Micrornas As Colorectal Cancer Biomarkers Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Colorectal cancer is a common pathology, accounting for approximately 8 of cancer deaths. The induction of a metastatic cascade heralds a potentially devastating endpoint with limited curative therapeutic options and early detection is therefore crucial. Novel diagnostic and predictive biomarkers are urgently needed as current candidates lack sensitivity and specificity, particularly in the context of early disease. MicroRNAs are a group of short non-coding RNAs which regulate gene expression at a post transcriptional level and have recently been shown to hold promise as faecal, circulating or tissue biomarkers. This review explores current progress and future challenges in this rapidly expanding field.


Colorectal cancer is a common malignancy and accounts for approximately 8% of all cancer deaths [1]. Despite advances in surgical technique and chemotherapeutic strategies, approximately 50% of patients diagnosed with colorectal cancer will die of the disease, most commonly due to the development of distal metastases [2]. The factors which precipitate this potentially devastating metastatic cascade remain poorly understood. A resultant inability to stratify patients means that individualised therapeutic options are at present, implausible. Selection of therapy is thereby based primarily on clinicopathological characteristics at diagnosis, highlighting an urgent need for novel diagnostic and predictive biomarkers [3]. Furthermore, colorectal cancer is a disease which has long been recognised to exhibit stepwise progression in carcinogenesis through the adenoma-carcinoma sequence and is thereby academically amenable to early detection [4]. Given the urgent need for novel biomarkers in colorectal cancer, it is unsurprising that much research is ongoing in this field. Over the past decade, a large number of potential biomarker candidates have been investigated ranging from proteins associated with cell proliferation and apoptosis [5], genetic alterations and epigenetic factors including DNA methylation, components of faeces including bile acids and short-chain fatty acids . Many have low sensitivity and specificity profiles and others have yet to be validated. The search for chemically stable and clinically applicable candidate biomarkers has recently led to enormous interest a remarkable group of small molecules termed microRNAs [3].

microRNA structure and function

MicroRNAs are a group of short non-coding molecules 18-25 nucleotides in length, found in both prokaryotes and eukaryotes, known to regulate gene expression at a post-transcriptional level [6, 7]. Two decades ago, Lee et al discovered that small non-coding RNAs, initially thought to be "junk", were the product of a gene (lin-4) which affected development in a transparent nematode Caenorhabditis Elegans [8]. Almost a decade later, Calin and colleagues published a seminal paper linking micro-RNAs to cancer for the first time [9]. In a model of B-cell chronic lymphocytic leukaemia, Calin et al identified a translocation-induced deletion at chromosome 13q14.3, with associated loss of miR-15a and miR-16-1[9]. More than 1000 microRNAs have since been described, with each assigned a numerical identifier upon discovery. It is now understood that microRNAs can function as oncogenes or tumour suppressor genes. Each microRNA holds the potential to regulate the expression of numerous protein-coding genes (tens to hundreds). MicroRNAs are involved in almost all cellular processes and dysregulation contributes to the pathogenesis of most human malignancies [8, 10, 11].

MicroRNAs are matured by a triphasic mechanism which commences in the nucleus with transcription of a pri-miRNA precursor agent [6]. Further processing by an RNA specific ribonuclease enzyme complex (DROSHA) then results in production of a second intermediate stem-loop precursor prior to transportation to the cytoplasm via exportin 5 [12]. Once at this location, an RNase III family endoribonuclease and haploinsufficient tumor suppressor (Dicer) processes mature miRNAs from the 5′ or 3′ arm of the precursor agent, resulting in the final microRNA product [6, 12].

microRNAs as biomarker candidates

Cancer is a diverse and multifactorial disease, and it is therefore unlikely that a single biomarker will detect all cancer of a given solid organ with high specificity and sensitivity. In this context, a panel of candidate microRNAs may provide a signature for a particular cancer or tumour type [13, 14]. While any detectable and reproducible molecular alteration at DNA, RNA or protein level has the potential to be investigated as a biomarker, not all are equally relevant in terms of clinical application. To address this, characteristics of an ideal cancer biomarker have been well described and are outlined in Box 1. Although at a relatively early stage in the translation from bench to bedside, microRNAs fulfil many of these criteria. These small molecules are detectable in blood and faeces, facilitating ease of detection. A biomarker must also be capable of differentiating between diseased and healthy individuals. Aberrant expression of microRNAs has now been reported in many cancers, including colorectal [15-19]. The exact mechanism of this dysregulation has yet to be elucidated although histone acetylation has been implicated and some reports suggest that DNA hypermethylation may result in down-regulation of specific microRNAs [10]. Of crucial importance in assessing the utility of microRNAs as biomarkers, is the discovery that microRNAs are highly specific to tissue type and tumour type [18, 20, 21]. Furthermore, altered expression of specific microRNA genes contributes to the initiation and progression of cancer [9] indicating a role for therapeutic intervention. Finally, micro-RNAs are extremely stable molecules, particularly in tissue and in the context of research, have been shown to maintain their integrity for up to a decade in formalin-fixed paraffin embedded colorectal tumour samples [19, 22]. Biomarkers may be divided into diagnostic, predictive and prognostic and the potential of microRNAs in these areas will be discussed in detail later in this manuscript (Table 1).

