Oncogene www.nature.

com/onc

REVIEW ARTICLE OPEN

MicroRNAs: circulating biomarkers for the early detection of

imperceptible cancers via biosensor and machine-learning
advances
1✉
Gavin A. D. Metcalf

© The Author(s) 2024

This review explores the topic of microRNAs (miRNAs) for improved early detection of imperceptible cancers, with potential to
advance precision medicine and improve patient outcomes. Historical research exploring miRNA’s role in cancer detection
collectively revealed initial hurdles in identifying specific miRNA signatures for early-stage and difficult-to-detect cancers. Early
studies faced challenges in establishing robust biomarker panels and overcoming the heterogeneity of cancer types. Despite this,
recent developments have supported the potential of miRNAs as sensitive and specific biomarkers for early cancer detection as well
as having demonstrated remarkable potential as diagnostic tools for imperceptible cancers, such as those with elusive symptoms or
challenging diagnostic criteria. This review discusses the advent of high-throughput technologies that have enabled
1234567890();,:

comprehensive detection and profiling of unique miRNA signatures associated with early-stage cancers. Furthermore,
advancements in bioinformatics and machine-learning techniques are considered, exploring the integration of multi-omics data
which have potential to enhance both the accuracy and reliability of miRNA-based cancer detection assays. Finally, perspectives on
the continuing development on technologies as well as discussion around challenges that remain, such as the need for
standardised protocols and addressing the complex interplay of miRNAs in cancer biology are conferred.

Oncogene (2024) 43:2135–2142; https://doi.org/10.1038/s41388-024-03076-3

CHALLENGES IN EARLY DETECTION OF DIFFICULT-TO-DETECT screening tools applicable to high-risk groups, high false positive
CANCERS rates in common imaging techniques along with invasive
Current trajectories suggest that global cancer cases will reach diagnostic biopsy procedures posing significant risks, especially
over 35 million by 2050 [1] and that almost half of all cancer cases in light of the recent pandemic [9, 10]. The early detection of Liver
that were diagnosed in England in 2018 were at advanced stages cancers is complicated by cirrhosis masking key diagnostic
(3 and 4) [2]. This highlights the urgency of early detection in symptoms, imaging limitations when considering small lesions
order to attempt to reduce the burden this will inevitably cause on due to poor arterial phase hyperenhancement of contrast agents
patients, their social support network, healthcare systems, as well and the challenges of liver biopsies [11, 12]. CNS tumours often
as local and global economies. present with subtle or non-specific symptoms due to their
However, despite this pressing need, early detection poses location, imaging struggles to differentiate between benign and
many challenges. The lack of systemic effects due to the small size malignant tumours and biopsy procedures carry high risks [13, 14].
of primary tumours, or metastases, in early stages of development These reasons, combined with technological advances in
often leads to the asymptomatic nature of disease progression. In biosensor and machine-learning development, have led to
addition, the reported lack of specific and sensitive screening tests heightened interest in the area of clinically pertinent microRNA
that follow standardised testing parameters collectively contribute (miRNA) biomarkers obtained from, and stable within, minimally
to delayed accurate detection. invasive biofluids, such as blood, saliva, and urine.
Certain well-documented cancers present unique challenges in
diagnosis, including pancreatic, non-small cell lung cancer
(NSCLC), liver, and central nervous system (CNS) tumours. All MICRORNAS
aforementioned cancers share the challenge of being asympto- MicroRNAs (miRNAs) are short, non-coding RNA molecules that are
matic in early stages of development, and often present with typically 18–25 nucleotides in length. The first miRNA, Lin-4, was
vague and non-specific symptoms as the disease advances [3–7]. discovered in Caenorhabditis elegans (C. elegans) in 1993 by
The deep-seated retroperitoneal location of the pancreas fre- Ambros and colleagues. Initial results suggested that the Lin-4
quently hinders early detection of cancer through common gene did not code for a protein but, instead, gave rise to a short
screening methods. In addition to this, there are no widely RNA sequence of 22 nucleotides in length that was observed to
adopted reliable screening tests [8]. NSCLC has limited effective interact with the 3′ untranslated region (UTR) of the Lin-4

School of Life Sciences, Faculty of Science and Engineering, Anglia Ruskin University, Cambridge, UK. ✉email: [email protected]
1

Received: 17 April 2024 Revised: 24 May 2024 Accepted: 29 May 2024

Published online: 5 June 2024
 G.A.D. Metcalf
2136

Fig. 1 MiRNA excretion mechanisms and extracellular survival. MiRNAs are actively and passively released by cells via various approaches,
including multivesicular bodies (MVB) and cellular excretion via exosomes, microvesicle formation achieved via membrane shedding, as well
as association with AGO-2 and HDL.

messenger RNA (mRNA), via Watson-Crick base pairing, repressing and cleaves the pre-miRNA generating a 19 to 23 nucleotide
its expression and initiating the larval stages of development [15]. duplex structure with a mature (master) and complementary
Research into miRNAs continued at a somewhat slow pace and in (passenger) strand. The mature miRNA is then integrated into the
2000 another pivotal miRNA was reported by Reinhart and RNA-induced silencing complex (RISC) chiefly consisting of
colleagues. Let-7 was identified as a key player in the timing of Argonaute Proteins 1–4 (AGO 1–4), Dicer, and TAR RNA-binding
larval development in the same model system, C. elegans. Notably, protein (TRBP) [22], which guides the RISC to specific mRNA
Let-7 is evolutionarily conserved across diverse species, including targets, leading to either translational repression or mRNA
Homo Sapiens, indicating a fundamental role in biological degradation [23].
processes [16]. The discoveries of Lin-4 and Let-7 collectively
reshaped the scientific landscape, leading to a new field of Circulating microRNAs
research with substantial potential to impact development, health, Freely circulating nucleic acids in blood were first denoted about
and disease across a wide range of species. Upon consideration of 60 years ago [24, 25], with subsequent research reporting that
cancer, a pioneering study by Calin and colleagues shed light on tumour-specific DNA and RNA were frequently found in the
the aberrant expression of specific miRNAs in Chronic Lympho- plasma of cancer patients [26, 27]. Historically it was a common
cytic Leukaemia (CLL). In particular, miR-15a and miR-16-1 genes, belief within the scientific community that RNA molecules would
often down-regulated or deleted in approximately 68% of CLL not be a suitable biomarker within blood samples due to the
cases, were found to target the B-cell lymphoma 2 (BCL-2) gene, presence of endogenous nucleases within plasma [28], however
providing a mechanistic link between their downregulation or upon the discovery of miRNAs within fixed tissues [29] this idea
deletion and the apoptotic resistance commonly observed in CLL was rapidly dismissed. This was further explored by Chen et al. [30]
cells. Thus, suggesting that the loss of these miRNAs contributes who identified a set of miRNAs that were consistently present and
to the pathogenicity of CLL, via promotion of cell survival [17]. The stable within serum samples. This groundbreaking work laid the
findings resulted in a growing field of research, contributing to a foundation for the exploration of circulating miRNAs as non-
broader understanding of the roles of miRNAs in cancer biology, invasive biomarkers for various diseases, including cancer.
offering potential avenues for the development of miRNA-based The release of miRNAs from cells into extracellular environ-
therapeutics, as well as pioneering efforts to identify miRNA ments, including blood, is believed to result from a variety of
signatures that could be used to distinguish between different mechanisms. These include active secretion via exosomes,
cancer types. microvesicle release, and protein-mediated export (such as high-
MiRNAs are coded for in both exons and introns of genes or density lipoproteins and AGO2) [31–34] (Fig. 1). The stability of
non-coding RNA transcripts [18] suggesting a link to host-gene circulating miRNAs is enhanced by their association with various
promoters. MiRNA-coding sequences are first transcribed by RNA carriers, as described, protecting them from nuclease- and
polymerase II (Pol II) into long primary transcripts (pri-mRNAs) [19]. protease-facilitated degradation [35, 36] as well as stabilising the
Pri-mRNAs are then capped and polyadenylated [20], and molecules when exposed to sample processing conditions such as
subsequently cleaved in the nucleus by Drosha and Pasha freeze-thaw cycles [37].
resulting in a hairpin-looped precursor nucleotide (pre-miRNA) The first reported case of using miRNAs as candidate
of approximately 60–75 nucleotides in length. Pre-miRNA is biomarkers for cancer was published by Lawrie and colleagues
actively transported out of the nucleus into the cytoplasm by in 2008 whereby blood serum concentrations of a panel of three
Exportin 5 [21] whereupon another RNAse III enzyme, Dicer, binds miRNAs (miR-155, miR-210 and miR-21) were compared in

