Circulating microRNA biomarker studies: pitfalls and potential solutions.
Additional confirmation of circulating miRNAs was provided over the course of 2008 by independent laboratories. Although other circulating nucleic acids as well as miRNAs within cells and tissues had been under investigation for some time, these reports gave the first indications that circulating miRNAs might serve as cancer biomarkers. Lawrie et al. found increased miRs-21, -155, and -210 in sera of diffuse large B-cell lymphoma patients (3). Tewari's group published an association of increased serum miR-141 with prostate cancer (4). A subsequent publication by Chen et al. measured increased miR-25 and miR-223 in serum of lung cancer cases (5). The positive reception of these and later reports was accompanied by enthusiasm that the "liquid biopsy" might obviate the need for more invasive testing and also allow early warning of oncogenesis.
miRNA Biology and Biomarker Suitability
miRNAs are short RNA molecules. The average mature miRNA is 21 or 22 nucleotides in length and is processed from a hairpin precursor. When incorporated into the RNA-induced silencing complex, a miRNA may bind to target sequences in other RNA molecules with some degree of complementarity (for a comprehensive review of miRNAs and their function, see (6)). mRNAs that are targeted by miRNAs--and their protein products--are often modestly downregulated. Several thousand human miRNAs have been reported or predicted. Only a small fraction of these miRNAs are sufficiently abundant in any given cell type to exert posttranslational regulation. For example, a single miRNA, miR-122, comprises the majority of miRNA copies in the hepatocyte.
The twin bases for the attractiveness of extracellular miRNA as biomarkers are stability and dysregulation in the diseased cell. While bound to Argonaute proteins, miRNAs are stable in the extracellular environment after release from cells, whether as unprotected ribonucleoprotein complexes or within membranous vesicles (7, 8). The released miRNA can be detected and quantitated as part of the "miRNome" of a biological fluid, such as plasma or serum. In diseased tissue, and especially in cancers, miRNA dysregulation has been well established. If the miRNA changes in a tumor, for example, are reflected in the circulation, invasive tissue biopsies could be replaced by straightforward assays of easily obtained blood products, and early warnings of tumorigenesis might be possible.
Over the past 7 years, fulfilling the biomarker promise of stable miRNAs in circulation has remained a work in progress, in both cancer research and studies of nonneoplastic disease. Commonly reported miRNA biomarkers are largely nonspecific, associated with a wide range of conditions and outcomes. Limited overlap has been observed between the findings of even very similar studies of the same disease. Methodological challenges have been offered as explanations for this variability (9-12). Will it be possible to advance the field and move miRNA biomarkers from theory to clinical utility? In this review, the diagnostic specificity and reproducibility of miRNA biomarkers in circulation are examined by way of several examples. The concept of the tumor biomarker and its relevance to circulating miRNA findings is then assessed. Finally, several suggestions are provided, including a redoubled focus on miRNA in extracellular vesicles or other carriers, to maximize the biomarker discovery prospects of future miRNA profiling.
Diagnostic Specificity and Reproducibility
MULTIPLE ASSOCIATIONS: THE CASE OF miR-141
As described in the first circulating miRNA publication, maternal blood concentrations of miR-141 increased during pregnancy (2). Mitchell et al. found that miR-141 distinguished prostate cancer cases from controls, but the authors correctly noted that aberrant expression had been reported in other cancers of epithelial origin, including breast, colon, and lung (4). Thus, even the first 6 months of circulating miRNA publication presaged specificity questions: what would an increased miR-141 concentration tell the clinician--that lung, colon, or breast cancer is present, or that the patient is pregnant?
More complications became apparent as more data became available. Whereas miR-141 increased in pregnancy, increased concentrations were associated with preeclampsia (13). Patients with lower miR-141 were more likely to have systemic lupus erythematosus (14). In cancers, higher concentrations of miR-141 in resected colorectal cancer patients were found to be associated with long-term and disease-free survival (15), on the one hand, but on the other, and seemingly more consistently, with poor outcomes including biochemical recurrence (16), increased metastasis (17) in prostate cancer, or poor prognosis in prostate cancer (18), hepatocellular carcinoma (19), and colon cancer (20). Results obtained in an identification cohort supported association of miR141 with bladder cancer; however, this was not confirmed in a validation cohort (21). One publication reported no differences in serum miR-141 concentrations of biopsy-positive and -negative prostate patients (although miR-141 was associated with higher Gleason score) (22). Another study found that miR-141 concentrations were similar in prostate cancer cases and controls (23). Some authors even reported difficulty detecting miR-141 in the circulation (24).
