Plasma exosomal miR-21 and miR-181a differentiates follicular from papillary thyroid cancer

Abstract

Thyroid cancer (TC) is the most common endocrine malignancy and its incidence has increased over the last few decades. As has been revealed by a number of studies, TC tissue’s micro-RNA (miRNA) profile may reflect histological features and the clinical behavior of tumor. However, alteration of the miRNA profile of plasma exosomes associated with TC development has to date not been explored. We isolated exosomes from plasma and assayed their characteristics using laser diffraction particle size analysis, atomic force microscopy, and western blotting. Next, we profiled cancer-associated miRNAs in plasma exosomes obtained from papillary TC patients, before and after surgical removal of the tumor. The diagnostic value of selected miRNAs was evaluated in a large cohort of patients displaying different statuses of thyroid nodule disease. MiRNA assessment was performed by RT-qPCR. In total, 60 patients with different types of thyroid nodal pathology were included in the study. Our results revealed that the development of papillary TC is associated with specific changes in exosomal miRNA profiles; this phenomenon can be used for differential diagnostics. MiRNA-31 was found to be over-represented in the plasma exosomes of patients with papillary TC vs. benign tumors, while miRNA-21 helped to distinguish between benign tumors and follicular TC. MiRNA-21 and MiRNA-181a-5p were found to be expressed reciprocally in the exosomes of patients with papillary and follicular TC, and their comparative assessment may help to distinguish between these types of TC with 100 % sensitivity and 77 % specificity.

Full text

ORIGINAL ARTICLE
Plasma exosomal miR-21 and miR-181a differentiates follicular
from papillary thyroid cancer
Roman Samsonov
1,2
&Vladimir Burdakov
3,4
&Tatiana Shtam
3,4
&Zamira Radzhabovа
2
&
Dmitry Vasilyev
2
&Evgenia Tsyrlina
2
&Sergey Titov
5
&Michail Ivanov
5
&Lev Berstein
2
&
Michael Filatov
3,4
&Nikolay Kolesnikov
5
&Hava Gil-Henn
6
&Anastasia Malek
1,2,6
Received: 5 December 2015 /Accepted: 1 May 2016
#International Society of Oncology and BioMarkers (ISOBM) 2016
Abstract Thyroid cancer (TC) is the most common endo-
crine malignancy and its incidence has increased over the last
few decades. As has been revealed by a number of studies, TC
tissue’s micro-RNA (miRNA) profile may reflect histological
features and the clinical behavior of tumor. However, alter-
ation of the miRNA profile of plasma exosomes associated
with TC development has to date not been explored. We iso-
lated exosomes from plasma and assayed their characteristics
using laser diffraction particle size analysis, atomic force mi-
croscopy, and western blotting. Next, we profiled cancer-
associated miRNAs in plasma exosomes obtained from pap-
illary TC patients, before and after surgical removal of the
tumor. The diagnostic value of selected miRNAs was evalu-
ated in a large cohort of patients displaying different statuses
of thyroid nodule disease. MiRNA assessment was performed
by RT-qPCR. In total, 60 patients with different types of thy-
roid nodal pathology were included in the study. Our results
revealed that the development of papillary TC is associated
with specific changes in exosomal miRNA profiles; this phe-
nomenon can be used for differential diagnostics. MiRNA-31
was found to be over-represented in the plasma exosomes of
patients with papillary TC vs. benign tumors, while miRNA-
21 helped to distinguish between benign tumors and follicular
TC. MiRNA-21 and MiRNA-181a-5p were found to be
expressed reciprocally in the exosomes of patients with pap-
illary and follicular TC, and their comparative assessment may
help to distinguish between these types of TC with 100 %
sensitivity and 77 % specificity.
Keywords Thyroid cancer .Exosomes .MicroRNA .
Diagnostics
Introduction
Thyroid cancer (TC) is a common endocrine malignancy with
a low death rate but increasing incidence. Differentiated TC,
which includes papillary and follicular types, comprises the
majority (90 %) of all malignant thyroid nodules and deter-
mines the increase of TC incidence. Despite relatively good
curability and a low mortality rate, challenges still exist in the
management of patients with differentiated TC. Improved ap-
proaches are required for screening patients with asymptom-
atic thyroid nodules [1,2], for differential diagnostics between
histological sub-types of TC [3,4] and for the TC patients
following up during and after therapy [5].
Micro-RNAs (MiRNAs) are small, non-coding single-
strand RNAs that regulate the expression of multiple
protein-coding genes at the post-transcriptional level and are
implicated in the regulation of virtually all signaling circuits
within a cell. Alteration of miRNA expression or regulatory
functions plays an essential role in the development and pro-
gression of human malignancies [6]. Assessment of the
*Anastasia Malek
anastasia@malek.com
1
Oncosystem Ltd, Hoshimina 11/1-207, Saint-Petersburg 194356,
Russia
2
NN Petrov Institute of Oncology, Leningradskaya 68,
Saint-Petersburg 197758, Russia
3
FSBI Petersburg Nuclear Physics Institute,
Gatchina, Saint-Petersburg 188300, Russia
4
Peter the Great St. Petersburg Polytechnic University,
Polytechnicheskaya 29, Saint-Petersburg 195251, Russia
5
Institute of Molecular and Cellular Biology SB RAS, Lavrentieva 8/
2, Novosibirsk 630090, Russia
6
Faculty of Medicine in the Galilee, Bar-Ilan University, Henrietta
Szold 8, Safed 13100, Israel
Tumor Biol.
DOI 10.1007/s13277-016-5065-3

miRNAs’expression profile in tumor tissues is being actively
explored as a promising approach for personalized cancer
therapy [7]. Specific alterations of the cellular miRNAs’ex-
pression and function have been reported in the case of TC [8,
9]. Recent investigations in these areas present significant ef-
fectiveness for improving diagnostics and the management of
this disease [10,11].