Oncogene (2024) 43:2135 – 2142

 G.A.D. Metcalf
2137
patients with diffuse large B-cell lymphoma compared to healthy discriminating between early NSCLC patients (TNM stages I, II and
control sera [38]. This pioneering publication demonstrated for the IIIA) and healthy individuals.
first time that miRNAs are clearly detectable in serum samples and Liver cancer, most commonly Hepatocellular carcinoma (HCC), is
that miRNAs have potential for clinical approaches in cancer commonly diagnosed at an advanced stage with poor prognosis
detection via minimally invasive means of sample sourcing. due to limited therapeutic options. Several studies have reported
the promise of miRNAs as biomarkers for early diagnosis of HCC
opening up the opportunity for a greater volume of treatment
RECENT DEVELOPMENTS OF MICRORNAS AS BIOMARKERS options and thus heightened prognostic outcomes. Amr et al. [46]
FOR EARLY CANCER DETECTION OF DIFFICULT-TO-DIAGNOSE explored the promise of two previously reported miRNAs (miR-122
CANCERS and miR-224) as biomarkers for early stage HCC diagnosis, in
As outlined, the emergence of miRNAs as promising biomarkers comparison to the conventional serum marker, alpha-fetoprotein
for early cancer detection has been a key development in the field (AFP), which has a modest accuracy of detecting early stage HCC
of molecular biology. Further to this, particular miRNA signatures [47]. Blood plasma samples of patients with HCC, preceded by
have been suggested to hold great impact when regarding chronic HCV infection (n = 40) were compared to non-HCC
efficient diagnosis of imperceptible cancers at an early stage. We controls, those with Chronic hepatitis C (CHC) (n = 40) and
shall go on to explore the identification and clinical validation of disease-free individuals (n = 20). RT-qPCR assays specific for hsa-
miRNA signatures, along with technological advances in biosensor miR-122 and hsa-miR-224 were run to assess the expression levels
development and machine learning used to detect, analyse, and within all samples obtained. Chief findings reported mean plasma
report key miRNA signatures. values of miR-122 and miR-224 were significantly lower and
higher, respectively, within HCC samples in comparison to HCC-
Identification and clinical validation of miRNA signatures free groups. Furthermore, results suggested that miR-122 and miR-
Pioneering research by Lu et al. [39] conducted a screen of 224 could predict HCC with comparable sensitivity (87.5% and
expression profiles of miRNAs from more than 300 patient-derived 92.5%, respectively), specificity (95 and 90%, respectively), and
biosamples, including multiple cancers. They were able to accuracy (0.96 and 0.94, respectively). In comparison, AFP
successfully classify poorly differentiated tumours using miRNA sensitivity and specificity results were much poorer at 57.5 and
profiles alone, which when compared to messenger RNA (mRNA) 95%, respectively. Additional studies further support the findings
counterparts which were highly inaccurate, emphasised the of reduced miR-122 and elevated miR-224 expression profiles as
potential of miRNA profiling in cancer diagnosis. Given the promising biomarkers for the early diagnosis of HCC [48–50].
growing promise that miRNA’s serve as biosample-based diag- Despite overall cancer-derived mortality declining over the past
nostic tools, the exploration of their application within the field of two decades, CNS malignancies still contribute to high mortality
imperceptible cancers has been growing steadily. rates [51]. This is believed to be largely due to a lack of accurate
In 2024, Shi et al. [40] explored miRNAs that could serve as mass screening methods for early detection often hindered by
biomarkers for the early detection of pancreatic cancer in patients factors such as the blood–brain–barrier (BBB). Cerebrospinal fluid
presenting with chronic pancreatitis. The study successfully (CSF) due to its function is in direct contact with any possible
implemented a robust rank aggregation (RRA) machine-learning pathology within the CNS and is a model biofluid source for
algorithm to aid in early pancreatic diagnosis via screening the biomarker detection, with relatively easy sourcing via a lumbar
expression profile of candidate miRNA biomarkers. Serum-derived puncture. MiRNAs have been widely reported to be abundant
miR-205-5p was identified as a promising predictor candidate that within CSF [52], with previous studies suggesting that dysregu-
was observed to discern between patients with pancreatitis and lated miRNA expression is associated with malignant CNS tumours
pancreatic cancer, with reported accuracy rates of 91.5%. [53, 54]. Promising results were reported in a small-scale study
Furthermore, results demonstrated that miR-205-5p expression conducted by Shalaby et al. [55], whereby the promise of miRNAs
could be used as a predictive marker for more advanced disease, were explored as novel biomarkers for medulloblastoma (MB)
with high-expression rates matching with poorer prognosis within detection. The study explored the presence of extracellular
R1/2 resection margins of tumour specimens (indicative of miRNAs within both cell culture medium as well as CSF samples
residual tumour) compared to R0 resection counterparts (resection obtained from MB and non-tumour control patients. Microarray
for cure or complete remission). analysis identified 268 high-expression miRNA profiles and 6 low-
NSCLC is another well-reported cancer that is difficult to expression miRNA profiles, in comparison to controls. Selected
diagnose in early stages, often due to asymptomatic presentation, miRNAs were chosen to validate these findings (including miR-
erroneous radiographic interpretation, or symptoms being 486-3p, miR-572, miR-3918, miR-4476, miR-615, miR-1290, miR-
wrongly attributed to other chronic respiratory conditions such 152a, miR-125b, and miR-1298) via RTq-PCR. The analysis
as chronic obstructive pulmonary disease (COPD) [41]. A study by confirmed the heightened presence of 4 key miRNAs (miR-1290,
Dong et al. [42] analysed the miRNA profiles of plasma samples miR-125a, miR-125b and miR-1298) within CSF from test samples.
derived from both NSCLC patients and healthy controls via a This pioneering, albeit small, study identified a small selection of
miRNA microarray. Real time quantitative RT-q-PCR was then used key miRNAs that were enriched within CSF samples from MB
to assess the expression levels of 11 different miRNAs that were patients, which contributed to additional larger-scale studies
upregulated. Three particular miRNAs, miR-1247-5p, miR-301b-3p being conducted. One such study was conducted by Kopkova
and miR-105-5p, were demonstrated to accurately distinguish et al. [56], which involved a two-phase (discovery and validation)
between patients with NSCLC and healthy individuals, with approach. During the initial discovery phase 89 CSF samples, taken
corresponding AUCs being reported as 0.769, 0.761, and 0.777, from patients with glioblastoma (n = 32), low-grade glioma
respectively. Such findings are supported by previous studies (n = 14), meningioma (n = 11), brain metastases (n = 13) and
[43, 44] which denote that miR-301b-3p is not only present at non-tumour controls (with normal-pressure hydrocephalus,
elevated rates in NSCLC, but pivotal to cancer cell characteristics n = 19), were screened via small RNAseq analysis. This identified
including proliferation, migration and invasion via targeting Rho miRNAs with altered expression when compared to controls,
GTPase activating protein, DLC1. Further analyses conducted by ultimately leading to a panel of 9 candidate miRNA biomarkers
Arab et al. [45] identified elevated miR-141 expression in plasma (let-7a, let-7b. miR-10a, miR-10b, miR-21-3p, miR-30e, miR-140,
samples from NSCLC patients and matched healthy controls, miR-196a and miR-196b) to be explored. During the validation
independent of age and sex. ROC plot analyses indicated high phase a further 126 pathology-matched CSF samples independent
sensitivity (82.7%) and specificity (98%) for miR-141 in terms of of the discovery cohort, including glioblastoma (n = 41), low-