Faced with this range of results and associations, are miR-141 and other miRNAs, such as the intensively investigated miR-21, really "prostate cancer markers" (25)? Or, for that matter, specific markers of any other disease? Or are these miRNAs, when increased in circulation, simply indicative of general disease states such as inflammation (25)? Specificity questions about miRNAs are not restricted to certain cancers. A recent review found that changes in circulating miR-16, -155, -21, -126, and -223 were each reported to be associated with at least 10 nonneoplastic conditions (26). Possibly, the utility of such miRNAs lies not in screening, but in the context of an already quite involved knowledge of the patient's condition (as seen, for example, in (22, 23)) and alongside other biomarkers.
BREAST CANCER STUDIES: WIDESPREAD LACK OF REPRODUCIBILITY PERSISTS
Specificity concerns may be accompanied and compounded by disparate findings by independent investigators. An important investigation of circulating miRNAs as breast cancer (BC) biomarkers was published in 2013 by Leidner et al. (27). Fifteen previous reports were compared with the authors' high-throughput sequencing data set. In 10 studies that measured miRNAs with individual quantitative PCR (qPCR) assays, only miR-21 (3 studies) and miR-155 (2 studies) were found to be upregulated in circulation by >1 group, even though 7 of the groups used the same normalization strategy. Five larger-scale profiling studies were also analyzed. In those studies, neither miR-21 nor miR-155 was consistently identified. In fact, 1 study reported downregulation of miR-21, and 2 reported downregulation of miR-155. There was no overlap of significantly differentially expressed miRNAs between the Leidner et al. data set and others. The authors noted that consistently discordant results outnumbered concordant results, justifying what they called "dampening enthusiasm" in their article title (27).
What of more recent developments? Between the Leidner et al. inclusion cutoff of July 2012 and September 2014, at least 26 additional publications have reported on circulating miRNAs in BC. To update the Leidner et al. findings, I examined these more recent publications as well. It is of course impossible to compare directly the results of all studies, since some have focused on distinct questions, such as response to particular therapies, prognostics vs diagnostics, metastasis, or genotype associations. Several investigators have measured only a small number of miRNAs, whereas others have performed genome-wide analysis. Nevertheless, 20 new publications were found to be directly relevant, in addition to 12 previous studies. A description of the methods and a citation list are provided in Supplemental Materials, which accompanies the online version of this article at http://www.clinchem.org/content/vol61/issue1.
Unfortunately, the more recent findings have not helped to build consensus. Differential regulation of a total of 143 miRNAs in plasma or serum was reportedly BC-associated across 32 publications (Fig. 1). Of these miRNAs, 100 were reported in only 1 publication. Discordant results were reported in different publications for 25 of the remaining 43 miRNAs, with discordance defined as different direction of regulation or differential regulation juxtaposed with at least 1 publication finding the miRNA to be a stable reference. An additional 8 miRNAs had fold regulation of <2 in a third or more of the supporting publications (a fold change that would be difficult to verify with qPCR in the clinical setting), or the magnitude of regulation could not be determined from the data provided.
Only 10 miRNAs, then, were found to be differentially regulated in the same direction and by >2-fold in >1 study (see online Supplemental Table 1). Just 1 miRNA, miR-126, was reported in >2 publications. The star arm (alternative sequence) of miR-126 was also reported in 2 of the same publications. Eight of the miRNAs were downregulated in BC cases--by 3.4- to 19-fold--a situation that, for reasons outlined below, could not be due directly to processes in the tumor tissue itself. The support for all 10 miRNAs derived entirely (9) or mostly (1) from 4 publications from the same institution, including 3 with overlapping authorship; the samples were presumably derived from the same or similar populations. One interpretation is that these findings underline the need to minimize preanalytic variables, including population diversity, sampling and processing protocol variations, and use of different sequencing or analysis facilities. This does not bode well, however, for clinical utility.
Strikingly, the same miRNAs and other small RNAs used and in some cases validated as stable references were almost all reported to be differentially regulated in BC serum or plasma in other publications. These included miR-16 (the most commonly used miRNA reference in qPCR experiments and also one of few miRNAs found to be differentially regulated in >2 BC publications), and miRs-10b, -30a, -30d,-103, -148b, -191, and -192, as well as the small RNA U6.
It should be noted that several other miRNA species were labeled as "discordant" in this analysis but deserve special mention. Differential expression of miR-21 was reported in 9 studies. In 7 studies, miR-21 was upregulated by 1.2- to 200-fold; in 1 study, the direction of regulation was unclear; and in 1 study, miR-21 was downregulated 4-fold. Although labeled here as "discordant," the weight of the evidence indicates upregulation of extracellular circulating miR-21 in BC plasma and serum. However, miR-21 is no more likely to be a specific BC marker than it is to be a specific prostate cancer marker. miR-21 was also associated with benign tumors (28) and is one of the most commonly reported biomarkers for other cancers and a variety of nonneoplastic conditions (26). miR-21 would be useful in BC only in combination with other, specific markers. As noted above, endothelial marker miR-126 was downregulated 2- to 33-fold in 3 studies, and its star arm was also downregulated. These findings might merit further research into how endothelia sense and signal BC states. Other miRNAs with occasionally reproduced findings included miR-155 and -223 (see online Supplemental Materials), but again, these have many reported disease associations.