In addition to intra-cellular regulatory functioning, miRNA
can be secreted by cells into interstitial spaces to shuttle the
regulatorysignal to neighboring and distant cells. Detection of
tumor-derived miRNA in various bodily fluids may also be
useful for both early cancer diagnostic and therapeutic man-
agement [12]. Thus, over-representation of miRNAs Let-7e,
miR-151, and miR-222 in the serum of patients with papillary
TC (PTC) have been reported by Yu et al. [13]. The relevance
of circulating miR-146b and miR-222 for the persistence of
PTC was demonstrated in another study performed by Lee
et al. [14]. Recent investigations have revealed the diagnostic
potency of circulating miR-95, miR-146b, miR-190, and miR-
155 for discrimination between benign lesions and PTC [15,
16]. However, all cited studies dealt with total circulating
miRNA, which, in fact, consists of different fractions:
miRNAs associated with components of RISC (Ago2),
miRNAs loaded into extracellular vesicles (exosomes,
ectosomes, apoptotic bodies), and miRNAs coupled with
high-density lipoprotein particles. Although the biological sig-
nificance of these variously packaged forms of circulating
miRNAs is still unclear, the diagnostic and disease prognostic
potentials of exosomal miRNAs continue to attract growing
attention [17]. By now, miRNAs secreted by tumor cells with-
in exosomes are being considered as an important factor of
disease progression as well as promising disease markers in
various cancer types including breast [18–20], colon [21],
lung [22], pancreatic [23,24], and other cancers [25]. The first
evaluation of miRNA content in PTC-derived exosomes in
vitro was recently performed [26]; however, circulating
exosomal miRNA in patients with TC has to date not been
investigated.
The aim of the present study was to explore whether anal-
ysis of miRNA isolated from the plasma exosomes of patients
with PTC could provide clinically useful information and
whether they have the potency for the development of a new
diagnostic approach. First, we profiled cancer-associated
miRNAs in plasma exosomes obtained from PTC patients
before and after surgical removal of a tumor and selected
miRNAs with well detectable and reduced after surgery ex-
pression levels. Next, the diagnostic values of these miRNAs
were evaluated in a large cohort of patients with different
thyroid nodal pathologies. The design of the study is schemat-
ically presented in Fig. 1. This approach allowed us to define
miRNAs over-represented in plasma exosomes of patients
with PTC or follicular TC (FTC), compared to those with
benign thyroid adenomas. More importantly, we were able
to define miRNAs with a reciprocal change of representation
in plasma exosomes of patients with PTC and FTC.
Comparative assessment of miR-21-5p/miR-181a or
miR-221/miR-181a amplification ratios allowed us to
distinguish patients with FTC from patients with PTC
with 100 % sensitivity and 77 or 87 % specificity, re-
spectively. Test specificity might be improved up to 87 % by
analysis of both ratios in parallel. These results demonstrated
that analysis of exosomal miRNA may provide clinically rel-
evant information and present a promising approach for per-
sonalizing TC therapy.
Material and methods
Patients and clinical samples
The study had Ethical Committee of N.N. Petrov Institute of
Oncology approval. Informed consent was obtained from all
individual participants included in the study. First group of
patients (n=10) included those who underwent radical thy-
roidectomy for PTC in the N.N. Petrov Institute of Oncology
from June to December 2014. All patients had a papillary
thyroid tumor not extending beyond thyroid gland capsule
and had neither regional nor distant metastases (T
2–3
N
0
M
0
).
All pathological findings (tumor histology, staging) were con-
firmed after surgery. Patients with metabolic, immune, or
blood-related diseases were excluded. Patient’s characteristics
are summarized in Table 1. Second group of patients (valida-
tion group, n=50) was recruited retrospectively and included
patients diagnosed with a different status of either benign or
cancerous thyroid nodules and treated in the N.N. Petrov
Institute of Oncology from 2011 to 2014.
Fig. 1 Study design
Tumo r Biol.

From the patients of the first group, blood was collected
twice 1–2 days before surgery and 7–10 days after. From the
patients of second validation group, blood was collected once
before surgery and onset of any kind of therapy. Blood was
collected in a violet-top blood collection tubes treated with
EDTA, centrifuged at 1500gfor 15 min at 4 °C. Plasma was
carefully isolated, frozen, and kept at −80 °C until use.
Surgical specimens of PTC and non-cancerous tissue of the
contralateral lobe were obtained from the patients of the first
group, snap frozen in liquid nitrogen within 15 min from
thyroidectomy, and kept at −80 °C until use.
Exosomes isolation and characterization
Forexosomesisolation,plasma(2ml)wasmelteduntil4°C
and diluted by cold PBS up to 4 ml. All further centrifugation
steps were performed at plus 4 °C. The diluted plasma sam-
ples were first centrifuged at 17,000×gfor 30 min to remove
cell debris. The supernatant was collected and exosomes were
pelleted by ultracentrifugation at 110,000×gfor 70 min. The
exosome pellet was washed in 1 ml of PBS and centrifuged
again at 110,000×gfor 70 min. Exosomal pellet was finally
diluted in either de-ionized water or lysis buffer depending to
further applications.
Size determination of isolated exosomes was performed
using a Microtrac S3500 particle analyzer (Microtrac Inc) ac-
cording to the manufacturer’s instructions. Atomic force mi-
croscopy has been used to visualize exosomes. Purified
exosomes were diluted in de-ionized water and adsorbed to
freshly cleaved mica sheets, fixed with 0.5 % glutaraldehyde,
rinsed with de-ionized water, and air-dried. The samples were
scanned in the air by a semi-contact method with a scanning
microscope of Solver Bio series (NT-MDT, Russia) equipped
with silicon test probe (type NSG01 (NT-MDT)), with a char-
acteristic stiffness of 5.5 N/m and a typical radius of curvature
of the tip (less than 10 nm). The initial amplitude of scanning
was set to 16 nA in current terms; SetPoint was adjusted to
half of the amplitude. Scanning was performed with a fre-
quency of 1.01 Hz. The images were processed using standard
software package (Image Analysis Nova).
Western blotting
Pellets of exosomes isolated from plasma were lysed in lysis
buffer (0.05 M Tris-HCl pH 7.4, 0.15 M NaCl, 1 mM EDTA,
1 % NP-40, 0.1 % SDS, and 0.5 % (w/v)Na-deoxychotale
supplemented with protease inhibitors) on the ice for 30 min
and then centrifuged at 17,000gfor 20 min. The protein ex-
tracts (20 mg) were normalized using Bio-Rad DC Protein
Assay (Bio-Rad Laboratories Inc, USA), separated by SDS-
PAGE, and transferred to the polyvinyl membrane. The mem-
branes were blocked and incubated with primary antibodies
against TSG101 (sc-7964), CD63 (sc-15363), Alix (sc-
49268), and β-Actin (sc-47778) at dilution 1:200 followed
by incubation with corresponding fluorescent-labeled second-
ary antibodies. Blots were evaluated with Odyssey CLx
Infrared Imaging System (LI-COR Biosciences, USA).