Oncogene (2024) 43:2135 – 2142

 G.A.D. Metcalf
2138
Table 1. Circulating miRNAs as clinical biomarkers for detection of difficult-to-diagnose cancers.
miRNA profiling Test Cancer Healthy Control vs. Cancer Study
sample Type Reference
Sensitivity Specificity AUC (95% CI)
miR-20a, miR-21, miR-25, miR-99a, miR-185, miR-191 Serum Pancreatic 89% 100% 0.992 [85]
(PDAC)
miR-30c-5p, miR-let7e-5p, miR340-5p, miR-223-3p, miR- Plasma Pancreatic 87% 88% 0.920 [86]
26a-5p, miR-340-3p, miR-335-5p, miR-23b-3p, miR-142-3p, and (PDAC)
miR-200c-3p, miR-148a-3p, miR-216a-5p, miR-145-5p, miR- Serum
200b-3p, miR-143-3p, miR-34a-5p, miR-429, miR-141-3p,
miR-1260b, miR-145-3p, miR-216b-3p, miR-200a-3p, miR-
1260a, miR-217-5p
miR-93-5p, miR-339-3p, miR-425-5p, miR-425-3p Plasma Pancreatic 80% 94.7% 0.885 [87]
(PDAC)
miR-15b, miR-27b Serum Lung 100% 84% 0.980 [88]
(NSCLC)
miR-155, miR-20a, miR-25, miR-296, miR-126, miR-223, Plasma Lung 81.8% 82.9% 0.879 [89]
miR-199a, miR-24, miR-152, miR-145, miR-let7f (NSCLC)
miR-31-5p, miR-210-3p, miR21-5p Sputum Lung 85.5% 91.7% 0.913 [90]
and (NSCLC)
Plasma
miR-1247-5p, miR-301b-3p, miR-105-5p Plasma Lung 72.5% 82.2% 0.815 [42]
(NSCLC)
miR-141 Plasma Lung 96.3% 99.3% 0.972 [45]
(NSCLC)
miR-16 and miR-122 Serum Liver (HCC) 58% 84% 0.803 [91]
miR-206, miR-141-3p, miR-433-3p, miR-1228-5p, miR-199a- Serum Liver (HCC) 86% 73% 0.887 [92]
5p, miR-122-5p, miR-192-5p, miR-26a-5p
miR-4661-5p, miR-4746-5p Serum Liver (HCC) 85% 89% 0.942 [93]
miR-92-3p, miR-107, miR-3126-5p Serum Liver (HCC) 97.5% 87.8% 0.962 [94]
miR-30e, miR-140 CSF CNS (Brain 76% 75% 0.776 [56]
Tumour)
miR-15b CSF CNS 90% 94.9% 0.960 [95]
(Glioma)
miR-15b Plasma CNS 100% 100% 1.000 [96]
(Glioma)
miR-210 Serum CNS 91.3% 91.27% 0.927 [97]
(Glioma)
miR microRNA, AUC area under the curve, PDAC pancreatic ductal adenocarcinoma, NSCLC non-small cell lung cancer, HCC hepatocellular carcinoma; CSF
cerebrospinal fluid, CNS central nervous system.

grade glioma (n = 8), meningioma (n = 44), brain metastases by the scientific community as a means of quantifying miRNA,
(n = 12) and non-tumour controls (with normal-pressure hydro- after first being described in published works by Chen et al. [57]. It
cephalus, n = 21), were processed via RTq-PCR. Findings identified is now deemed as an orthogonal and gold-standard approach for
that combination patterns of 5 key miRNAs (miR-30e, miR-140, let- validating miRNA expression differences within sample sets.
7n, miR-10a and miR-21-3p) served as promising biomarker tools However, due to limitations of this approach, including RNA-
in discerning between healthy donors and cancers, with high sensitivity, specificity and cross-reactivity, and normalisation, other
sensitivity and specificity. Furthermore, the study also identified a techniques are now becoming established as validation
promising prognostic panel in glioblastoma patients, with median approaches when exploring miRNA detection and profiling. One
overall survival (OS) scores in patients with elevated levels of miR- technique, NanoString’s nCounter® microRNA assay, was reported
10b and miR-196b observed to be 9 months, compared to low- as having advantages in terms of improved sensitivity and
expression concentrations equating to median OS of 16.5 months. specificity over conventional techniques (such as RTq-PCR) within
Collective outputs from key studies, inclusive of particular a pioneering paper in 2014 by Mestdagh et al. [58], a finding that
aforementioned studies from above, can be found in Table 1, has been further supported within the literature [59, 60]. The
further determining the efficacy of circulating miRNAs as technology works without relying upon reverse transcription or
biomarkers for imperceptible cancers. amplification and is capable of highly multiplexed analysis of
samples. This is achieved via direct digital detection of RNA or
Technological biosensor advances DNA molecules of interest using colour-coded pairs of probes
Due to the nature of miRNAs being short and highly conserved, (capture and reporter) which hybridise to the target molecules.
resulting in high homology, presents challenges in detection that The high-throughput approach facilitates reliable and reproduci-
is specific and sensitive. Overcoming such challenges relies upon ble expression profiling of up to 800 genes in a single assay, with
technological advances in their detection, of which there have input biomarker molecules as low as 1 ng (RNA). Another well-
been an abundance in recent years. RTq-PCR was rapidly accepted reported sensing technology is that of Next Generation RNA