Against the scattered positive findings, multiple studies have also reported completely negative results. These include a relatively large microarray study in which sera from 410 matched cases and controls were profiled comprehensively by Affymetrix chips (29). Only small differences were found, "ranging from 4% to 19%"--58 Clinical Chemistry 61:1(2015) i.e., differences that would not be reliably detectable with qPCR assays--and in any case, not one of these was significant (29).
miRNAs: Direct Markers of Neoplastic Growth, or of General Disease States?
CIRCULATING miRNA DOWNREGULATION IS NOTA RESULT OF DOWNREGULATION IN THE TUMOR
In its infancy, a neoplastic growth that displaces, not replaces, healthy tissue could not possibly be directly responsible for a decline of any miRNA in circulation, even if expression of that miRNA were completely shut off in the growth. Shutdown of a miRNA in cancer cells might result from genetic rearrangement or epigenetic silencing (30), but downregulation in circulation could happen only if the growing tumor negatively affected expression of the miRNA in other cells or, less plausibly, influenced stability of the miRNA in circulation. In contrast, a metastatic growth that replaces the healthy tissue source of a miRNA could conceivably reduce circulating concentrations of a miRNA to a degree commensurate with the percentage of tissue replaced. It seems likely that a cancer that replaced the majority of all sources of a particular miRNA would be fatal, or at least easily diagnosed without recourse to circulating biomarkers. Reductions in circulating miRNAs, if confirmed, are probably not the result of changed expression in the tumor and are, instead, nonspecific responses to the presence of neoplastic growth.
TUMOR RNA IS UNLIKELY TO EXPLAIN UPREGULATED CIRCULATING miRNA
Could a tumor directly increase the amount of specific circulating miRNAs? Here, the answer is a very qualified yes. We must first consider that most extracellular miRNAs in circulation have a blood or endothelial cell origin (31,32). miRNAs abundant in these cells and their "releasates" (the total extracellular fractions derived from these cells) are highly unlikely to be direct biomarker candidates. The fraction of extracellular RNA deriving from tissue is unknown and variable, but small. The fraction hailing from any given tissue would be determined by the amount of that tissue in the body, its relative vascularization, and other factors, but would certainly be even smaller. Finally, the fraction of circulating miRNA from a tumor would depend on the tumor's size and access to blood supply and the difference in miRNA expression compared with surrounding healthy tissue.
To continue with the BC comparison, consider the optimal situation of a hypothetical miRNA produced only in breast tissue. For a 65-kg individual with breast weight of around 1.5 kg and breast volume of around 1700 [cm.sup.3], a stage I tumor of 0.5 cm in diameter might be only four-thousandths of 1% of total breast tissue. A much larger 5-cm tumor might represent about 4% of total breast volume. In these 2 scenarios, the stage I tumor would have to contribute 50 000-fold more RNA to the blood than healthy tissue to reach even a 2-fold increase of the hypothetical miRNA in circulation. This increase would need to be achieved through some combination of upregulation in the cell, selective export, and enhanced stability in circulation. The large tumor would need only a 50-fold differential. Of course, the assumption that breast tissue is the only contributor of a circulating miRNA is inaccurate for most of the fairly ubiquitous miRNAs that have been proposed as biomarkers. Thus, for most tumors and miRNAs, upregulation by many orders of magnitude would be required to affect circulating concentrations of the miRNA. Such levels of overexpression might well induce toxicity. Furthermore, such drastic upregulation of specific miRNAs in tumor tissue is largely inconsistent with the literature. Compared with adjacent healthy tissue, fold changes of <2 are not at all uncommon (33). Although direct contributions of a tumor to circulating miRNA upregulation are more plausible than responsibility for downregulation, few tumors would release enough copies of a normally expressed miRNA to achieve a measurable increase in the peripheral blood. A miRNA-based liquid biopsy could not possibly detect the contributions of small, otherwise undetectable growths. Again, as for the downregulated miRNA, it is more likely that any upregulation observed in circulation is the result of a response or responses to the presence of neoplasm.
POSSIBLE EXCEPTION: NORMALLY SILENCED miRNAS
In contrast, a hypothetical low-abundance or normally unexpressed miRNA that through an oncogenetic event becomes highly expressed in cancer cells could make an excellent biomarker. It is entirely conceivable that even a small tumor could generate enough of a rare miRNA to be detected in circulation. For good reasons, miRNAs that are not found in the majority of samples are often filtered out of analysis, however. In some studies, these potential markers may have been overlooked. Interestingly, Leidner et al. discarded common blood miRNAs before analysis, effectively focusing on just these rare miRNAs (27). The possibility of reanalysis and discovery is another argument for public availability of raw and normalized data. However, this category of biomarker might not be found in a sizable fraction of patients.
