MicroRNAs (miRNAs) are short strands of RNA that act as post-transcriptional regulators. Research in various disease processes from cancer to cardiovascular disease has found that miRNAs play a role in disease pathogenesis and have potential as biomarkers and therapeutic agents. However, as with all biomarkers particularly those that appear to be the next best thing, we need to proceed with caution.
When reviewing the potential of miRNAs as biomarkers we need to examine the problems that must be overcome before implementing routine use. Often a good approach is to go back to basics and examine whether miRNAs possess some of the characteristics of the ideal biomarker, involving consideration of clinical, analytical and practical criteria:
a) An ideal marker should have optimal clinical sensitivity and specificity. It is suggested that miRNA would provide increased sensitivity. In addition, miRNA expression profiles have been demonstrated to be pathognomonic, or tissue-specific; however, this does not necessarily translate into disease specificity.
b) Detection and quantification should be robust, rapid, simple, accurate, reproducible and inexpensive. MiRNA detection currently falls short of all of these prerequisites. There is often a low correlation of results obtained from different platforms or even from the same platform using reagents from different vendors. Standardization of these assays is a challenge for the near future. Data normalization, an often underestimated aspect of data processing, is crucially important and a major concern in obtaining accurate results. The lack of standardization and implementation of normalization methods is especially challenging. Some of the discrepant results between miRNA studies may be due, in part, to the application of different normalization methods.
c) Biomarker levels should be proportional to the degree of severity of pathology. Various studies have confirmed this; for example, the severity of drug-induced liver injury has been shown to correlate with the level of specific plasma miRNAs.
d) Minimally invasive procedure for obtaining samples. Measurement of circulating miRNAs would merely require a blood sample, making it highly convenient for the patient, physician and medical laboratory.
e) Minimal and predictable pre-analytical variability. Currently there are no universally implemented guidelines for the collection, preparation, and extraction of samples for miRNA analysis. Differences in specimen type (tissue, whole blood, plasma, or serum) can introduce a profound effect on miRNA concentrations. Furthermore, miRNA content in both plasma and serum can be influenced by cell remnant contamination from erythrocytes, leukocytes, or platelets. Biological variance, which has been largely neglected, is another extremely important variable that may affect the clinical utility of miRNAs.
From a laboratorian’s viewpoint, before applying large-scale effort to miRNA biomarker discovery, baseline parameters such as pre-analytical, intra-individual, and inter-individual variability of miRNAs must be explored and defined. Steps must be taken to carefully select miRNA-profiling platforms and data analysis methods for the acquisition of clinically meaningful and dependable data. Furthermore, standardization of miRNA biomarker evaluation is urgently needed.
It is worth mentioning that the ultimate expectation of miRNAs as novel biomarkers is to enhance the ability to optimally manage the patient. Thus, a key question to ask is whether miRNA biomarkers would significantly improve the diagnostic workup and eventually lead to positive targeted clinical outcome when compared to already in-use diagnostic standards.