RNA extraction
RNA extraction from pelleted exosomes or tissue was per-
formed with RNA isolation Kit (BioSilica, Russia) in accor-
dance with manufacture’s protocol based on conventional
spin-column technology. The RNA concentration and purity
were confirmed using absorbance measurements at wave-
lengths of 260/280 nm ratio with a NanoDrop 2000 (Thermo
Scientific, USA).
MiRNA profiling by RT-PCR
To profile tumor-derived miRNA in plasma exosomes, meth-
od of universal miRNA reverse transcription followed by
miRNA-specific qPCR was applied. The RNA was first
polyadenylated and reverse transcribed using miRCURY
LNA Universal RT microRNA Polyadenylation and cDNA
synthesis Kit (Exiqon, Denmark). Then, quantitative PCR
was performed using Cancer Focus microRNA PCR Panels
and ExiLENT SYBR Green master mix (both from Exiqon,
Denmark) on CFX96 Touch™Real-Time PCR Detection
System (Bio-Rad, USA). UniSp6 synthetic oligos was
used as a standard for absolute quantification; relative
Tabl e 1 Characteristics of
patients included in the study Diagnosis Stage Age (averaged) Gender (F/M) Number
Preliminary screening
1. Papillary thyroid cancer T
2–3
N
0
M
0
54.5 8/2 10
Validation groups
2. Benign thyroid adenoma 53.4 7/1 8
3. Follicular thyroid cancer T
1–3
N
0–1a
M
0
42.8 8/0 8
4. Papillary thyroid cancer T
1–4
N
0
M
0
53.0 9/1 10
5. Papillary thyroid cancer T
(any)
N
1a–1b
M
0
54.0 7/3 10
6. Papillary thyroid cancer T
(any)
N(
any)
M
1
49.8 11/3 14
Tumor Biol.

quantification was performed using either hsa-miR-191-5p or
Ct value averaged through the panel as reference [27].
Both methods gave similar results. Two samples from
one patient were assayed twice; inter-plate result vari-
ability did not exceed 5 % of miRNA expression level after
normalization using Ct value averaged through each panel as
reference. Each of other 18 samples was assayed by single
analysis.
Assessment of selected miRNA by RT-PCR
To assay individual miRNAs, reverse transcription was per-
formed using miRNA-specific stem-looped primers.
Afterwards, cDNA was diluted and amplified using two
PCR primers (miR-specific and universal) and Taqman probe.
This approach is supposed to be more sensitive comparing to
one used for screening [28]. Primers and probes were de-
signed and synthesized by JSC BVector-Best^(Vector-Best,
Russia); sequences are listed in Table 2. All assays were done
in triplicate and results were averaged. Data relative quantifi-
cation was performed using either U6 snRNA or Ct value
averaged through all assays as reference [27]. Using the
above-mentioned method, we averaged results of 200 assays
to evaluate miRNA expression in tissues (20 tissue sam-
ples × 10 miRNAs) and 500 assays to evaluate miRNA ex-
pression in plasma-derived exosomes (50 plasma sam-
ples × 10 miRNAs).
Statistics
All statistical calculations were performed by GraphPad soft-
ware. Wilcoxon matched-pairs signed-rank test was applied to
compare miRNAs’expression in groups of paired samples
(PTC, microscopically normal thyroid tissue); the Mann–
Whitney nonparametric test was applied to compare exosomal
miRNA expression between groups of patients.
Results
Isolation and characterization of plasma exosomes
Membrane micro-vesicles were isolated from plasma by ultra-
centrifugation. As was defined by laser diffraction particle size
analysis of 10 samples, the population of isolated particles
contained two major sub-populations with an averaged diam-
eter of 50 nm (±4.6) and 240 nm (±21.6). A representative
example of measurement is shown in Fig. 2a. We supposed
that the sub-population of particles with a hydrodynamic di-
ameter less than 100 nm was composed of plasma exosomes,
while particles larger than 200 nm may present an admixture
of shedding vesicles such as ectosomes. Atomic force micros-
copy revealed particles with an almost spherical form and size
Tabl e 2 Primers and probes used for miRNA-specific RT-qPCR
lp-126 gtcgtgtctgaggctcactgagacctattcgcacctcgacacgaccgcattatt
f-126 tctcactcgtaccgtgagt
ur3 ctgaggctcactgagacct
p-126 (r6g)-attcgcacc(t-bhq1)cgacacgaccgcattatt-p
lp-145 cgtgtcgccttgtagcacgaccttattcgcaccctcgacacgacagggattc
f-145 acacgtccagttttcccag
ur8 gccttgtagcacgacctta
p-145 (r6g)-ttcgcaccc(t-bhq1)cgacacgacagggattc-p
lp-146a gaggagaggtgaaccagacagacacaatcccacctcctctcctcaacccatg
f-146a aaccaaccatgagaactgaattc
ur5 gtgaaccagacagacacaa
p-146a (r6g)-tcccacc(t-bhq1)cctctcctcaacccatg-p
lp-150 gtcgtgtcgtgaagcagacagacacaatcgcacctcgacacgaccactggtac
f-150 cacactcatctctcccaacc
ur4 gtgaagcagacagacacaa
p-150 (r6g)-tcgcacc(t-bhq1)cgacacgaccactggtac-p
lp-155 gtcagagcgctcttctagcaccactctatcctaccctcgctctgacaccccta
f-155 cccagcttaatgctaatcgtga
ur10 gctcttctagcaccactcta
p-155 (r6g)-tcctaccc(t-bhq1)cgctctgacaccccta-p
lp-181a gaggagaggcctcacaccactaaccatcctcacctcctctcctcactcacc
f-181a ccgcaa+cattcaacgct
ur14 gcctcacaccactaacca
p-181a (r6g)-tca+cctcc(t-bhq1)ctcctcactcaccg-p
lp-206 gtcgtgtctgaggctcactgagacctattcgcacctcgacacgacccacacac
f-206 ccacagctggaatgtaaggaa
ur3 ctgaggctcactgagacct
p-206 (r6g)-attcgcacc(t-bhq1)cgacacgacccacacac-p
lp-21 gtcgtgtctgaggctcactgagacctattcgcacctgacacgactcaacatc
f-21 acgcagctagcttatcagact
ur3 ctgaggctcactgagacct
p-21 (r6g)-attcgca+cc(t-bhq1)gacacgactcaacatcag-p
lp-221 gtcgtgtctgaggctcactgagacctattcgcacctcgacacgacgaaaccca
f-221 cagcagctacattgtctgc
ur3 ctgaggctcactgagacct
p-221 (r6g)-attcgcacc(t-bhq1)cgacacgacgaaacccag-p
lp-223 gtcagagcgtcttctagcacgacttcatcctcacctcgctctgactggggtat
f-223 cactcatctgtcagtttgtcaa
ur6 gtcttctagcacgacttca
p-223 (r6g)-tcctcacc(t-bhq1)cgctctgactggggtat-p
lp-31 gtcgtgtctgaggctcactgagacctattcgcacctcgacacgacagctatgc
f-31 caagcaggcaagatgctg
ur3 ctgaggctcactgagacct
p-31 (r6g)-attcgcacc(t-bhq1)cgacacgacagctatgc-p
lp-u6 gtcgtgtctgaggctgactgagacctattcgcacctgacacgacggccatgc
f-u6 ggccgcatacagagaagatta
r-u6 ctgaggctgactgagacct
p-u6 (r6g)-attcgcacc(t-bhq1)gacacgacggccatgc-p
LP loop primer for reverse transcription, FPCR forward primer, UR
universal PCR reverse primer, PPCR probe
Tumo r Biol.