In summary, the contribution of neoplastic cells to total circulating extracellular miRNA is in most cases relatively small and often undetectable, such that any confirmed changes in circulation must be traced instead to responses to the presence of cancer. Downregulated circulating miRNAs in particular cannot be well explained by dysregulation in the tumor. Furthermore, the contribution of tumors is correspondingly more meager in early stages of cancer, just when reliable results of a liquid biopsy might be most desirable. Effects of the tumor on surrounding tissue or broader systems, or responses of, e.g., the immune system to the presence of the tumor are more plausible explanations for the appearance of altered miRNA profiles in circulation. These "danger signals" are likely nonspecific, in both cancer and non-neoplastic diseases.
POTENTIAL ADVANCES: EXTRACELLULAR VESICLE FRACTIONATION AND SAMPLING PROTOCOLS
If the goal is to sense the tumor itself remotely, narrowing the search field will be necessary; in essence, translating into reality the sole source assumption in the illustration above. Rather than whole plasma or serum, it would be useful to examine only those miRNAs released from the tumor, or from the tumor and its surrounding tissue. There are 2 complementary approaches that might help to narrow the search.
BLOOD PRODUCT FRACTIONATION
How is miRNA carried and protected in the blood? Extracellular miRNAs are associated abundantly with Argonaute complexes (7, 8, 34), which may be released nonspecifically. Lipoprotein particles have also been described as carriers, although their significance remains unclear (35). Finally, a minority of extracellular miRNA is associated with extracellular vesicles (EVs) (1, 8, 36). Of these 3 classes of carrier, EVs, which arise from plasma or internal membranes of progenitor cells (37), are the most likely to permit assessment of cellular source on the basis of surface markers, and to be isolated and purified by means of these markers. In this way, a hypothetical marker, assuming sufficient surface occupancy on the membranes of the progenitor cell and the departing vesicle, would provide enrichment opportunities. An abundant central nervous system (CNS) marker, for example, might allow isolation of CNS-origin EVs from the deafening background of more abundant EVs and other complexes.
To be sure, the EV as carrier of small RNA is not a novel concept. For several decades, it has been known that small RNAs are released in retroviral particles, which could rightly be considered a type of EV. The most abundant and first recognized of these RNAs were cellular transfer RNA molecules, which serve as primers for retroviral cDNA synthesis (38). That is, the retrovirus carries these molecules to new host cells and subverts them for a specific function. In terms of miRNAs, the first report of circulating miRNAs described their insensitivity to 0.2-[micro]m filtration and presaged the need for ultracentrifugation studies for further characterization (2). Later in 2008, Skog et al. recognized the potential of extracellular vesicles to contain biomarkers of glioblastoma (39). Taylor and Gercel-Taylor reported epithelial cell adhesion molecule (EpCAM)-positive EVs as carriers of RNA in blood of ovarian cancer patients (40). Strikingly, cancer stages were segregated perfectly (i.e., with no overlap between groups) by protein concentration of EpCAM-positive EVs, suggesting EV protein, if verified, as a powerful marker that might eclipse RNA. Recent work has also indicated that the concentration of RNA in EVs is quite low (41). Nevertheless, these studies advanced the concept that characteristics of EV populations, including RNA content, could diagnose cancer and differentiate between cancer stages. Despite the recognition of EV miRNA since the start of the rush to identify circulating miRNA, though, relatively few studies have examined this or other potentially informative fractions of plasma, serum, and other biofluids.
For some organ systems and biofluids, sampling could be performed before EVs of interest are diluted in the bloodstream or other spaces. Li et al. (42) recently revealed the possibilities of local sampling combined with EV isolation for bile. Just as bile EVs would be highly diluted after release into the small intestine, tumor EVs exiting the tissue will be quickly diluted in blood. Similarly, finding brain EVs in cerebrospinal fluid might be preferable to searching for the same vesicles in the peripheral blood. Whether or not tissue-specific EV can be isolated by surface markers, local sampling, although perhaps challenging to standardize and control for, could provide a relatively undiluted version of the miRNA releasate from an organ. The relatively invasive nature of these techniques and focus on specific organs would preclude any use in routine screening. However, the locoregional sampling approach might allow collection of markers for response to therapy in advanced disease.
Challenges, Opportunities, and Recommendations
PROFILING OF UNFRACTIONATED PLASMA OR SERUM
For additional profiling of unfractionated plasma or serum, the foremost concerns are to establish standardized sampling and processing protocols (43-46); account for the effects of RNA extraction methods (9, 47, 48), profiling platforms (49, 50), and analysis (46); and develop and adhere to norms for reporting of methods and data (12, 51-54). These issues have been addressed adequately elsewhere and also apply to other biofluids and more-focused studies on EVs. To this list, I would add, though, that investigators should carefully consider whether single-miRNA studies, motivated by tumor expression findings, remain useful and have sufficient theoretical underpinning to justify use of precious resources including patient samples. It may be that more comprehensive profiling, not less, is needed to assess the utility of circulating miRNA. Perhaps improvements in technology and its accessibility will allow a renaissance of circulating miRNA studies, similar to what has occurred in the field of circulating DNA biomarkers, once also prone to conflicting results.