Oncogene (2024) 43:2135 – 2142

 G.A.D. Metcalf
2139
sequencing (RNAseq) as utilised in the previously discussed study cancers. This was represented as: C(n, m), where ‘m’ is a cancer
conducted by Kopkova et al. [56]. RNAseq, developed upon classic stage-specific miRNA signature and ‘n’ is a pool of 7,117 candidate
Sanger sequencing, offers a parallel sequencing by synthesis (SBS) miRNAs from 4,667 patients with 15 cancer types, including HCC,
approach that has many advantages over other sequencing obtained from The Cancer Genome Atlas (TCGA). The pan-cancer
approaches. Not only does it offer large data outputs (300 kb up to analysis of miRNA signatures collectively proposed that three
multiple terabases/run) but has high sensitivity, quantitative miRNAs (let-7i-3p, miR-362-3p and miR-3651) could discriminate
precision, as well as the ability to screen a high number of between tumour and non-tumour samples and extensively
samples in parallel (high throughput in nature) [61]. However, contributed to stage predictions amongst 8 of the 15 cancer
despite this there are some challenges associated when applying types analysed.
this technology to miRNA screening. These include misannotation Other combined machine learning and bioinformatics
of novel or underexpressed miRNA [62, 63], sequencing biases approaches have been described as a means of identifying
based on library preparation protocol variations, and extensive potential diagnostic pancreatic cancer-specific miRNA biomarkers
sample processing [64]. As a means of circumnavigating some of [68]. The study analysed serum-derived miRNA expression profiles
these challenges new technologies are being developed within from three datasets (independent) obtained from the Gene
the scientific community. Recently, Cai et al. [65] reported a Expression Omnibus (GEO) database. Collectively, three machine-
molecular probe-based system that allows for amplification-free learning algorithms (Support Vector Machine-Recursive Feature
multiplexed detection of microRNAs within unprocessed biofluid Elimination, Least Absolute Shrinkage and Selection Operator
samples. This electro-optical sensing platform functions via regression analysis, and Random Forest) identified three key
custom probes combining a DNA carrier and molecular beacon candidate miRNAs (miR-4648, miR-125b-1-3p, and miR-3201)
labelled with a reversible fluorophore and quencher in a hairpin which displayed promise as diagnostic biomarkers due to
structure which upon binding to the miRNA-target unhinges and exhibiting altered differential expression patterns. The combined
restores fluorescence. The DNA-carrier acts as a barcode with its model described exhibited notable performance and accuracy in
length resulting in differential electrical signal upon nanopore both training and validation processes, with reported AUC values
(nanopipette) detection. This technology displayed femtomolar of 0.926 and 0.935, respectively.
sensitivity thus supporting detection of sub-nanomolar concen- While the promise of such approaches is putative, limitations of
trations of miRNAs within biofluids without the need of deep learning models within the cancer field exist. Such
amplification steps, and single-base mismatch selectivity essential limitations include the lack of large phenotypically characterised
for discriminating between miRNA’s high homology due to their open-source datasets often as a result of high processing costs
short sequence length, on unprocessed blood serum samples and restricted sample availability. Furthermore, storage of tumour
obtained from Prostate Cancer patients. One limitation of this samples is typically within formalin-fixed paraffin-embedded
strategy is the electrical resolution of DNA with nanopore; (FFPE) blocks which commonly results in RNA degradation [69]
however this could be resolved via reconstruction of nanopores and DNA crosslinking [70], thus making samples unsuitable for
using smaller pore sizes or heightened bandwidths. Furthermore, profiling and data production. AI uncertainty also needs to be
environmental conditions such as temperature and surrounding considered, with most approaches resulting in point-estimate
humidity could also affect the nanopipette setup. However, (predictive) methods, which could, when applied to the clinic,
despite such limitations, due to the nature of this strategy, panels result in overconfident predictions and inaccurate diagnosis.
of known miRNA signatures could be screened simultaneously Bayesian approaches, such as the recently reported Epistemic
within biosamples such as serum (as reported) as well as other Invariance in Cancer Classification (EpICC) [71], could however
biosamples including CSF thus aiding towards its use for miRNA help resolve this. Despite such limitations, the invaluable
detection in imperceptible CNS cancers. Another study reported implications of (often combined) computational techniques holds
impressive limit of detection (LOD) improvements in comparison great hope for accurately and promptly predicting the presence,
to the techniques described. Kanik et al. [66] presented a sensitive and stage, of cancers including imperceptible neoplasm.
and multiplexed digital microarray setup that uses a polarisation-
enhanced single-particle reflectance imaging sensor (SP-IRIS) to
detect plasmonic gold nanorod probes that hybridise to miRNA CURRENT CHALLENGES AND RECOMMENDATIONS IN THE USE
targets. Within proof of concept studies, a LOD of 100 attomolar OF MICRORNAS FOR EARLY CANCER DETECTION
was reported when detecting synthetic miR-223-3p as a target, As described, miRNAs meet key characteristics of being deemed
showing marked improvements over more common femtomolar promising candidate clinical biomarkers due to their stable
ranges. Furthermore, the assay itself was conducted in 35 min properties and ubiquity within a plethora of readily accessible
highlighting its value within a healthcare setting for rapid biofluids, obtained via non- and minimally invasive means.
diagnostic applications. Storage of biofluids is an important consideration when consider-
ing biomarkers, however, miRNAs have been reported to be
Integration of Bioinformatics and machine learning considerably stable in a variety of biofluids, including blood, CSF,
Bioinformatics and machine-learning approaches have shown urine, saliva, and lacrimal fluid held at room temperature for short-
incredible aid towards processing and interpretation of biological term storage up to 96 h [72], and long-term storage for months at
data. In current biomedical research wet experimentation and a time at low (−20 °C) and ultralow (−80 °C) temperatures [73, 74].
bioinformatics analytics are equally pivotal – with large complex Furthermore, routine laboratory processes such as freeze-thaw
datasets needing increasingly sophisticated management and cycles are commonly reported to not adversely affect miRNA
analysis systems to gain biotic knowledge for therapeutic and quality [75, 76]. Conversely, technical considerations in terms of
diagnostic applications. This is true of all disease states, but processing are frequently reported constraints when considering
especially so for cancer due to its complex and multifaceted freely circulating miRNAs within blood samples. Studies have
nature. The introduction of evolutionary supervised AI learning reported variations of miRNA concentrations in samples due to
methods is key to improving early-stage cancer diagnosis and contamination by haemolysis during processing [77], as well as
therapeutic decisions, collectively aiding towards heightened rates altered miRNA profiles as a result of platelet activation [78] –
of remission and overall survival. One such evolutionary learning although the extent of this could be affected by donor age,
method is CancerSig [67] which followed a bi-objective combina- gender, and race [79]. Upon consideration of the wider application
torial genetic algorithm, with the end goal of identifying the of cancer biology this effect may not be particularly disruptive as
miRNA signatures that could aid in early-stage detection of known erythrocyte and platelet-derived miRNAs could be omitted

