Oesophageal cancer is the eighth most common cancer type worldwide, and the sixth most common source of cancer mortality accounting for over 400,000 deaths and 456,000 new cases every year (2012 figures). The most frequent type of this cancer is oesophageal squamous cell carcinoma (ESCC) representing 50–60% of cases worldwide, rising to ~90% of cases in countries of Asia. ESCC is an aggressive malignancy and because most patients present with advanced disease, it is typified by a poor prognostic outcome (overall 5-year survival of 10–20%). There is therefore great clinical interest in better and earlier diagnosis. Although endoscopy, the current diagnostic technique of choice for ESCC, can be effective in detecting early stage ESCC, its high cost and invasiveness precludes widespread use as a screening tool. In contrast, non-invasive diagnostic techniques such as blood tests are not only suitable for screening, more economic and more comfortable for patients but most importantly are optimal for repeat sampling. This characteristic represents a potential paradigm change in cancer management as drug response and disease progression can be followed on a regular basis allowing the fine-tuning of treatment regimens to match the patient experience, in other words a personalised medical approach. Circulating microRNAs (miRNAs) are particularly attractive candidates as non-invasive biomarkers as tumour-associated miRNAs present in the blood are present at much higher levels than their counterpart circulating DNA, and unlike other RNA species, miRNAs can resist the high levels of RNase that are present in biological fluids.
The publication of the manuscript ‘Circulating microRNAs in plasma of patients with oesophageal squamous cell carcinoma (ESCC)’ by Komatsu et al. in this journal was the first description of circulating miRNAs in the plasma of ESCC patients (https://www.nature.com/articles/bjc2011198). In their highly cited publication, the authors measured levels of miR-21, miR-184, miR-221 and miR-375 in plasma samples taken from ESCC patients and healthy donors. These miRNAs were chosen on the basis of previous studies that showed an importance in ESCC tumour tissue. Interestingly, none of these of these miRNAs were found to be differentially expressed in a previous study that used NGS to identify miRNAs in sera from ESCC patients (https://academic.oup.com/clinchem/article/56/12/1871/5622231). In this publication, Komatsu et al. observed significant up-regulation in the levels of miR-21 and miR-184, although the latter was only detectable in half of cases, down regulation of miR-375 and no difference in levels of miR-221 between plasma from 20 ESCC patients and 10 healthy controls. They validated the differential expression of miR-21 and miR-375 in an extended cohort of 50 ESCC patients and 20 healthy controls and found an association of miR-21 levels with the presence of vascular invasion and disease recurrence. They also investigated whether differential expression of miR-21 and miR-375 in plasma was reflected in the tumour. Although they observed higher levels of miR-21 and lower levels of miR-375 in primary ESCC tissue when compared to normal mucosa, they did not see any difference between cases that had high or low expression of these miRNAs in plasma. Similar findings have since been observed in many other studies (in many different cancer types), presumably reflecting intratumoural heterogeneity and/or non-passive release of miRNAs from tumour tissue into the circulation. The authors further demonstrated that plasma levels of miR-21 decreased significantly after surgery, although there were no differences in levels of miR-375. Similarly changes in the levels of miR-21 in post-operative plasma samples have subsequently been reported in other cancer types including gastric, breast and lung cancer.
Since this publication there have been many follow-up publications that have not only repeated the work of Komatsu and colleagues but expanded ESCC blood-based biomarkers studies to other circulating miRNAs and beyond to other non-coding (nc)RNA species including long ncRNA (lnc)RNA and circular (circ)RNAs. Yao et al. carried out a meta-analysis of 33 published studies of ESCC circulating miRNAs representing 3156 ESCC patients and 2304 healthy controls and 43 different miRNAs (https://www.jcancer.org/v09p2876.htm). According to their analyses both miR-21 and miR-375 levels had high diagnostic power with an overall AUC value of 0.840 (ESCC (n = 586 and n = 408 respectively) and healthy controls (n = 379 and n = 2017 respectively)) with data taken from 7 and 4 articles respectively. Although it should be noted that this meta-analysis did not take into account either source (i.e., plasma or sera) nor the ethnicity of the patients both of which are likely to be important confounding factors.
Despite the original findings of Komatsu et al. being successfully replicated in independent studies, in common with many other circulating miRNA biomarker studies, there have been some apparently contradictory reports as well. For example, although several studies also reported the down-regulation of miR-375 in plasma (and sera), in another study this miRNA was observed to be significantly up-regulated in plasma obtained from 38 ESCC patients when compared to 19 healthy controls (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4509208/). There are many reasons for such disparities including biological, technical and analytical differences between studies especially as the low quantity and low quality of RNA present in biological fluids results in a low signal-to-noise ratio that is particularly sensitive to variability. This is further compounded by a high degree of variability in the expression of circulating RNA between individuals, even within healthy populations. Therefore, there is a clear need for standardisation in this area, or at least standardisation in the reporting, perhaps similar to existing MIAME guidelines. Even though great strides have been made in the short time since this manuscript was published there remains much to do to translate that potential into clinical reality. Nowhere is the divide between the public demand for liquid biopsies and the current state of the art more evident than the recent trial of Elizabeth Holmes.
COMPETING INTERESTS
The author declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.