below 250 nm, which corresponds to established exosome
characteristics (Fig. 2b, c). As revealed by western blotting,
the protein content of isolated particles was enriched by
exosomal markers (components of ESCRT complex ALIX,
Tsg101, and exosomal tetraspanin CD63), compared to the
plasma as a whole (Fig. 3). These results indicated that the
population of isolated micro-vesicles predominantly com-
prises exosomes.
Profiling of cancer-associated miRNAs from the plasma
exosomes of PTC patients
To perform a preliminary screening and to determine whether
exosomal miRNA concentration changes can be associated
with PTC, blood was collected from patients (n.10) prior to
and 1 week after surgery. We assumed surgical removal of a
tumor to results in a reduction of the circulation of tumor-
derived exosomes, to be reflected by specific alterations in
the exosomal miRNA profile. All patients included in the
analysis had a similar clinical status (Table 1).
To focus our assay on cancer-associated changes of
exosomal miRNA profiles and to minimize detection of alter-
ations associated with surgical trauma, we applied Cancer
Focus microRNA PCR panels that allowed us to assay 84
human miRNAs related to cancer. In total, 20 samples obtain-
ed from 10 patients before and 7–10 days after surgery were
analyzed. A detectable level of miRNA expression was ob-
served in 36 % of cases (608 from 1680 (84 × 20)). From 17 to
50 miRNAs were detected in each RNA sample assayed. The
presence of detectable levels of each miRNA among the sam-
ples varied considerably: two miRNAs (miR-16-5p and miR-
186-5p) werenot detected in any of the 20 samples includedin
the analysis; 40 miRNAs were detected in less than half of the
samples (<10); and 42 miRNAs were detected in half or more
Fig. 2 Characteristics of
membrane vesicles isolated from
plasma. aThe size distribution of
vesicles was determined using the
Microtrac S2500 particle
analyzer. b2D images of vesicles
obtained by atomic force
microscopy (AFM). c3D images
of vesicles generated by Image
Analysis Nava software
Fig. 3 Western blot analysis. Equal amount of protein lysates obtained
from entire plasma and pelleted exosomes was assayed using antibodies
against exosomal markers Alix, CD63, and TSG101. Amount of beta-
actin was evaluated as loading control
Tumor Biol.

of the samples (≥10). Normalized expression levels of these
42 miRNAs are shown using a scatter plot in Fig. 4a.Error
bars on the graph indicate the standard deviation of the mean
value for each miRNA expression. Some of these miRNAs
exhibited quite stable expression levels, with little difference
between samples. Fluctuations in expression levels for most of
the other miRNAs had a rather random character. A decrease
in expression level following surgery was observed for several
miRNAs. Among them, nine miRNAs (miR-126-3p, miR-
145-5p, miR-146a-5p, miRNA-181a-5p, miR-206, miR-21-
5p, miR221-3p and miR-223-3p, miR-31-5p) have shown sta-
tistically significant (Pvalue varied from 0.2 to 0.5) surgery-
associated reduction and were selected for further analysis.
Analysis of selected miRNAs expression in papillary TC
tissue
Tumortissueissupposedtobeaprimarysourceofcancer-
associated miRNA; however, other tissues and cells affected
by cancerous disease or surgical intervention may also influ-
ence the profile of exosomal miRNAs in plasma. Тoconfirm
that selected miRNAs can originate from a tumor, we assayed
their expression in a tumor and in normal tissues of removed
thyroid glands.
Total RNA was isolated from 10 paired samples, and the
expression level of nine selected miRNAs was estimated by
miRNA-specific RT, followed by qPCR. All assays were com-
pleted in triplicate and results were averaged and normalized.
The expression level of selected miRNAs in PTC (n= 10) and
in macroscopically normal thyroid tissue (n= 10) was com-
pared using the Wilcoxon matched-pairs signed-ranks test
(Fig. 4b). All nine miRNAs tested were over-expressed in
PTC tissue, while over-expression of only five miRNAs was
statistically significant: miR-31-5p, miR-146a-5p, miR-21-
5p, miR-221-3p, and miR-181a-5p. Based on these results,
we concluded that these five miRNAs were likely the content
of exosomes secreted by PTC cells and may therefore prove
the most reliable as disease markers.
Fig. 4 Cancer-associated
miRNA expression in plasma
exosomes and cancer tissue of
patients with PTC (n.10). a
Exosomal miRNA obtained from
10 PTC patients twice (before and
after surgery) was assayed by
Cancer Focus RT-qPCR Panel.
Results are normalized and
expression levels of 42 miRNAs
detected in more than half tested
samples are illustrated by scatter
plot. Expression fluctuations are
shown by error bars. MiRNA
whose expression level decreased
after surgical tumor removal are
indicated and marked as gray
spots.bExpression of these miRs
was evaluated in PTC and normal
tissue of removed thyroid gland.
Statistical significance of
observed over-expression in
cancer tissue was evaluated by
Wilcoxon matched-pairs signed
rank test and indicated as single
asterisk (<0.05), double asterisks
(0.005)
Tumo r Biol.

