 [120.358], P<0.05) (Fig. 1A). In contrast, localized CRC demonstrated higher expression of miR-24 compared to metastatic group (12.900 [42.376] vs. 0.024 [4.680], P<0.01] (Fig. 1B). These findings indicated that the expression of miRNAs was altered in CRC metastasis, particularly the upregulation of miR-21 and the downregulation of miR-24. In turn, metastatic CRC showed a higher ratio of miR-21/24 (P<0.01), as seen in Fig. 1D. No statistically significant differences were found in miR-145 levels between groups (Fig. 1C). This study additionally performed a sub-analysis to compare miRNA levels across various organ sites of metastasis. Patients with liver metastasis had a decreased level of miR-24 compared to those without (0.006 [3.136] vs. 10.192 [42.282], P<0.01). Conversely, patients with lung and bone metastases exhibited elevated levels of miR-21 (64.325 [201.528] vs. 4.781 [36.490], P<0.01; 198.422 [not available] vs. 8.799 [47.725], P<0.05) The ratio of miR-21/24 differed statistically only in patients with and without liver metastases (Table 2).
Expression of miR-21, miR-24, miR-145, and the miR-21/24 Ratio with Various Metastases
(A-C) Comparison of miR-21, miR-24, and miR-145 expression in localized versus metastatic CRC. (D) Significant differences in miR-21/24 ratio were also observed between the groups. CRC, colorectal cancer; miR-21, micro-RNA 21; miR-24, micro-RNA 24; miR-145, micro-RNA 145.
The miR-21/24 ratio was excluded from the multivariate analysis to minimize multicollinearity. Our study revealed that miR-21 overexpression was identified as an independent predictor of synchronous metastasis in CRC (odds ratio [OR], 1.016; 95% confidence interval [CI], 1.003–1.030; P<0.05). Based on the sub-analysis of metastatic site, there was a higher likelihood of lung and bone metastases associated with upregulation of miR-21, with ORs of 1.011 (95% CI, 1.002–1.020) and 1.022 (95% CI, 1.001–1.043), respectively (Table 3). Both miR-24 and miR-145 failed to demonstrate a comparable association.
miR-21 as an Independent Predictor of Colorectal Cancer Metastases, Specifically in the Lung and Bone
3. The Predictive Value of miR-21/24 Ratio in Relation to Metastatic Status
Based on the observed alterations in miR-24 and miR-21 expression in CRC tissues, we performed an analysis of the miR-21/24 ratio to evaluate whether it could improve the accuracy of their diagnostic value prediction. ROC analysis showed that the miR-21/24 ratio had the best diagnostic accuracy to predict synchronous metastases, with an AUC of 0.833, outperforming miR-21 and miR-24 as a single marker. Using a cutoff value of 20.463, the miR-21/24 ratio demonstrated a PPV of 94.4, suggesting its utility as biomarker for metastases. Further analysis that subdivided metastases according to organ site showed that the miR-21/24 ratio was a good predictor of liver metastases, with an AUC of 0.801 and PPV of 81.2%. In addition, the overexpression of miR-21 as a single marker was also identified as a reliable predictor of lung metastases in CRC, with an AUC of 0.805, sensitivity of 77.8%, and specificity of 83.3% (Table 4, Fig. 2).
Diagnostic Accuracy of miR-21, miR-24, and miR-21/24 in Predicting Various Metastases
(A) ROC curves of miR-21, miR-24, and miR-21/24 ratio to differentiate localized and metastatic CRC. (B) ROC curves of miR-24 and miR-21/24 ratio to differentiate CRC patients with and without liver metastases. (C) ROC curves of miR-21 to differentiate CRC patients with and without lung metastases. ROC, receiver operating characteristic; miR-21, micro-RNA 21; miR-24, micro-RNA 24; CRC, colorectal cancer.