ISOLATION OF EVs
For isolation of EVs, much uncharted territory lies ahead. The challenges of comparing studies will be heightened while different sample processing and EV isolation methods abound (55, 56). Even the nomenclature of EVs is unsettled (57): is an "exosome" defined by size, biogenesis, surface markers, or a combination of features? And what, if any, are the distinguishing features of an exosome and a "microvesicle" of similar size after release from the cell? These questions have prompted some individuals in the scientific community to gather the various proposed types under the "extracellular vesicle" term.
ACHIEVING AND ASSESSING EV PURITY
Whatever the nomenclature, EVs and their subsets will be useful for narrowing the search for diagnostic and prognostic miRNAs only if they can be isolated reproducibly and to high levels of purity while retaining sufficient material for downstream analyses. In this regard, Webber and Clayton were right to ask, "How pure are your vesicles?" (58). Against a backdrop of gold-standard techniques that require ultracentrifugation and time-consuming gradients (59, 60) but may themselves benefit from careful optimization (61), numerous commercial kits for isolation and purification of EVs have been developed. These kits may be useful, but they should not be relied on off-the-shelf without rigorous comparison to legacy techniques, particularly since some appear to consist of nonspecific precipitation reagents. In performing comparisons, the purity of EV subpopulations is often imputed from detection of proteins thought to be enriched in exosomes. However, this approach does not establish purity, only the presence of a target protein. Isolation of extracellular vesicles should not be confused with total number of detected particles, total recovered protein, detection of presumed target proteins, or detection of miRNAs or other RNA species. A combined approach to assessing purity was recently recommended (58). Development and use of vesicle standards that can be spiked into EV preparations at different stages of purification will also be important.
The issues of specificity and reproducibility that affect whole biofluid research will also apply to EVs unless reliable enrichment of organ-, cell-, or disease-specific vesicles occurs. A major focus of upcoming research must be the development of these techniques. Along the way, we must also answer pressing questions. For example, what percentage of EVs in circulation derives from tissue? To what extent can EVs be isolated by tissue or cell type of origin with antibody-based methods? Although flow cytometry can reliably separate cell types on the basis of surface markers, the surface occupancy of traditional markers must be considered when so little surface area is present as on a small EV. How many molecules per particle are needed to identify or immunoprecipitate EV? In terms of size-based separation, there is evidence that in cancer studies, characteristic vesicles may be discarded in standard stepped centrifugation protocols. These large vesicles, dubbed "large oncosomes," approach the size of platelets and are both rich in potential biomarkers and preferentially released by cancer cells (62). Finally, assuming efficiency of EV isolation, can enough RNA be reliably extracted from a small minority of total EVs in clinically relevant fluid volumes to support representative RNA profiling? In the era of single-cell analysis, it is important to remember that the volume of an average-sized vesicle is 6-7 orders of magnitude smaller than that of its parent cell.
OPTIMIZATION, STANDARDIZATION, AND THE NOVELTY-DRIVEN SCIENTIFIC FUNDING MODEL
In the US, the White House Office of Science and Technology Policy recently issued a request for information regarding reproducibility initiatives in science (63). Considering the positive effects on biomarker studies and beyond, it might be hoped that funding bodies will take steps to complement the current emphasis on novelty in awarding resources with increased funds and programs for reproduction studies. One approach would be to make assessment of methods and replication and reproduction of results a part of grant review, asking the study section to rank positively applications that include methods comparisons, reproduction of existing findings, or provisions for independent verification of anticipated results as aims or subaims of the proposed project. Only the availability of funds will ensure that rigorous foundations are laid for reproducible biomarker findings.
To date, circulating miRNA research has not resulted in highly specific, validated disease markers. Focusing on extracellular vesicles and advanced sampling and isolation techniques may help to maximize the likelihood that useful markers will be validated in the future. However, any confirmation that occurs--whether in whole biofluid or EV studies--will stem from careful optimization and standardization of conditions. These research questions may not be considered entirely novel or innovative, but they must be addressed if EV-based miRNA profiling is to succeed in translating basic research into the clinic.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.
(1.) Valadi H, Ekstrom K, Bossios A, Sjostrand M, Lee JJ, Lotvall JO. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 2007; 9:654 -9.
(2.) Chim SS, Shing TK, Hung EC, Leung TY, Lau TK, Chiu RW, et al. Detection and characterization of placental microRNAs in maternal plasma. Clin Chem 2008; 54: 482-90.