Oncogene (2024) 43:2135 – 2142

 G.A.D. Metcalf
2140
from screening profiles, should a large enough pool of miRNA the integration of bioinformatics and machine learning, such as
signatures be explored. However, when using samples for the evolutionary supervised AI learning method CancerSig, enable
cardiovascular biomarkers, for example acute myocardial infarc- further identification of promising miRNA signatures via predictive
tion where platelets play a fundamental pathophysiological role, modelling in a scalable and reproducible automated means. Yet,
this could diminish diagnostic value extensively [80]. despite the promise of miRNAs as cancer biomarkers, challenges
While aforementioned donor age, gender and race are reported exist when considering their clinical application. Many of these
to potentially affect haemolysis rates and platelet activation challenges, although being intrinsic in nature, could be circumna-
cascades, physiological variables could also have an impact on vigated however, via united collaborative efforts within the
miRNA concentrations, thus questioning the validity of “healthy scientific community, sharing datasets through open-source
control” samples used within studies. A number of studies have platforms, applying standardised protocols, and fully embracing
explored variables including age [81], gender [82], and BMI [83, 84] technological advances.
with findings suggesting altered expression profiles of particular
miRNAs in comparison to counterparts. Although there is an
increasing number of research outputs, there is a lack of REFERENCES
consistent evidence of which miRNAs are associated with 1. Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global
particular confounding factors, thereby highlighting the necessity cancer statistics 2022: GLOBOCAN estimates of incidence and mortality world-
for further research. wide for 36 cancers in 185 countries. CA Cancer J Clin. 2024. https://doi.org/
10.3322/caac.21834.
Circulating miRNAs concentrations are considerably lower than
2. Ahmad AS, Offman J, Delon C, North BV, Shelton J, Sasieni PD. Years of life lost
those within cells and tissues, presenting challenges in accurate due to cancer in the United Kingdom from 1988 to 2017. Br J Cancer.
detection and quantification. Current gold-standard approaches, 2023;129:1558–68. https://doi.org/10.1038/s41416-023-02422-8
including NGS, enable relatively low miRNA input values, whereas 3. Chari ST. Detecting early pancreatic cancer: problems and prospects. Semin
other recently developed novel approaches as discussed above Oncol. 2007;34:284–94. https://doi.org/10.1053/j.seminoncol.2007.05.005
have the ability to detect attomolar ranges [66]. This attempts to 4. Zhang L, Sanagapalli S, Stoita A. Challenges in diagnosis of pancreatic cancer.
circumnavigate low concentrations, however due to the hetero- World J Gastroenterol. 2018;24:2047–60. https://doi.org/10.3748/
genous nature of cancer, the variability of miRNA expression wjg.v24.i19.2047
amongst populations, individuals, tissues, and cells as well as the 5. Wood DE, Kazerooni EA, Aberle D, Berman A, Brown LM, Eapen GA, et al. NCCN
Guidelines® Insights: Lung Cancer Screening, Version 1.2022. J Natl Compr Cancer
dynamic nature of miRNA regulation, there is an absence of
Netw. 2022;20:754–64. https://doi.org/10.6004/jnccn.2022.0036
unanimously agreed robust internal controls. This creates a 6. Corner J, Hopkinson J, Roffe L. Experience of health changes and reasons for
challenge that can only be truly overcome by combining several delay in seeking care: a UK study of the months prior to the diagnosis of lung
factors including agreed consensus amongst the scientific cancer. Soc Sci Med. 2024;62:1381–91. https://doi.org/10.1016/
community (aligned under internationally recognised SOPs) for j.socscimed.2005.08.012
testing, consistent validation studies, and crucially open-access 7. Sun VC-Y, Sarna L. Symptom management in hepatocellular carcinoma. Clin J
data-sharing repositories. Oncol Nurs. 2008;12:759–66. https://doi.org/10.1188/08.cjon.759-766
Lastly, while there are an ever-increasing number of techniques 8. Waleleng BJ, Adiwinata R, Wenas NT, Haroen H, Rotty L, Gosal F, et al. Screening
being reported to profile and quantify stable miRNAs obtained of pancreatic cancer: target population, optimal timing and how? Ann Med Surg.
2022;84. https://doi.org/10.1016/j.amsu.2022.104814.
from biofluids, there is yet to be an optimised standard for cancer
9. Liam C-K, Liam Y-S, Poh M-E, Wong C-K. Accuracy of lung cancer staging in the
biomarker discovery within clinical settings. Furthermore, multidisciplinary team setting. Transl Lung Cancer Res. 2020;9:1654–66. https://
although there is a pressing need for accurate diagnostic tests, doi.org/10.21037/tlcr.2019.11.28
there is also a growing view within the scientific community that a 10. Huber RM, Cavic M, Kerpel-Fronius A, Viola L, Field J, Jiang L, et al. Lung cancer
turning point from the lack of sensitive and specific screening screening considerations during respiratory infection outbreaks, epidemics or
tests to increasingly sensitive detection methods can, in fact, make pandemics: an international association for the study of lung cancer early
it difficult to distinguish between clinically insignificant changes detection and screening committee report. J Thorac Oncol. 2024;17:228–38.
and lesions that could result in aggressive, life-threatening, https://doi.org/10.1016/j.jtho.2021.11.008
cancers. This issue, while not yet reported within the field of 11. Choi J-Y, Lee J-M, Sirlin CB. CT and MR imaging diagnosis and staging of hepa-
tocellular carcinoma: Part II. extracellular agents, hepatobiliary agents, and
miRNA biomarkers, is something that needs to be carefully
ancillary imaging features. Radiology. 2014;273:30–50. https://doi.org/10.1148/
observed moving forward. radiol.14132362
12. Di Tommaso L, Spadaccini M, Donadon M, Personeni N, Elamin A, Aghemo A,
et al. Role of liver biopsy in hepatocellular carcinoma. World J Gastroenterol.
CONCLUSION 2019;25:6041–52. https://doi.org/10.3748/wjg.v25.i40.6041
Early detection of cancer is vital in attempts to mitigate the 13. Walter FM, Penfold C, Joannides A, Saji S, Johnson M, Watts C, et al. Missed
growing burden of the disease, with the ambition of increasing opportunities for diagnosing brain tumours in primary care: a qualitative study of
patient remission and overall survival rates. However, despite this patient experiences. Br J Gen Pract. 2024;69:e224–35. https://doi.org/10.3399/
pressing need, a myriad of challenges exist when considering bjgp19x701861
14. Jozsa F, Gaier C, Ma Y, Kitchen N, McEvoy A, Miserocchi A, et al. Safety and
imperceptible cancers including the asymptomatic nature of many
efficacy of brain biopsy: results from a single institution retrospective cohort
tumours, comorbidities, and a lack of sensitive and specific study. Brain Spine. 2023;3:101763. https://doi.org/10.1016/j.bas.2023.101763
screening tests. MiRNAs present as promising early diagnostic 15. Ambros V. Control of developmental timing in Caenorhabditis elegans. Curr Opin
cancer biomarkers, with highly sensitive and specific diagnostic Genet Dev. 2000;10:428–33. https://doi.org/10.1016/s0959-437x(00)00108-8
signatures being well reported within the literature for a variety of 16. Reinhart BJ, Slack FJ, Basson M, Pasquinelli AE, Bettinger JC, Rougvie AE, et al. The
difficult-to-diagnose cancers such as NSCLC and HCC, as well as 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis ele-
being able to accurately discern between patients with pancrea- gans. Nature. 2000;403:901–6. https://doi.org/10.1038/35002607
titis and pancreatic cancer. The field of diagnostics continues to 17. Calin GA, Ferracin M, Cimmino A, Di Leva G, Shimizu M, Wojcik SE, et al. A
drive forward with enhanced detection of miRNAs being reported. MicroRNA signature associated with prognosis and progression in chronic lym-
phocytic leukemia. N Engl J Med. 2005;353:1793–801. https://doi.org/10.1056/
Commercial platforms such as Nanostring’s nCounter ® microRNA
nejmoa050995
assay as well as Next Generation RNA Sequencing (RNAseq) 18. Rodriguez A. Identification of mammalian microRNA host genes and transcription
outperform what was once considered the gold-standard method, units. Genome Res. 2004;14:1902–10. https://doi.org/10.1101/gr.2722704
RT-qPCR, in terms of sensitivity and specificity, with other novel 19. Lee Y, Kim M, Han J, Yeom K-H, Lee S, Baek SH, et al. MicroRNA genes are
sensing technologies offering even greater sensitivity and LOD transcribed by RNA polymerase II. EMBO J. 2024;23:4051–60. https://doi.org/
rates as well as heightened multiplexing capabilities. Additionally, 10.1038/sj.emboj.7600385