Analysis of selected miRNAs in the plasma exosomes
of patients with different thyroid nodular diseases
The goal of the study was to evaluate whether exosomal
miRNA profiles may reflect the development of TC and
whether its assessment may be of clinical relevance. With this
purpose in mind, the expression behavior of nine selected
miRNAs observed in a preliminary assay was explored among
a large cohort of patients with different nodular thyroid pa-
thologies. Thus, patients with benign thyroid adenomas (goi-
ter), FTC, small intra-nodal, locally extended, and metastatic
PTC were enrolled in the study (Fig. 1, Table 1).
First, we quantified selected miRNAs in plasma exosomes
isolated from patients with benign thyroid adenomas and pa-
tients with PTC. It was found that miR-126-3p, miR-145-5p,
and miR-31-5p were significantly over-expressed in plasma
exosomes of cancer patients (Fig. 5a). MiR-31-5p was shown
to be downregulated in benign thyroid adenomas compared to
surrounding tissues [29] while, in addition to our own re-
search, others [30] have reported over-expression of miR-
31-5p in PTC tissues. These results suggest that miR-31-5p
may be implemented in the regulation of malignant transfor-
mation of the thyroid goiter. Assuming thyroid nodules are a
source of exosome-enclosed miRNAs in plasma, exosomal
miR-31-5p assessment can be used for differential diagnostics
of benign thyroid adenomas and PTC. Interestingly,
miR-126-3p and miR-145-5p were shown to be down-
regulated in PTC tissues [9]; moreover, both of these
miRNAs appeared to play a tumor-suppressive role in the
context of thyroid cancer [31–33]. Thus, over-representation
of these miRNAs in plasma exosomes of cancer patients
might reflect a difference between patterns of cellular and
exosomal miRNA due to the selective character of miRNA
incorporation into exosomes.
A comparison of groups of PTC patients at different dis-
ease stages did not reveal any miRNAs that could reflect
disease progression. A comparison of nine selected
miRNAs’expression in a group of patients with benign thy-
roid adenomas and FTC indicated miR-21-5p as a possible
marker for differential diagnostics between these two condi-
tions (Fig. 5b).
Finally, we aimed to explore whether two main histological
forms of differentiated TC (PTC and FTC) may be distin-
guished by exosomal miRNA profiling. To do so, we com-
pared the expression of nine selected miRNAs in correspond-
ing groups of patients. As shown in Fig. 6, miR-21-5p (and
miR-221-3p, though not statistically significant) was over-
expressed in the plasma exosomes of patients with FTC, while
miR-181a was over-expressed in plasma exosomes of patients
with PTC. This observation may reflect the reciprocal charac-
ter of these miRNAs’expression and their function in follic-
ular and papillary types of TC. Next, we analyzed the miR-21-
5p/miR-181a (miR-221/miR-181a) expression ratio in plasma
exosomes of patients with FTC and PTC. Thus, in a group of
42 patients (Table 1; nn.3,4,5,6) recruited for this study, follic-
ular types of cancer were distinguished from papillary types
with 100 % sensitivity and 77 % specificity using Ct (miR-21-
5p) < Ct (miR-181a) criteria. Analysis of the miR-221 and
miR-181a amplification ratio gave 100 % sensitivity and
76 % specificity; however, false-positive diagnostic results
(wrong diagnosis of FTC in the case of microscopically con-
firmed PTC) partially overlapped. Thus, the test specificity
may be improved up to 87 % by analysis of both the miR-
21-5p/miR-181a and the miR-221/miR-181a ratios and by
considering the coincidence of two negative results (Table 3).
This analysis gave the same results being performed by a com-
parison of normalized values as well as by a direct comparison
of the two miRNA amplification rates assessed in plasma
exosome sample of individual patient. It is practically impor-
tant, because an absence of reliable normalization methods and
reference miRNAs is considered as serious weakness in the
context of miRNA-based diagnostic approaches.
Fig. 5 Comparison of selected
miRNA representation in plasma
exosomes of patients with benign
thyroid adenomas (n.8) and
thyroid cancer (n.42). Statistical
significance is evaluated using
Mann–Whitney Utest and
indicated as single asterisk
(P< 0.05), triple asterisks
(P< 0.0005). aAdenoma vs.
papillary thyroid cancer. b
Adenoma vs. follicular thyroid
cancer
Tumor Biol.

Discussion
MiRNAs, both cellular [34] and secreted [35], contribute to
nearly all aspects of carcinogenesis including tumor initiation,
morphogenesis of tumor-associated stroma, neo-angiogene-
sis, immune surveillance, invasion, metastasis, maintenance
of cancer stem cells, drug resistance, and disease recurrence.
In the field of TC, great efforts have been made to explore the
role of cellular miRNA. For instance, from more than 2500
known miRNAs,about 400 are expressed at a significant level
in the thyroid gland tissue [36] and the specific pattern of
miRNAs is being dysregulated during malignant transforma-
tion. Thus, over-expression of miR-146b, miR-155, miR-187,
miR-197, miR-221, miR-222, and miR-224 was shown to be
specific enough to be used as a diagnostic marker providing
with 100 % sensitivity, 94 % specificity, and 95 % accuracy of
cancer detection [37]. The clinical utility of the microRNA
expression panel was further evaluated, revealing that the de-
cision model based on the expression of miR-146b, miR-155,
and miR-221 had a validityof 97.7 % and reliability of 78.4 %
[38]. TC-associated changes of secreted (extracellular)
miRNA are also being intensively studied as well. For in-
stance, a significantly elevated level of Let-7e, miR-151-5p,
and miR-222 in the circulation of PTC patient vs. patients with
benign nodules or healthy donors was revealed in large-scale
study including 245 subjects (42). Authors of other investiga-
tions have indicated miR-146b and miR-155 as promising
markers of PTC (44). Another study showed that level of
miR-181a-5p was significantly increased in the plasma of
TC patients compared with both control donors and with pa-
tients with other types of cancer (43). However, in total, results
of investigations focused on circulating miRNAs appear to be
fewer and less consistent than results of tumor-specimen stud-
ies. One of the possible reasons for such discordance is the
very complex composition of circulating miRNAs arising
from different tissues and playing different biological roles.
Assessment of isolated exosomal fractions of circulating
miRNA presents a promising approach to improve specificity
and clinical relevance of extracellular miRNA analysis [39].