(3.) Lawrie CH, Gal S, Dunlop HM, Pushkaran B, Liggins AP, Pulford K, et al. Detection of elevated levels of tumour-associated microRNAs in serum of patients with diffuse large B-cell lymphoma. Br J Haematol 2008; 141: 672-5.
(4.) Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, Pogosova-Agadjanyan EL, et al. Circulating microRNAs as stable blood-based markers for cancer detection. Proc Natl Acad Sci U S A 2008; 105:10513-8.
(5.) Chen X, Ba Y, Ma L, Cai X, Yin Y, Wang K, et al. Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and other diseases. Cell Res 2008; 18:997-1006.
(6.) Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell 2009; 136:215-33.
(7.) Turchinovich A, Weiz L, Langheinz A, Burwinkel B. Characterization of extracellular circulating microRNA. Nucleic Acids Res 2011; 39:7223-33.
(8.) Arroyo JD, Chevillet JR, Kroh EM, Ruf IK, Pritchard CC, Gibson DF, et al. Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma. Proc Natl Acad Sci U S A 2011; 108: 5003-8.
(9.) Moldovan L, Batte KE, Trgovcich J, Wisler J, Marsh CB, Piper M. Methodological challenges in utilizing miRNAs as circulating biomarkers. J Cell Mol Med 2014; 18:371-90.
(10.) Pritchard CC, Cheng HH, Tewari M. MicroRNA profiling: approaches and considerations. Nat Rev Genet 2012; 13:358-69.
(11.) Watson AK, Witwer KW. Do platform-specific factors explain microRNA profiling disparities? Clin Chem 2012; 58:472- 4.
(12.) Kirschner MB, van Zandwijk N, Reid G. Cell-free microRNAs: potential biomarkers in need of standardized reporting. Front Genet 2013; 4:56.
(13.) Li H, Ge Q, Guo L, Lu Z. Maternal plasma miRNAs expression in preeclamptic pregnancies. Biomed Res Int 2013:970265.
(14.) Wang G, Tam LS, Li EK, Kwan BC, Chow KM, Luk CC, et al. Serum and urinary free microRNA level in patients with systemic lupus erythematosus. Lupus 2011; 20: 493-500.
(15.) Diaz T, Tejero R, Moreno I, Ferrer G, Cordeiro A, Artells R, et al. Role of miR-200 family members in survival of colorectal cancer patients treated with fluoropyrimidines. J Surg Oncol 2014; 109:676 - 83.
(16.) Selth LA, Townley SL, Bert AG, Stricker PD, Sutherland PD, Horvath LG, et al. Circulating microRNAs predict biochemical recurrence in prostate cancer patients. Br J Cancer 2013; 109:641-50.
(17.) Bryant RJ, Pawlowski T, Catto JW, Marsden G, Vessella RL, Rhees B, et al. Changes in circulating microRNA levels associated with prostate cancer. Br J Cancer 2012; 106:768-74.
(18.) Gonzales JC, Fink LM, Goodman OB Jr, Symanowski JT, Vogelzang NJ, Ward DC. Comparison of circulating microRNA 141 to circulating tumor cells, lactate dehydrogenase, and prostate-specific antigen for determining treatment response in patients with metastatic prostate cancer. Clin Genitourin Cancer 2011; 9:39-45.
(19.) Yeh TS, Wang F, Chen TC, Yeh CN, Yu MC, Jan YY, et al. Expression profile of microRNA-200 family in hepatocellular carcinoma with bile duct tumor thrombus. Ann Surg 2014; 259:346 -54.
(20.) Cheng H, Zhang L, Cogdell DE, Zheng H, Schetter AJ, Nykter M, et al. Circulating plasma MiR-141 is a novel biomarker for metastatic colon cancer and predicts poor prognosis. PLoS One 2011; 6:e17745.
(21.) Scheffer AR, Holdenrieder S, Kristiansen G, von Ruecker A, Muller SC, Ellinger J. Circulating microRNAs in serum: novel biomarkers for patients with bladder cancer? World J Urol 2014; 32:353-8.
(22.) Westermann AM, Schmidt D, Holdenrieder S, Moritz R, Semjonow A, Schmidt M, et al. Serum microRNAs as biomarkers in patients undergoing prostate biopsy: results from a prospective multi-center study. Anticancer Res 2014; 34:665-9.
(23.) Yaman Agaoglu F, Kovancilar M, Dizdar Y, Darendeliler E, Holdenrieder S, Dalay N, et al. Investigation of miR-21, miR-141, and miR-221 in blood circulation of patients with prostate cancer. Tumour Biol 2011; 32: 583-8.
(24.) Brunet Vega A, Pericay C, Moya I, Ferrer A, Dotor E, Pisa A, et al. microRNA expression profile in stage III colorectal cancer: circulating miR-18a and miR-29a as promising biomarkers. Oncol Rep 2013; 30:320-6.