Oncogene (2024) 43:2135 – 2142

 G.A.D. Metcalf
2141
20. Cai X. Human microRNAs are processed from capped, polyadenylated transcripts 45. Arab A, Karimipoor M, Irani S, Kiani A, Zeinali S, Tafsiri E, et al. Potential circulating
that can also function as mRNAs. RNA. 2004;10:1957–66. https://doi.org/10.1261/ miRNA signature for early detection of NSCLC. Cancer Genet. 2017;216217:150–8.
rna.7135204 https://doi.org/10.1016/j.cancergen.2017.07.006
21. Yi R. Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin 46. Amr KS, Elmawgoud Atia HA, Elazeem Elbnhawy RA, Ezzat WM. Early diagnostic
RNAs. Genes Dev. 2003;17:3011–6. https://doi.org/10.1101/gad.1158803 evaluation of miR-122 and miR-224 as biomarkers for hepatocellular carcinoma.
22. Pratt AJ, MacRae IJ. The RNA-induced silencing complex: a versatile gene- Genes Dis. 2017;4:215–21. https://doi.org/10.1016/j.gendis.2017.10.003
silencing machine. J Biol Chem. 2009;284:17897–901. https://doi.org/10.1074/ 47. Lu M, Kong X, Wang H, Huang G, Ye C, He Z. A novel microRNAs expression
jbc.r900012200 signature for hepatocellular carcinoma diagnosis and prognosis. Oncotarget.
23. Medley JC, Panzade G, Zinovyeva AY. microRNA strand selection: Unwinding the 2017;8:8775–84. https://doi.org/10.18632/oncotarget.14452
rules. WIREs RNA. 2020. https://doi.org/10.1002/wrna.1627. 48. Shehab-Eldeen S, Nada A, Abou-Elela D, El-Naidany S, Arafat E, Omar T. Diag-
24. Bendich A, Wilczok T, Borenfreund E. Circulating DNA as a possible factor in nostic performance of microRNA-122 and microRNA-224 in Hepatitis C virus-
oncogenesis. Science. 1965;148:374–6. https://doi.org/10.1126/ induced hepatocellular carcinoma (HCC). Asian Pac J Cancer Prev.
science.148.3668.374 2019;20:2515–22. https://doi.org/10.31557/apjcp.2019.20.8.2515
25. Mandel P, Metais P. Nuclear acids in human blood plasma. C R Seances Soc Biol 49. Wei X, Liu H, Li X, Liu X. Overexpression of MiR122 promotes apoptosis of
Fil. 1948;142:241–3. hepatocellular carcinoma via targeting TLR4. Ann Hepatol. 2019;18:869–78.
26. Hasselmann DO, Rappl G, Rössler M, Ugurel S, Tilgen W, Reinhold U. Detection of https://doi.org/10.1016/j.aohep.2019.07.005
tumor-associated circulating mRNA in serum, plasma and blood cells from 50. Nomair AM, Issa NM, Madkour MA, Shamseya MM. The clinical significance of
patients with disseminated malignant melanoma. Oncol Rep. 2001. https:// serum miRNA-224 expression in hepatocellular carcinoma. Clin Exp Hepatol.
doi.org/10.3892/or.8.1.115. 2020;6:20–7. https://doi.org/10.5114/ceh.2020.93052
27. Stroun M, Anker P, Maurice P, Gahan PB. Circulating nucleic acids in higher 51. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin.
organisms. Int Rev Cytol. 1977;51:1–48. https://doi.org/10.1016/s0074-7696(08) 2020;70:7–30. https://doi.org/10.3322/caac.21590
60225-9 52. Kopkova A, Sana J, Fadrus P, Machackova T, Vecera M, Vybihal V, et al. MicroRNA
28. Kamm RC, Smith AG. Ribonuclease activity in human plasma. Clin Biochem. isolation and quantification in cerebrospinal fluid: a comparative methodical
1972;5:198–200. https://doi.org/10.1016/S0009-9120(72)80033-X study. PLoS ONE. 2018;13:e0208580. https://doi.org/10.1371/journal.pone.0208580
29. Xi Y, Nakajima G, Gavin E, Morris CG, Kudo K, Hayashi K, et al. Systematic analysis 53. Teplyuk NM, Mollenhauer B, Gabriely G, Giese A, Kim E, Smolsky M, et al.
of microRNA expression of RNA extracted from fresh frozen and formalin-fixed MicroRNAs in cerebrospinal fluid identify glioblastoma and metastatic brain
paraffin-embedded samples. RNA. 2007;13:1668–74. https://doi.org/10.1261/ cancers and reflect disease activity. Neuro Oncol. 2012;14:689–700. https://
rna.642907 doi.org/10.1093/neuonc/nos074
30. Chen X, Ba Y, Ma L, Cai X, Yin Y, Wang K, et al. Characterization of microRNAs in 54. Akers JC, Hua W, Li H, Ramakrishnan V, Yang Z, Quan K, et al. A cerebrospinal fluid
serum: a novel class of biomarkers for diagnosis of cancer and other diseases. Cell microRNA signature as biomarker for glioblastoma. Oncotarget. 2017;8:68769–79.
Res. 2008;18:997–1006. https://doi.org/10.1038/cr.2008.282 https://doi.org/10.18632/oncotarget.18332
31. Hu G, Drescher K, Chen X. Exosomal miRNAs: biological properties and ther- 55. Shalaby T, Fiaschetti G, Baulande S, Gerber N, Baumgartner M, Grotzer M.
apeutic potential. Front Genet. 2012;3. https://doi.org/10.3389/fgene.2012.00056. Detection and quantification of extracellular microRNAs in medulloblastoma.
32. Turchinovich A, Weiz L, Burwinkel B. Extracellular miRNAs: the mystery of their J Cancer Metastasis Treat. 2015;1:67. https://doi.org/10.4103/2394-
origin and function. Trends Biochem Sci. 2012;37:460–5. https://doi.org/10.1016/ 4722.157068
j.tibs.2012.08.003 56. Kopkova A, Sana J, Machackova T, Vecera M, Radova L, Trachtova K, et al. Cer-
33. Valadi H, Ekström K, Bossios A, Sjöstrand M, Lee JJ, Lötvall JO. Exosome-mediated ebrospinal fluid microRNA signatures as diagnostic biomarkers in brain tumors.
transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange Cancers. 2019;11:1546. https://doi.org/10.3390/cancers11101546