DISCUSSION
This study provides the first evidence of in vivo alteration in miRNAs expression in Indonesian patients with metastatic CRC. We observed that, in contrast to localized CRC, tumors in patients with distant metastases showed lower expression of miR-24. A prior study had provided consistent evidence indicating that the downregulation of miR-24-3p is associated with higher cell proliferation, migration, and invasion in CRC [17]. An in vitro study demonstrated that the expression of miR-24 reduced the metastatic potential of cancer cells via translational repression of p130Cas [18]. In contrast, another study demonstrated that miR-24 was overexpressed in natural killer cells from CRC patients, in comparison to healthy controls. It was suggested that miR-24 acts as an oncogene since its overexpression reduced the secretions of interferon-γ and tumor necrosis factor-α via Paxillin targeting [19]. Even within the same cancer type, this bidirectional effect could be driven by differences in miRNA targets depending on the specific environment [20].
We also found that miR-21 overexpression was an independent predictor of synchronous metastases in CRC. This finding is consistent with an earlier study that reported a higher expression of miR-21 in metastatic CRC [21]. It has been discovered that the miR-21 plays an important role in promoting CRC metastases by downregulating programmed cell death 4 (PDCD4) [22]. Silencing of PDCD4 resulted in the upregulation of the IL-6/STAT3 signaling pathway, leading to the EMT and metastatic processes [23,24]. Subsequent bioinformatic analysis had identified a conserved target-site for miR-21 in the PDCD4-30-UTR at the 228-249 nt [22]. miR-145 was one of the first 2 miRNAs identified in association with CRC, along with miR-143 [25]. However, in our study, there were no significant differences in miR-145 expression between the metastatic and the localized CRC. Conversely, a previous study demonstrated that miR-145 prevents metastasis in CRC by specifically targeting fascin-1 [26]. This negative finding suggests that metastatic CRC in Indonesia may not primarily involve activation of the epidermal growth factor receptor (EGFR) signaling pathway, as low expression of miR-145 correlates with overexpression of kirsten rat sarcoma viral oncogene homolog (KRAS) and v-Raf murine sarcoma viral oncogene homolog B1 (BRAF), leading to increased proliferation and migration of cancer cells [27].
Distant metastases were observed in 41.3% of CRC patients in our study. The incidence of metastases in CRC had been previously reported to be as high as 20% to 25% [28-30]. The lack of a nationwide CRC screening program in Indonesia and the relatively limited availability of colonoscopy facilities, in comparison to Western countries, could potentially account for the disparity in incidence rates. The liver was found to be the most prevalent metastatic site, accounting for 76.9% of all metastatic CRC cases in our study. This finding aligns with an earlier study that reported the liver as the most common site for metastasis in 70% of CRC patients [31]. As the primary collector of intestinal mesenteric drainage, the hepatic portal venous system makes the liver the most frequent location for CRC metastases. Liver metastases were present in around 15% of patients upon diagnosis, and more than half of patients with CRC might develop liver metastases at a certain point in their lives [32]. There is mounting evidence that altered expression of miRNAs plays an important role in the pathogenesis of metastases in CRC, either as tumor suppressors or oncogenes, modulating the expression of the mRNAs of their target [33]. The genetic heterogeneity of cancer, where distinct molecules can be altered by different mechanisms, is useful for targeted treatment of CRC in a specific population [34]. A growing number of evidence suggests the EMT as an important factor in tumor invasion and metastases. EMT is a process whereby epithelial cells lose their epithelial state and develop mesenchymal features. This phenomenon is characterized by a reduction in cell polarity, weakening of cell adhesion, and enhanced cell motility. This allows the cancer cells to invade the surrounding tissues, enter the lymphatic or blood vessels, travel to distant locations through the circulatory and lymphatic systems, and ultimately establish themselves in a metastatic environment [35].
Compared to individual miRNAs, the present study found that the miR-21/24 ratio was more accurate in predicting synchronous metastases in CRC with an area under the curve (AUC) of 0.833. To our knowledge, this is the first study to reveal that miR-21/24 ratio in tissues may accurately distinguish CRC patients with and without distant metastases. Since the liver was the most prevalent location of metastasis in our study, we found that the miR-21/24 ratio was a reliable marker of liver metastases (AUC=0.801). With an AUC of 0.805 and 0.726, respectively, our study was the first to demonstrate the use of miR-21 and miR-24 for differentiating CRC patients with lung and liver metastases. From an alternative clinical standpoint, results from earlier CRC studies suggest that the miR-21/24 ratio could be helpful in discriminating between normal and malignant tissues. Despite using plasma samples, it was discovered that the miR-21/24 ratio is significantly higher in metastatic CRC than in localized CRC [16]. Employing this type of ratio as a biomarker also offers several advantages. It eliminates the need for an internal reference, enhances diagnostic potential by improving discrimination and specificity, and reduces the risk of missing a metastasis event [34].