Current colorectal cancer biomarkers

While recent advances in molecular technology have resulted in the discovery of many putative colorectal cancer biomarkers, the markers commonly used in clinical practice at present remain limited (Table 2). The majority of national screening programmes currently centre on the use of colonoscopy or faecal immunohistochemistry/occult blood testing [23]. Colonoscopy is an expensive, invasive procedure which carries a significant risk of intestinal perforation [23]. Faecal testing is more cost effective and less invasive but sacrifices diagnostic performance [24]. Guaiac-based faecal occult blood testing has relatively low specificity and sensitivity in the detection of colorectal cancer (33-50%) and this becomes even lower in the context of colonic adenomas (11%)[24-26]. Faecal immunochemical test is a newer method of faecal haemoglobin measurement and has been shown to have improved sensitivity for colorectal cancer and advanced adenomas, although some studies report that FOBT specificity is better [27].  Carcinoembryonic antigen, a high molecular weight glycoprotein which is a member of the immunoglobulin super family, has been used for many years as a circulatory colorectal cancer biomarker and is recommended by prominent advisory agencies [28]. However, it has shown only modest success when assessed according to characteristics of an ideal biomarker. While it is relatively cost effective and non-invasive, as a diagnostic biomarker it exhibits low sensitivity and specificity in particular in the setting of early cancers [29]. It is of more use in the setting of measuring response to treatment although levels may be high in healthy smokers or in response to inflammatory conditions such as Inflammatory Bowel Disease (IBD) or pancreatitis [29]. Recently, focus has remained on genetic markers including DYPD (dihydropyrimidine dehydrogenase), KRAS ((Kirsten rat sarcoma viral oncogene) and BRAF (v-RAF murine sarcoma viral oncogene

homolog B1). However, a recent review found that the majority remain in the discovery phase and require clinical validation [29].

microRNAs as diagnostic and prognostic biomarkers in colorectal cancer

Numerous microRNAs have been reported to be differentially expressed in colorectal cancer to date [19, 30] (Table 3). As previously discussed, the stepwise progression of colorectal cancer from resectable polyp to tumour renders it a disease amenable to the use of diagnostic biomarkers for timely intervention. There is a known link between inflammation, dysregulated microRNA expression and colorectal cancer [31]. Therefore, microRNAs represent exciting putative diagnostic biomarkers in this malignancy. Although it may be pragmatic to ultimately consider a panel of microRNAs as a colorectal cancer "multimarker", many individual microRNAs have also shown promise as circulating markers. For example, a recent study by Huang et al powered by over 150 colorectal cancer patients found plasma miR-29a and miR-92a combined to display 83.0% sensitivity and 84.7% specificity in discriminating colorectal cancer from healthy controls [32]. Considering the weaknesses of circulating markers in current clinical use, these results are promising. A recent systematic review by Ma et al identified six microRNAs which are most consistently reported differentially expressed in colorectal cancer after reviewing all published studies which examined differential expression between tumour and tumour associated normal samples in colorectal cancer [3]. The microRNAs identified in this study were mir-106a (up regulated), miR-125a, miR-133a, miR-145, miR-30a-3p and miR-139 [3]. Interestingly, all six were downregulated in colorectal cancer tissue relative to adjacent normal tissue [3]. In pursuit of non-invasive diagnostic biomarkers, research into faecal microRNA has yielded interesting results. Link et al found higher expression of miR-21 and -106a in patients with adenomas and colorectal tumours, compared with healthy controls, although numbers were small (n=29 cancers, n=8 controls). Ahmed et al proposed that a panel of micro-RNA isolated from faecal samples may hold greater sensitivity and specificity than currently used screening genomic, methylomic or proteomic methods for colon cancer [33]. Regardless of specific putative targets, it is widely accepted that microRNAs can be extracted from stool easily and reproducibly, providing enormous scope for future research in this area.