between cells. Nat Cell Biol. 2007;9:654–9. https://doi.org/10.1038/ncb1596 57. Chen C, Ridzon DA, Broomer AJ, Zhou Z, Lee DH, Nguyen JT, et al. Realtime
34. Li L, Zhu D, Huang L, Zhang J, Bian Z, Chen X, et al. Argonaute 2 complexes quantification of microRNAs by stem–loop RT–PCR. Nucleic Acids Res.
selectively protect the circulating microRNAs in cell-secreted microvesicles. PLoS 2005;33:e179. https://doi.org/10.1093/nar/gni178
ONE. 2012;7:e46957. https://doi.org/10.1371/journal.pone.0046957 58. Mestdagh P, Hartmann N, Baeriswyl L, Andreasen D, Bernard N, Chen C, et al.
35. Elkayam E, Kuhn C-D, Tocilj A, Haase AD, Greene E, Hannon GJ, et al. The structure Evaluation of quantitative miRNA expression platforms in the microRNA quality
of human argonaute-2 in complex with miR-20a. Cell. 2012;150:100–10. https:// control (miRQC) study. Nat Methods. 2014;11:809–15. https://doi.org/10.1038/
doi.org/10.1016/j.cell.2012.05.017 nmeth.3014
36. Hüttenhofer A, Mayer G. Circulating miRNAs as biomarkers of kidney disease. Clin 59. Foye C, Yan IK, David W, Shukla N, Habboush Y, Chase L, et al. Comparison of
Kidney J. 2016:sfw075. https://doi.org/10.1093/ckj/sfw075. miRNA quantitation by Nanostring in serum and plasma samples. PLoS ONE.
37. Chevillet JR, Kang Q, Ruf IK, Briggs HA, Vojtech LN, Hughes SM, et al. Quantitative 2017;12:e0189165. https://doi.org/10.1371/journal.pone.0189165
and stoichiometric analysis of the microRNA content of exosomes. Proc Natl Acad 60. Armstrong DA, Green BB, Seigne JD, Schned AR, Marsit CJ. MicroRNA molecular
Sci USA. 2014;111:14888–93. https://doi.org/10.1073/pnas.1408301111 profiling from matched tumor and bio-fluids in bladder cancer. Mol Cancer.
38. Lawrie CH, Gal S, Dunlop HM, Pushkaran B, Liggins AP, Pulford K, et al. Detection 2015;14. https://doi.org/10.1186/s12943-015-0466-2.
of elevated levels of tumour-associated microRNAs in serum of patients with 61. Buschmann D, Haberberger A, Kirchner B, Spornraft M, Riedmaier I, Schelling G,
diffuse large B-cell lymphoma. Br J Haematol. 2008;141:672–5. https://doi.org/ et al. Toward reliable biomarker signatures in the age of liquid biopsies - how to
10.1111/j.1365-2141.2008.07077.x standardize the small RNA-Seq workflow. Nucleic Acids Res. 2016;44:5995–6018.
39. Lu J, Getz G, Miska EA, Alvarez-Saavedra E, Lamb J, Peck D, et al. MicroRNA https://doi.org/10.1093/nar/gkw545
expression profiles classify human cancers. Nature. 2005;435:834–8. https:// 62. Fromm B, Høye E, Domanska D, Zhong X, Aparicio-Puerta E, Ovchinnikov VV, et al.
doi.org/10.1038/nature03702 MirGeneDB 2.1: toward a complete sampling of all major animal phyla. Nucleic
40. Shi W, Wartmann T, Accuffi S, Al-Madhi S, Perrakis A, Kahlert C, et al. Integrating a Acids Res. 2021;50:D204–10. https://doi.org/10.1093/nar/gkab1101
microRNA signature as a liquid biopsy-based tool for the early diagnosis and 63. Fromm B, Zhong X, Tarbier M, Friedlander MR, Hackenberg M. The limits of
prediction of potential therapeutic targets in pancreatic cancer. Br J Cancer. human microRNA annotation have been met. RNA. 2022;28:rna.079098.122.
2024;130:125–34. https://doi.org/10.1038/s41416-023-02488-4 https://doi.org/10.1261/rna.079098.122
41. Ciello Adel, Franchi P, Contegiacomo A, Cicchetti G, Bonomo L, Larici AR. Missed 64. Khamina K, Diendorfer AB, Skalicky S, Weigl M, Pultar M, Krammer TL, et al. A
lung cancer: when, where, and why? Diagn Interv Radiol. 2017;23:118–26. https:// microRNA next-generation-sequencing discovery assay (miND) for genome-scale
doi.org/10.5152/dir.2016.16187 analysis and absolute quantitation of circulating microrna biomarkers. Int J Mol
42. Dong X, Chang M, Song X, Ding S, Xie L, Song X. Plasma miR‐1247‐5p, miR‐301b‐3p Sci. 2022;23:1226. https://doi.org/10.3390/ijms23031226
and miR‐105‐5p as potential biomarkers for early diagnosis of non‐small cell lung 65. Cai S, Pataillot-Meakin T, Shibakawa A, Ren R, Bevan CL, Ladame S, et al. Single-
cancer. Thorac Cancer. 2020;12:539–48. https://doi.org/10.1111/1759-7714.13800 molecule amplification-free multiplexed detection of circulating microRNA can-
43. Liu H, Ma X, Niu N, Zhao J, Lu C, Yang F, et al. MIR-301b-3p promotes lung cer biomarkers from serum. Nat Commun. 2021;12. https://doi.org/10.1038/
adenocarcinoma cell proliferation, migration and invasion by targeting DLC1. s41467-021-23497-y
Technol Cancer Res Treat. 2021;20:153303382199003. https://doi.org/10.1177/ 66. Ekiz Kanik F, Celebi I, Sevenler D, Tanriverdi K, Lortlar Ünlü N, Freedman JE, et al.
1533033821990036 Attomolar sensitivity microRNA detection using real-time digital microarrays. Sci
44. Li P, Xing W, Xu J, Yuan D, Liang G, Liu B, et al. microRNA-301b-3p down- Rep. 2022;12. https://doi.org/10.1038/s41598-022-19912-z
regulation underlies a novel inhibitory role of long non-coding RNA MBNL1-AS1 67. Sathipati SY, Tsai M, Shukla SK, Ho S-Y. Artificial intelligence-driven pan-cancer
in non-small cell lung cancer. Stem Cell Res Ther. 2019;10. https://doi.org/ analysis reveals miRNA signatures for cancer stage prediction. Hum Genet
10.1186/s13287-019-1235-8. Genomics Adv. 2023;4:100190. https://doi.org/10.1016/j.xhgg.2023.100190