(25.) Egidi MG, Cochetti G, Serva MR, Guelfi G, Zampini D, Mechelli L, et al. Circulating microRNAs and kallikreins before and after radical prostatectomy: are they really prostate cancer markers? Biomed Res Int 2013: 241780.
(26.) Haider BA, Baras AS, McCall MN, Hertel JA, Cornish TC, Halushka MK. A critical evaluation of microRNA biomarkers in non-neoplastic disease. PLoS One 2014; 9: e89565.
(27.) Leidner RS, Li L, Thompson CL. Dampening enthusiasm for circulating microRNA in breast cancer. PLoS One 2013; 8:e57841.
(28.) Schwarzenbach H, Milde-Langosch K, Steinbach B, Muller V, Pantel K. Diagnostic potential of PTEN-targeting miR-214 in the blood of breast cancer patients. Breast Cancer Res Treat 2012; 134:933-41.
(29.) Godfrey AC, Xu Z, Weinberg CR, Getts RC, Wade PA, DeRoo LA, et al. Serum microRNA expression as an early marker for breast cancer risk in prospectively collected samples from the Sister Study cohort. Breast Cancer Res 2013; 15:R42.
(30.) Croce CM. Causes and consequences of microRNA dysregulation in cancer. Nat Rev Genet 2009; 10:704-14.
(31.) Duttagupta R, Jiang R, Gollub J, Getts RC, Jones KW. Impact of cellular miRNAs on circulating miRNA biomarker signatures. PLoS One 2011; 6:e20769.
(32.) Gyorgy B, Szabo TG, Pasztoi M, Pal Z, Misjak P, Aradi B, et al. Membrane vesicles, current state-of-the-art: emerging role of extracellular vesicles. Cell Mol Life Sci 2011; 68:2667-88.
(33.) Yan LX, Huang XF, Shao Q, Huang MY, Deng L, Wu QL, et al. MicroRNA miR-21 overexpression in human breast cancer is associated with advanced clinical stage, lymph node metastasis and patient poor prognosis. RNA 2008; 14:2348-60.
(34.) Turchinovich A, Burwinkel B. Distinct AGO1 and AGO2 associated miRNA profiles in human cells and blood plasma. RNA Biol 2012; 9:1066-75.
(35.) Wagner J, Riwanto M, Besler C, Knau A, Fichtlscherer S, Roxe T, et al. Characterization of levels and cellular transfer of circulating lipoprotein-bound microRNAs. Arterioscler Thromb Vasc Biol 2013; 33:1392-400.
(36.) Hunter MP, Ismail N, Zhang X, Aguda BD, Lee EJ, Yu L, et al. Detection of microRNA expression in human peripheral blood microvesicles. PLoS One 2008; 3:e3694.
(37.) Raposo G, Stoorvogel W. Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol 2013; 200:373-83.
(38.) Sawyer RC, Harada F, Dahlberg JE. Virion-associated RNA primer for Rous sarcoma virus DNA synthesis: isolation from uninfected cells. J Virol 1974; 13:1302-11.
(39.) Skog J, Wurdinger T, van Rijn S, Meijer DH, Gainche L, Sena-Esteves M, et al. Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers. Nat Cell Biol 2008; 10:1470-6.
(40.) Taylor DD, Gercel-Taylor C. MicroRNA signatures of tumor-derived exosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol 2008; 110:13-21.
(41.) Chevillet JR, Kang Q, Ruf IK, Briggs HA, Vojtech LN, Hughes SM, et al. Quantitative and stoichiometric analysis of the microRNA content of exosomes. Proc Natl Acad Sci U S A 2014; 111:14888-93.
(42.) Li L, Masica D, Ishida M, Tomuleasa C, Umegaki S, Kalloo AN, et al. Human bile contains microRNA-laden extracellular vesicles that can be used for cholangiocarcinoma diagnosis. Hepatology 2014; 60:896-907.
(43.) Cheng HH, Yi HS, Kim Y, Kroh EM, Chien JW, Eaton KD, et al. Plasma processing conditions substantially influence circulating microRNA biomarker levels. PLoS One 2013; 8:e64795.
(44.) Pritchard CC, Kroh E, Wood B, Arroyo JD, Dougherty KJ, Miyaji MM, et al. Blood cell origin of circulating microRNAs: a cautionary note for cancer biomarker studies. Cancer Prev Res (Phila) 2012; 5:492-7.
(45.) Kim DJ, Linnstaedt S, Palma J, Park JC, Ntrivalas E, Kwak-Kim JY, et al. Plasma components affect accuracy of circulating cancer-related microRNA quantitation. J Mol Diagn 2012; 14:71-80.
(46.) McDonald JS, Milosevic D, Reddi HV, Grebe SK, Algeciras-Schimnich A. Analysis of circulating microRNA: preanalytical and analytical challenges. Clin Chem 2011; 57:833-40.