Exosomal miRNA attracts a special interest [17] due to the
advantages of assessment and result evaluation in comparison
with other components of exosomes. The correlation of the
circulating exosome amount and/or content with cancerous
disease progression or prognosis has been established for dif-
ferent malignancies including breast [40], lung [41], colon
[21], and other cancer types. In the field of TC research, one
single study focused on miRNA content of tumor cell-derived
exosomes has been published so far. Lee et al. [26]have
shown that cultured in vitro PTC cells release exosomes con-
taining miR-146b and miR-222. Since over-expression of
these two miRNAs in PCT cells was demonstrated by many
authors, secretion of these molecules within exosomes can be
expected. However, our study is a first attempt to estimate TC-
associated alteration of miRNA in exosomes derived from
plasma of patients. We demonstrated an elevation of exosomal
content of miR-126, miR-145, and miR-31 associated with
PTC and miR-21 associated with FTC. More interestingly,
the reciprocal character of miR-21-5p (miR-221-3p) and
miR-181a-5p expression in PTC and FTC was revealed.
These results have clinical relevance because they indicate
the possibility of distinguishing FTC from PTC through the
comparative assessment of two exosomal miRNAs.
Despite promising results, our study highlighted challenges
of exosomal miRNA analysis and obtained clinically mean-
ingful information from exosomal miRNA profiling. The
method of exosome isolation represents a crucial issue con-
sidered by one recent investigation of Tawari’s group [42].
The authors performed parallel quantification of both
exosomes and miRNA molecule numbers in plasma. It was
Fig. 6 Comparison of selected miRNA representation in plasma
exosomes of patients with follicular (n.8) vs. papillary (n.34) thyroid
cancer. Statistical significance is evaluated using Mann–Whitney Utest
and indicated as single asterisk (P<0.05) and triple asterisks
(P< 0.0005)
Tabl e 3 Diagnostic value of
exosomal miR-21/miR-181a
(miR221/miR-181a) ratio in
distinguishing follicular from
papillary thyroid cancer
Sensitivity Specificity PPV False pos NPV False neg
1. Ct (miR-21) < Ct (miR-181a) 1 0,77 0,57 0,43 1 0
2. Ct (miR-221) < Ct (miR-181a) 1 0,76 0,44 0,55 1 0
3. B1^∩B2^1 0,87 0,66 0,33 1 0
Tumo r Biol.

shown that the ratio of individual miRNA molecules to
exosomes was significantly less than one, i.e., around 100
exosomes corresponded to only one copy of the given
miRNA. This revealed that most individual exosomes in stan-
dard preparations do not carry a biologically relevant number
of miRNAs. Assuming that plasma exosome populations are
quite dynamic and influenced by various physiological and
pathological factors, results by Chevillet and co-authors have
significantly challenged the diagnostic potency of exosomal
miRNA analysis [42]. In a screening assay included in our
study, we observed a high variability of cancer-associated
miRNA profiles in the plasma exosomes of patients with
PTC (Fig. 4a), which may reflect both the individual charac-
teristics of patients as well as not enough representative
exosome templates analyzed on each occasion. An approach
overcoming this obstacle appears to be selective isolation of
tumor-derived exosomes using immune-affine capturing of
tumor-specific proteins on exosome surfaces: HER2 [43],
TM9SF4 [44], and GPC1 [45].
The method of miRNA quantification is another important
aspect that can crucially affect results of analysis. In our study,
we used two different RT-qPCT techniques where miRNA
either tailed with a common sequence and was then reverse
transcribed with universal primers or it was reverse tran-
scribed individually using miRNA-specific stem-loop primers
[28]. The first technique was applied for screening when a
number of amplicons (miRNA) were analyzed from a limited
amount of starting material. The second technique (miRNA-
specific RT) is considered to be more sensitive and was ap-
plied for the analysis of several selected miRNAs from a large
sample collection. In contrast to the screening assay, which
was enable to detect 36 % of miRNAs, this technique allowed
us to attain readable and comparable amplification rates in
more than 80 % of cases. This difference indicated the possi-
bility for optimizing a method of conventional RT-qPCR anal-
ysis. Moreover, new methods of absolute miRNA quantifica-
tion have become available. For example, droplet digital PCR
was recently explored by the Negrini group [46]. The authors
demonstrated that amounts of various miRNAs in the blood
flow are different, well distinguishable, and have specific fluc-
tuation amplitudes. Although this study was not focused on
exosomal fractions of circulating miRNA, the results infer that
the amount of specific miRNAs in the circulation is not ran-
dom and has a physiological range. Importantly, an increase of
miR-181a in the plasma of patients with TC compared to
patients with other cancer types was demonstrated in the re-
port of Ferracin et al. [46], which is in agreement with our own
results. In conclusion, the sensitivity of exosomal miRNA
analysis can be improved by the concomitant evaluation of
tumor-specific proteins on the exosome surface and exosomal
miRNA, as was recently demonstrated [47].
In addition to general methodological issues, exosomal
miRNA-based tests must be well integrated in existing
diagnostic and patient management algorithms to provide
clinically relevant decision-making results. Thus, the manage-
ment of patients with asymptomatic thyroid nodules, the dif-
ferential diagnostics of histological sub-types, and the identi-
fication of aggressive forms of differentiated TC still present
challenges in clinical practice. As has been revealed in a num-
ber of studies, thyroid nodule tissue miRNA profiles may
reflect the clinicopathologic [48,49] and prognostic charac-
teristics [10] of tumors, as relevant information is supposed to
be Benclosed^within tumor-derived exosomes. In our study,
we demonstrated the possibility of distinguishing FTC
from PTC through the comparative assessment of two
exosomal miRNAs: miR-21-5p (miR-221-3p) and miR-
181a-5p. Nevertheless, new investigations are required to re-
veal other clinically relevant correlations and to provide
with a basis for developing of new exosomal miRNA-
based diagnostic tools.
Acknowledgments This study was supported by Oncosystem Ltd. for
(to R.S. and A.M.) and by grants from the Helmsley Trust Fund (to
H.G.H).
Compliance with ethical standards The study had Ethical Committee
of N.N. Petrov Institute of Oncology approval. Informed consent was
obtained from all individual participants included in the study.
Conflicts of interest None
References
1. Gharib H. Changing trends in thyroid practice: understand-
ing nodular thyroid disease. Endocr Pract. 2004;10(1):31–9.
doi:10.4158/EP.10.1.31.
2. Dean DS, Gharib H. Epidemiology of thyroid nodules. Best
Pract Res Clin Endocrinol Metab. 2008;22(6):901–11. doi:
10.1016/j.beem.2008.09.019.