miRNA profiling in specific subsets of CRC patients may be beneficial for personalizing treatment and providing prognostic information. Our study suggests that the miR-21/24 ratio has potential as a biomarker for earlier detection of CRC patients at high risk of metastasis following colonoscopy. Careful prioritization of extensive workup for metastases can be achieved using the miR-21/24 ratio data obtained from biopsies. While this study provides novel insights into the role of miR-21, miR-24, and their ratio in predicting synchronous metastases in CRC, several limitations should be acknowledged. First, our analysis was limited to tumor biopsy samples, and we did not include plasma samples from the same cohort. This restricts our ability to directly compare circulating and tissue-based miRNA expression, which could provide a more comprehensive understanding of their biomarker potential. Future studies should incorporate plasma-based miRNA profiling to evaluate their noninvasive diagnostic and prognostic utility. Second, although we identified significant alterations in miR-21 and miR-24 expression, our study did not assess their downstream molecular targets or signaling pathways in CRC metastasis. Given the complexity of miRNA-mediated regulation, further investigations should focus on elucidating the functional impact of these miRNAs on key metastatic regulators, such as those involved in EMT, cell anoikis, and tumor microenvironment interactions. Additionally, we acknowledge that the designation of miRNAs as strictly oncogenic or tumor suppressive can be overly simplistic, as their roles are often context-dependent. A deeper mechanistic study, including transcriptomic and proteomic analyses, would be valuable in clarifying their functional roles in CRC progression. Lastly, as this study was conducted in a single-center Indonesian cohort, the generalizability of our findings should be validated in larger, multi-center studies with diverse populations. Addressing these limitations in future research will help refine the clinical application of miRNA biomarkers and improve early detection and risk stratification for metastatic CRC.
Despite the limitations, our study provides the first evidence of in vivo alterations in miR-21 and miR-24 expression in Indonesian CRC patients, highlighting their potential as biomarkers for detecting synchronous metastases. By identifying the miR-21/24 ratio as a strong predictor of metastasis, our findings lay the groundwork for future research on noninvasive miRNA-based diagnostics and targeted therapeutic strategies in CRC.
In conclusion, altered expression levels of miR-21 and miR-24 correlated with a higher incidence of synchronous metastases in Indonesian CRC. The mir-21/24 ratio, which combined the two, demonstrated significant potential as a biomarker for detecting synchronous metastases in CRC.
Notes
Funding Source
This research supported by Directorate of Research and Community Service (DRTPM), Ministry of Education, Culture, Research and Technology, Republic Indonesia and Universitas Airlangga (Nos. 575/UN3.14/PT/2020; 576/UN3.14/PT/2020).
Conflict of Interest
No potential conflict of interest relevant to this article was reported.
Data Availability Statement
All study-related data have included in the publication or provided as supplementary information.
Author Contributions
Conceptualization: Rezkitha YAA, Hidayat AA, Miftahussurur M. Data curation: Hidayat AA, Normalina I, Alfaray RI, Matsumoto T, Yamaoka Y. Formal analysis: Hidayat AA, Normalina I. Funding acquisition; Investigation: Rezkitha YAA, Miftahussurur M. Methodology: Lusida MI, Alfaray RI, Matsumoto T, Yamaoka Y. Project administration: Rezkitha YAA. Resources: Miftahussurur M. Supervision: Lusida MI, Yamaoka Y, Miftahussurur M. Validation: Matsumoto T, Yamaoka Y, Miftahussurur M. Visualization: Hidayat AA. Writing – original draft: Rezkitha YAA, Hidayat AA, Normalina, I. Writing – review & editing: Yamaoka Y, Miftahussurur M, Alfaray RI, Lusida MI, Matsumoto T. Approval of final manuscript: all authors.
Supplementary Material
Supplementary materials are available at the Intestinal Research website (https://www.irjournal.org).
RT-qPCR Primer Sequences