(47.) Kim YK, Yeo J, Kim B, Ha M, Kim VN. Short structured RNAs with low GC content are selectively lost during extraction from a small number of cells. Mol Cell 2012; 46:893-5.
(48.) McAlexander MA, Phillips MJ, Witwer KW. Comparison of methods for miRNA extraction from plasma and quantitative recovery of RNA from cerebrospinal fluid. Front Genet 2013; 4:83.
(49.) Leshkowitz D, Horn-Saban S, Parmet Y, Feldmesser E. Differences in microRNA detection levels are technology and sequence dependent. RNA 2013; 19:527-38.
(50.) Mestdagh P, Hartmann N, Baeriswyl L, Andreasen D, Bernard N, Chen C, et al. Evaluation of quantitative miRNA expression platforms in the microRNA quality control (miRQC) study. Nat Methods 2014; 11:809 -15.
(51.) Witwer KW. Data submission and quality in microRNA biomarker profiling. Clin Chem 2013; 59: 392-400.
(52.) Baggerly K. More Data, Please! Clin Chem 2013; 59: 459-61.
(53.) Bustin SA, Benes V, Garson JA, Hellemans J, Huggett J, Kubista M, et al. The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem 2009; 55:611-22.
(54.) Ioannidis JP, Allison DB, Ball CA, Coulibaly I, Cui X, Culhane AC, et al. Repeatability of published microarray gene expression analyses. Nat Genet 2009; 41:149-55.
(55.) Witwer KW, Buzas EI, Bemis LT, Bora A, Lasser C, Lotvall J, et al. Standardization of sample collection, isolation and analysis methods in extracellular vesicle research. J Extracell Vesicles 2013; 2:doi:10.3402/jev.v2i0.20360.
(56.) Yuana Y, Bertina RM, Osanto S. Pre-analytical and analytical issues in the analysis of blood microparticles. Thromb Haemost 2011; 105:396-408.
(57.) Gould SJ, Raposo G. As we wait: coping with an imperfect nomenclature for extracellular vesicles. J Extracell Vesicles 2013; 2:doi:10.3402/jev.v2i0.20389.
(58.) Webber J, Clayton A. How pure are your vesicles? J Extracell Vesicles 2013; 2:doi:10.3402/jev.v2i0.19861.
(59.) Thery C, Amigorena S, Raposo G, Clayton A. Isolation and characterization of exosomes from cell culture supernatants and biological fluids. Curr Protoc Cell Biol 2006; Chapter 3:Unit 3.22.
(60.) Palma J, Yaddanapudi SC, Pigati L, Havens MA, Jeong S, Weiner GA, et al. MicroRNAs are exported from malignant cells in customized particles. Nucleic Acids Res 2012; 40:9125-38.
(61.) Cvjetkovic A, Lotvall J, Lasser C. The influence of rotor type and centrifugation time on the yield and purity of extracellular vesicles. J Extracell Vesicles 2014; 3:doi: 10.3402/jev.v3.23111.
(62.) Morello M, Minciacchi VR, de Candia P, Yang J, Posadas E, Kim H, et al. Large oncosomes mediate intercellular transfer of functional microRNA. Cell Cycle 2013; 12: 3526-36.
(63.) Science and Technology Policy Office, National Economic Council. Strategy for American innovation. Fed Register 2014; 79:44064-8.
Kenneth W. Witwer  *
 Department of Molecular and Comparative Pathoblology, The Johns Hopkins University School of Medicine, Baltimore, MD.
* Address correspondence to the author at: Department of Molecular and Comparative Pathobiology, The Johns Hopkins University School of Medicine, 733 North Broadway, Miller Research Bldg, Room 829, Baltimore, MD 21205. Fax 410-955-9823; e-mail firstname.lastname@example.org.
Received September 12, 2014; accepted October 30, 2014.
Previously published online at DOI: 10.1373/clinchem.2014.221341
 Nonstandard abbreviations: miRNA, microRNA; BC, breast cancer; qPCR, quantitative PCR; EV, extracellular vesicle; CNS, central nervous system; EpCAM, epithelial cell adhesion molecule.
Fig. 1. Little overlap between reported miRNA BC biomarkers. Low fold change 8 Concordant 10 Related references 9 One references 100 Discordant 25 Of 143 circulating miRNAs reported to be differentially regulated, 100 were supported by just 1 reference; 25 others had discordant results across publications. Of the remaining 18 miRNAs, 8 had fold changes too low to be confirmed. Of 10 concordant results, 9 were supported entirely by publications from the same institution and had authors in common. Note: Table made from pie chart.
|Printer friendly Cite/link Email Feedback|
|Author:||Witwer, Kenneth W.|
|Date:||Jan 1, 2015|
|Next Article:||Sorting the wheat from the chaff in familial hypercholesterolemia.|