3. Kato MA, Fahey 3rd TJ. Molecular markers in thyroid can-
cer diagnostics. Surg Clin North Am. 2009;89(5):1139–55.
doi:10.1016/j.suc.2009.06.012.
4. Wang CC, Friedman L, Kennedy GC, Wang H, Kebebew E,
Steward DL, et al. A large multicenter correlation study of thyroid
nodule cytopathology and histopathology. Thyroid. 2011;21(3):
243–51. doi:10.1089/thy.2010.0243.
5. Yip L, Kelly L, Shuai Y, Armstrong MJ, Nikiforov YE, Carty SE, et
al. MicroRNA signature distinguishes the degree of aggressiveness
of papillary thyroid carcinoma.Ann Surg Oncol. 2011;18(7):2035–
41. doi:10.1245/s10434-011-1733-0.
6. Di Leva G, Garofalo M, Croce CM. MicroRNAs in cancer.
Annu Rev Pathol. 2014;9:287–314. doi:10.1146/annurev-
pathol-012513-104715.
7. http://www.rosettagenomics.com/.
8. Lee JC, Gundara JS, Glover A, Serpell J, Sidhu SB.
MicroRNA expression profiles in the management of pap-
illary thyroid cancer. Oncologist. 2014;19(11):1141–7. doi:
10.1634/theoncologist.2014-0135.
9. Cancer Genome Atlas Research N. Integrated genomic characteri-
zation of papillary thyroid carcinoma. Cell. 2014;159(3):676–90.
doi:10.1016/j.cell.2014.09.050.
Tumor Biol.

10. Dettmer MS, Perren A, Moch H, Komminoth P, Nikiforov YE,
Nikiforova MN. MicroRNA profile of poorly differentiated thyroid
carcinomas: new diagnostic and prognostic insights. J Mol
Endocrinol. 2014;52(2):181–9. doi:10.1530/JME-13-0266.
11. Zhang Y, Zhong Q, Chen X, Fang J, Huang Z. Diagnostic value of
microRNAs in discriminating malignant thyroid nodules from be-
nign ones on fine-needle aspiration samples. Tumour Biol.
2014;35(9):9343–53. doi:10.1007/s13277-014-2209-1.
12. Schwarzenbach H, Nishida N, Calin GA, Pantel K. Clinical rele-
vance of circulating cell-free microRNAs in cancer. Nat Rev Clin
Oncol. 2014;11(3):145–56. doi:10.1038/nrclinonc.2014.5.
13. Yu S, Liu Y, Wang J, Guo Z, Zhang Q, Yu F, et al. Circulating
microRNA profiles as potential biomarkers for diagnosis of papil-
lary thyroid carcinoma. J Clin Endocrinol Metab. 2012;97(6):
2084–92. doi:10.1210/jc.2011-3059.
14. Lee JC, Zhao JT, Clifton-Bligh RJ, Gill A, Gundara JS, Ip JC, et al.
MicroRNA-222 and microRNA-146b are tissue and circulating
biomarkers of recurrent papillary thyroid cancer. Cancer.
2013;119(24):4358–65. doi:10.1002/cncr.28254.
15. Cantara S, Pilli T, Sebastiani G, Cevenini G, Busonero G, Cardinale
S, et al. Circulating miRNA95 and miRNA190 are sensitive
markers for the differential diagnosis of thyroid nodules in a
Caucasian population. J Clin Endocrinol Metab. 2014;99(11):
4190–8. doi:10.1210/jc.2014-1923.
16. Lee YS, Lim YS, Lee JC, Wang SG, Park HY, Kim SY, et al.
Differential expression levels of plasma-derived miR-146b and
miR-155 in papillary thyroid cancer. Oral Oncol. 2015;51(1):77–
83. doi:10.1016/j.oraloncology.2014.10.006.
17. Sato-Kuwabara Y, Melo SA, Soares FA, Calin GA. The fusion of
two worlds: non-coding RNAs and extracellular vesicles—diagnos-
tic and therapeutic implications (review). Int J Oncol. 2015;46(1):
17–27. doi:10.3892/ijo.2014.2712.
18. Chen WX, Cai YQ, Lv MM, Chen L, Zhong SL, Ma TF, et al.
Exosomes from docetaxel-resistant breast cancer cells alter
chemosensitivity by delivering microRNAs. Tumour Biol.
2014;35(10):9649–59. doi:10.1007/s13277-014-2242-0.
19. Hannafon BN, Carpenter KJ, Berry WL, Janknecht R, Dooley WC,
Ding WQ. Exosome-mediated microRNA signaling from breast
cancer cells is altered by the anti-angiogenesis agent
docosahexaenoic acid (DHA). Mol Cancer. 2015;14:133.
doi:10.1186/s12943-015-0400-7.
20. Melo SA, Sugimoto H, O’Connell JT, Kato N, Villanueva A, Vidal
A, et al. Cancer exosomes perform cell-independent microRNA
biogenesis and promote tumorigenesis. Cancer Cell. 2014;26(5):
707–21. doi:10.1016/j.ccell.2014.09.005.
21. Ogata-Kawata H, Izumiya M, Kurioka D, Honma Y, Yamada Y,
Furuta K, et al. Circulating exosomal microRNAs as biomarkers of
colon cancer. PLoS One. 2014;9(4):e92921. doi:10.1371/journal.
pone.0092921.
22. Fujita Y, Kuwano K, Ochiya T, Takeshita F. The impact of
extracellular vesicle-encapsulated circulating microRNAs in
lung cancer research. BioMed Res Int. 2014;2014:486413.
doi:10.1155/2014/486413.
23. Que R, Ding G, Chen J, Cao L. Analysis of serum exosomal
microRNAs and clinicopathologic features of patients with
pancreatic adenocarcinoma. World J Surg Oncol. 2013;11:219.
doi:10.1186/1477-7819-11-219.
24. Zoller M. Pancreatic cancer diagnosis by free and exosomal
miRNA. World J Gastrointest Pathophysiol. 2013;4(4):74–90.
doi:10.4291/wjgp.v4.i4.74.
25. Whiteside TL. The potential of tumor-derived exosomes for nonin-
vasive cancer monitoring. Expert Rev Mol Diagn. 2015;1293:310.
26. Lee JC, Zhao JT, Gundara J, Serpell J, Bach LA, Sidhu S. Papillary
thyroid cancer-derived exosomes contain miRNA-146b and
miRNA-222. J Surg Res. 2015;196(1):39–48. doi:10.1016/j.
jss.2015.02.027.