Recently, an exciting association has been shown between microRNA expression and microsatellite instability (MSI) stratification into subgroups such as low MSI and HNPCC-associated, which may prove interesting in terms of risk stratification [34]. Schepeler et al similarly found that in a cohort of stage II colon cancers, MSI status for the majority of cancers could be correctly predicted based on microRNA expression profiles [35].

In terms of predictive biomarkers, colorectal cancer is a disease in grave need of novel targets, particularly in rectal cancer. In patients with locally advanced rectal cancer, standard therapy includes neoadjuvant chemoradiotherapy with fluoropyrimidines (capecitabine or 5-fluorouracil) followed by radical surgical resection [36]. However, a significant proportion of tumours respond only partially or not at all to neoadjuvant chemotherapy [36]. Not only have these patients been exposed to potentially toxic side effects without benefit, they are also at risk of recurrence with resultant poor prognosis. Micro-RNAs may play a role as predictive biomarkers in delineating response to treatment. In one study, expression of high levels of miR-196a was associated with chemo sensitivity towards platin derivatives [37]. Although numbers were small (n=20), Svoboda et al found that miR-215,miR-190b and miR-29b-2* to be overexpressed in non-responders, and let-7e, miR-196b, miR-450a, miR-450b-5p and miR-99a* to be overexpressed in responders [36]. Using these miRNAs, the authors successfully identified 9 of 10 responders and 9 of 10 non-responders (p < 0.05) [36].

In terms of prognosis, micro-RNAs which correlate with poor prognosis have been identified in tumour samples from patients with colorectal cancer [38, 39]. Tumours expressing high levels of mir-21 and mir-200 appear to exhibit worse prognosis and a more aggressive disease phenotype, while mir-29a expression is associated with greater degrees of nodal positivity [19, 39]. The exact mechanism of associations of this type has yet to be elucidated; however some insights can be found in the literature. For example, one study which reports an association between overexpression of miR-200c and shorter survival time, used sequencing analysis to show that miR-200c expression was strongly associated with the mutation status of the p53 tumour suppressor gene [40]. Mir-143 and mir-145 were amongst the first microRNAs found to be dysregulated in colon cancer [41]. Most recently, miRNA-143 expression levels have been shown by Pichler et al to serve as an independent prognostic biomarker for colorectal cancer in tissue studies of 77 patients [42].

Conclusion and future perspectives.

Initially, investigation of carcinogenesis focussed solely on epithelial tumour cells. It is now that a bidirectional relationship between stromal and epithelial cells impacts heavily on core tumour characteristics including degree of angiogenesis and inflammation, all of which are reflected in patient prognosis [43]. MicroRNAs may play a crucial functional role in mediating this cellular crosstalk [44]. Most recently focus has been placed on exosomal microRNA contained within small nano-vesicles [45]. Most recently, research has revealed that exosomal microRNAs may affect physiological and pathological cell conditions, and may be useful as novel therapeutic agents [46]. This rapidly expanding area of research is likely to yield interesting findings in the coming months. In summary, micro-RNAs hold enormous potential as putative diagnostic, predictive and prognostic markers for patients with colorectal cancer, detectable in both circulation and faeces. Not only have over 400 microRNAs been shown to be differentially expressed in colorectal cancer, growing evidence suggests a critical functional role in the colorectal tumour microenvironment [21, 47, 48]. Furthermore, microRNAs have been shown to provide crucial information regarding response to therapy, microsatellite instability and core tumour characteristics. Although microRNAs as a group of molecules represent exciting putative biomarkers, progression from bench to bedside requires phased validation with large-scale cohorts including early cancers and adenomas.

Executive Summary

Colorectal cancer is a disease in urgent need of diagnostic, prognostic and predictive biomarkers in order to improve detection and individualise therapy.

Current biomarkers in clinical use are suboptimal in terms of sensitivity and specificity.

microRNAs are a group of small non-coding molecules known to regulate gene expression at a post transcriptional level

microRNAs hold enormous potential as putative colorectal cancer biomarkers particularly in the area of early detection and prediction of chemotherapeutic response, areas in which current clinical biomarkers lack efficacy.

Since cancer is a multicellular, multifactorial disease, a panel of microRNAs may exhibit increased performance.

Future considerations include elucidation of the role of exosomal microRNA and validation of proposed targets with large-scale cohort studies, in particular focussing on early disease.