Oncogene (2024) 43:2135 – 2142

 G.A.D. Metcalf
2142
68. Chi H, Chen H, Wang R, Zhang J, Jiang L, Zhang S, et al. Proposing new early 88. Hennessey PT, Sanford T, Choudhary A, Mydlarz WW, Brown D, Adai AT, et al.
detection indicators for pancreatic cancer: Combining machine learning and Serum microrna biomarkers for detection of non-small cell lung cancer. PLoS
neural networks for serum miRNA-based diagnostic model. Front Oncol. 2023;13. ONE. 2012;7. https://doi.org/10.1371/journal.pone.0032307
https://doi.org/10.3389/fonc.2023.1244578 89. Sanfiorenzo C, Ilie MI, Belaid A, Barlési F, Mouroux J, Marquette CH, et al. Two
69. Levin Y, Talsania K, Tran B, Shetty J, Zhao Y, Mehta M. Optimization for panels of plasma microRNAs as non-invasive biomarkers for prediction of
sequencing and analysis of degraded FFPE-RNA Samples. J Vis Exp. 2020. https:// recurrence in resectable NSCLC. PLoS ONE. 2013;8. https://doi.org/10.1371/
doi.org/10.3791/61060 journal.pone.0054596.
70. Ottestad AL, Emdal EF, Grønberg BH, Halvorsen TO, Dai HY. Fragmentation 90. Liao J, Shen J, Leng Q, Qin M, Zhan M, Jiang F. MicroRNA-based biomarkers for
assessment of FFPE DNA helps in evaluating NGS library complexity and inter- diagnosis of non-small cell lung cancer (NSCLC). Thorac Cancer. 2020;11:762–8.
pretation of NGS results. Exp Mol Pathol. 2022;126:104771. https://doi.org/ https://doi.org/10.1111/1759-7714.13337
10.1016/j.yexmp.2022.104771 91. Fang Y, Yan D, Wang L, Zhang J, He Q. Circulating microRNAs (miR-16, miR-22,
71. Joshi P, Dhar R. EpICC: a Bayesian neural network model with uncertainty cor- miR-122) expression and early diagnosis of hepatocellular carcinoma. J Clin Lab
rection for a more accurate classification of cancer. Sci Rep. 2022;12. https:// Anal. 2022;36. https://doi.org/10.1002/jcla.24541
doi.org/10.1038/s41598-022-18874-6 92. Tan Y, Ge G, Pan T, Wen D, Chen L, Yu X, et al. A serum microrna panel as
72. Weber JA, Baxter DH, Zhang S, Huang DY, How Huang K, Jen Lee M, et al. The potential biomarkers for hepatocellular carcinoma related with hepatitis b virus.
microRNA spectrum in 12 body fluids. Clin Chem. 2010;56:1733–41. https:// PLoS ONE. 2014;9:e66577. https://doi.org/10.1371/journal.pone.0107986
doi.org/10.1373/clinchem.2010.147405 93. Cho HJ, Baek GO, Seo CW, Ahn HR, Sung S, Son JA, et al. Exosomal microRNA-4661-
73. Glinge C, Clauss S, Boddum K, Jabbari R, Jabbari J, Risgaard B, et al. Stability of 5p–based serum panel as a potential diagnostic biomarker for early-stage hepato-
circulating blood-based microRNAs – pre-analytic methodological considera- cellular carcinoma. Cancer Med. 2020;9:5459–72. https://doi.org/10.1002/cam4.3230
tions. PLoS ONE. 2017;12:e0167969. https://doi.org/10.1371/ 94. Zhang Y, Li T, Qiu Y, Zhang T, Guo P, Ma X, et al. Serum microRNA panel for early
journal.pone.0167969 diagnosis of the onset of hepatocellular carcinoma. Medicine. 2017;96. https://
74. Kupec T, Bleilevens A, Iborra S, Najjari L, Wittenborn J, Maurer J, et al. Stability of doi.org/10.1097/MD.0000000000005642
circulating microRNAs in serum. PLoS ONE. 2022;17:e0268958. https://doi.org/ 95. Baraniskin A, Kuhnhenn J, Schlegel U, Maghnouj A, Zöllner H, Schmiegel W, et al.
10.1371/journal.pone.0268958 Identification of microRNAs in the cerebrospinal fluid as biomarker for the diagnosis
75. Matias-Garcia PR, Wilson R, Mussack V, Reischl E, Waldenberger M, Gieger C, et al. of glioma. Neuro Oncol. 2012;14:29–33. https://doi.org/10.1093/neuonc/nor169
Impact of long-term storage and freeze-thawing on eight circulating microRNAs 96. Ivo D’urso P, Fernando D’urso O, Damiano Gianfreda C, Mezzolla V, Storelli C,
in plasma samples. PLoS ONE. 2020;15:e0227648. https://doi.org/10.1371/ Marsigliante S. miR-15b and miR-21 as circulating biomarkers for diagnosis of
journal.pone.0227648 glioma. Curr Genomics. 2015;16:304–11.
76. Chan S-F, He C, Kai-Rui KG, Zou R. Preanalytic methodological considerations and 97. Lai NS, Wu DG, Fang XG, Lin YC, Chen SS, Li ZB, et al. Serum microRNA-210 as a
sample quality control of circulating miRNAs. J Mol Diagn. 2023;25:438–53. potential noninvasive biomarker for the diagnosis and prognosis of glioma. Br J
https://doi.org/10.1016/j.jmoldx.2023.03.005 Cancer. 2015;112:1241–6. https://doi.org/10.1038/bjc.2015.91
77. MacLellan SA, MacAulay C, Lam S, Garnis C. Pre-profiling factors influencing
serum microRNA levels. BMC Clin Pathol. 2014;14. https://doi.org/10.1186/1472-
6890-14-27. AUTHOR CONTRIBUTIONS
78. Mitchell AJ, Gray WD, Hayek SS, Ko Y-A, Thomas S, Rooney K, et al. Platelets All contributions in this article were from the single author.
confound the measurement of extracellular miRNA in archived plasma. Sci Rep.
2016;6:32651. https://doi.org/10.1038/srep32651
79. Simon LM, Edelstein LC, Nagalla S, Woodley AB, Chen ES, Kong X, et al. Human
platelet microRNA-mRNA networks associated with age and gender revealed by COMPETING INTERESTS
integrated plateletomics. Blood. 2014;123:e37–45. https://doi.org/10.1182/blood- The author declares no competing interests.
2013-12-544692
80. Becker KC, Kwee LC, Neely ML, Grass E, Jakubowski JA, Fox KAA, et al. Circulating
microRNA profiling in Non-ST elevated coronary artery syndrome highlights ADDITIONAL INFORMATION
genomic associations with serial platelet reactivity measurements. Sci Rep. Correspondence and requests for materials should be addressed to
2020;10:6169. https://doi.org/10.1038/s41598-020-63263-6 Gavin A. D. Metcalf.
81. Dluzen DF, Noren Hooten N, De S, Wood WH, Zhang Y, Becker KG, et al. Extra-
cellular RNA profiles with human age. Aging Cell. 2018;17:e12785. https://doi.org/ Reprints and permission information is available at http://www.nature.com/
10.1111/acel.12785 reprints
82. Meder B, Backes C, Haas J, Leidinger P, Stähler C, Großmann T, et al. Influence
of the confounding factors age and sex on microRNA profiles from peripheral Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims
blood. Clin Chem. 2014;60:1200–8. https://doi.org/10.1373/ in published maps and institutional affiliations.
clinchem.2014.224238
83. Lischka J, Schanzer A, Hojreh A, Ba-Ssalamah A, de Gier C, Valent I, et al. Circu-
lating microRNAs 34a, 122, and 192 are linked to obesity-associated inflammation
and metabolic disease in pediatric patients. Int J Obes. 2021;45:1763–72. https://
doi.org/10.1038/s41366-021-00842-1 Open Access This article is licensed under a Creative Commons
84. Kasiappan R, Rajarajan D. Role of MicroRNA regulation in obesity-associated Attribution 4.0 International License, which permits use, sharing,
breast cancer: nutritional perspectives. Adv Nutr Int Rev J. 2017;8:868–88. https:// adaptation, distribution and reproduction in any medium or format, as long as you give
doi.org/10.3945/an.117.015800 appropriate credit to the original author(s) and the source, provide a link to the Creative
85. Liu R, Chen X, Du Y, Yao W, Shen L, Wang C, et al. Serum microRNA expression Commons licence, and indicate if changes were made. The images or other third party
profile as a biomarker in the diagnosis and prognosis of pancreatic cancer. Clin material in this article are included in the article’s Creative Commons licence, unless
Chem. 2012;58:610–8. https://doi.org/10.1373/clinchem.2011.172767 indicated otherwise in a credit line to the material. If material is not included in the
86. Nakamura K, Zhu Z, Roy S, Jun E, Han H, Munoz RM, et al. An exosome-based article’s Creative Commons licence and your intended use is not permitted by statutory
transcriptomic signature for noninvasive, early detection of patients with pan- regulation or exceeds the permitted use, you will need to obtain permission directly
creatic ductal adenocarcinoma: a multicenter cohort study. Gastroenterology. from the copyright holder. To view a copy of this licence, visit http://
2022;163:1252–66.e2. creativecommons.org/licenses/by/4.0/.
87. Makler A, Asghar W. Exosomal miRNA biomarker panel for pancreatic ductal
adenocarcinoma detection in patient plasma: a pilot study. Int J Mol Sci. 2023;24.
https://doi.org/10.3390/ijms24065081. © The Author(s) 2024

Oncogene (2024) 43:2135 – 2142