27. Mestdagh P, Van Vlierberghe P, De Weer A, Muth D, Westermann
F, Speleman F, et al. A novel and universal method for microRNA
RT-qPCR data normalization. Genome Biol. 2009;10(6):R64. doi:
10.1186/gb-2009-10-6-r64.
28. Benes V, Castoldi M. Expression profiling of microRNA using real-
time quantitative PCR, how to use it and what is available.
Methods. 2010;50(4):244–9. doi:10.1016/j.ymeth.2010.01.026.
29. Ferraz C, Lorenz S, Wojtas B, Bornstein SR, Paschke R, Eszlinger
M. Inverse correlation of miRNA and cell cycle-associated
genes suggests influence of miRNA on benign thyroid nod-
ule tumorigenesis. J Clin En docrinol Metab. 2013;98(1):E8–16.
doi:10.1210/jc.2012-2564.
30. Suresh R, Sethi S, Ali S, Giorgadze T, Sarkar FH. Differential
expression of MicroRNAs in papillary thyroid carcinoma and their
role in racial disparity. J Cancer Sci Ther. 2015;7(5):145–54. doi:
10.4172/1948-5956.1000340.
31. Xiong Y, Kotian S, Zeiger MA, Zhang L, Kebebew E. miR-126-3p
inhibits thyroid cancer cell growth and metastasis, and is associated
with aggressive thyroid cancer. PLoS One. 2015;10(8):e0130496.
doi:10.1371/journal.pone.0130496.
32. Boufraqech M, Zhang L, Jain M, Patel D, Ellis R, Xiong Y,
et al. miR-145 suppresses thyroid cancer growth and metas-
tasis and targets AKT3. Endocr Relat Cancer. 2014;21(4):517–31.
doi:10.1530/ERC-14-0077.
33. Gu Y, Li D, Luo Q, Wei C, Song H, Hua K, et al. MicroRNA-145
inhibits human papillary cancerTPC1 cell proliferation by targeting
DUSP6. Int J Clin Exp Med. 2015;8(6):8590–8.
34. Acunzo M, Romano G, Wernicke D, Croce CM. MicroRNA and
cancer—a brief overview. Adv Biol Regul. 2015;57:1–9. doi:
10.1016/j.jbior.2014.09.013.
35. Kosaka N. Decoding the secret of cancer by means of extracellular
vesicles. J Clin Med. 2016;5(2). doi:10.3390/jcm5020022.
36. Swierniak M, Wojcicka A, Czetwertynska M, Stachlewska E,
Maciag M, Wiechno W, et al. In-depth characterization of the
microRNA transcriptome in normal thyroid and papillary thyroid
carcinoma. J Clin Endocrinol Metab. 2013;98(8):E1401–9.
doi:10.1210/jc.2013-1214.
37. Nikiforova MN, Tseng GC, Steward D, Diorio D, Nikiforov YE.
MicroRNA expression profiling of thyroid tumors: biological sig-
nificance and diagnostic utility. J Clin Endocrinol Metab.
2008;93(5):1600–8. doi:10.1210/jc.2007-2696.
38. Agretti P, Ferrarini E, Rago T, Candelieri A, De Marco G, Dimida
A, et al. MicroRNA expression profile helps to distinguish benign
nodules from papillary thyroid carcinomas starting from cells of
fine-needle aspiration. Eur J Endocrinol. 2012;167(3):393–400.
doi:10.1530/EJE-12-0400.
39. Whiteside TL. The potential of tumor-derived exosomes for nonin-
vasive cancer monitoring. Expert Rev Mol Diagn. 2015;15(10):
1293–310. doi:10.1586/14737159.2015.1071666.
40. Lowry MC, Gallagher WM, O’Driscoll L. The role of exosomes in
breast cancer. Clin Chem. 2015;61(12):1457–65. doi:10.1373/
clinchem.2015.240028.
41. Taverna S, Giallombardo M, Gil-Bazo I, Carreca AP, Castiglia M,
Chacartegui J, et al. Exosomes isolation and characterization in
serum is feasible in non-small cell lung cancer patients: critical
analysis of evidence and potential role in clinical practice.
Oncotarget. 2016. 10.18632/oncotarget.7638.
42. 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(41):14888–93. doi:10.1073/pnas.1408301111.
43. Koga K, Matsumoto K, Akiyoshi T, Kubo M, Yamanaka N, Tasaki
A, et al. Purification, characterization and biological significance of
tumor-derived exosomes. Anticancer Res. 2005;25(6A):3703–7.
44. Lozupone F, Kont V, Logozzi A, Talpsepp K, Oja T, Kubo A, et al.
TM9SF4 level of expression on exosomes as new marker of
Tumo r Biol.

malignancy in human cancer. First scientific meeting of ISEV-
International Society for Extracellular Vesicles April 18-21 2012.
Gothenburg: University of Gothenburg; 2012. p. 19.
45. Melo SA, Luecke LB, Kahlert C, Fernandez AF, Gammon ST,
Kaye J, et al. Glypican-1 identifies cancer exosomes and
detects early pancreatic cancer. Nature. 2015;523(7559):
177–82. doi:10.1038/nature14581.
46. Ferracin M, Lupini L, Salamon I, Saccenti E, Zanzi MV, Rocchi A,
et al. Absolute quantification of cell-free microRNAs in cancer
patients. Oncotarget. 2015;6(16):14545–55.
47. Madhavan B, Yue S, Galli U, Rana S, Gross W, Muller M, et al.
Combined evaluation of a panel of protein and miRNA serum-
exosome biomarkers for pancreatic cancer diagnosis in-
creases sensitivity and specificity. Int J Cancer. 2015;136(11):
2616–27. doi:10.1002/ijc.29324.
48. Aragon Han P, Weng CH, Khawaja HT, Nagarajan N, Schneider
EB, Umbricht CB, et al. MicroRNA expression and association
with clinicopathologic features in papillary thyroid cancer: a sys-
tematic review. Thyroid. 2015. doi:10.1089/thy.2015.0193.
49. Dettmer M, Perren A, Moch H, Komminoth P, Nikiforov YE,
Nikiforova MN. Comprehensive MicroRNA expression pro-
filing identifies novel markers in follicular variant of papil-
lary thyroid carcinoma. Thyroid. 2013;23(11):1383–9. doi:
10.1089/thy.2012.0632.
Tumor Biol.