Society for Endocrinology
Challenges in circulating miRNA analysis in adrenocortical tumors
Bálint Vékony1,2,*, Gábor Nyirő1,2,3,*, Henriett Butz3,4,5, Bálint Kende Szeredás1,2, Viktória Tóth6,7, Peter Ferdinandy6,7,8, Attila Patócs 3,4,5 and Peter Igaz 1,2
1Department of Endocrinology, Faculty of Medicine, Semmelweis University, Budapest, Hungary
2Department of Internal Medicine and Oncology, Faculty of Medicine, Semmelweis University, Budapest, Hungary
3Department of Laboratory Medicine, Faculty of Medicine, Semmelweis University, Budapest, Hungary 4HUN-REN-SU Hereditary Tumours Research Group, Budapest, Hungary
5Department of Molecular Genetics and the National Tumour Biology Laboratory, National Institute of Oncology, Budapest, Hungary
6Department of Pharmacology and Pharmacotherapy, Faculty of Medicine, Semmelweis University, Budapest, Hungary 7Center for Pharmacology and Drug Research & Development, Semmelweis University, Budapest, Hungary 8Pharmahungary Group, Szeged, Hungary
Correspondence should be addressed to P Igaz: igaz.peter@semmelweis.hu
*(B Vékony and G Nyirő contributed equally to this work)
Abstract
The differentiation of benign and malignant adrenocortical tumors is of major clinical relevance. Circulating microRNAs (miRNAs) hold promise as blood-borne biomarkers of adrenocortical cancer (ACC). There are, however, many difficulties with their use, including technical and biological standardization challenges. Our aim was to evaluate the interchangeability of quantitative polymerase chain reaction (qPCR) and digital PCR (dPCR) for measuring circulating miRNAs and to investigate whether K2- and K3-EDTA as anticoagulants influence the measurements. Blood samples were drawn simultaneously from 20 participants into K2- and K3-EDTA tubes. Three miRNAs shown to be associated with ACC (miR-483-5p, miR-210-3p, miR-21-5p), together with two controls (miR-16-5p, cel-miR-39-3p), were analyzed using RT-qPCR and dPCR. qPCR and dPCR results showed different correlations in K2- and K3-EDTA samples, with K2 performing better regarding 4Ct values. Moreover, proportional biases related to low or high miRNA expressions between the two methods were observed. In qPCR measurements, K3-EDTA samples showed larger standard deviations, particularly for cel-miR-39. While raw Ct values differed between K2- and K3-EDTA only for miR-483-5p, ACt values showed statistically significant differences across all miRNAs except for miR-483-5p. dPCR results were not affected by the choice of anticoagulant. In conclusion, this is the first study demonstrating that dPCR and qPCR results are not easily interchangeable for circulating miRNA, particularly for abundant or rare miRNAs, making cross-validation studies challenging. K2- and K3-EDTA could potentially influence qPCR outcomes, underscoring the need for standardized protocols. A consensus-based methodology could improve reproducibility, enhancing miRNA-based biomarker utility in adrenocortical tumor diagnostics.
Keywords: microRNA; dPCR; qPCR; adrenocortical carcinoma; EDTA
Introduction
Adrenal tumors are relatively common in the adult population, and their incidence rises with age (Fassnacht et al. 2023). Most of these tumors are benign; however, differential diagnosis toward
adrenocortical carcinoma (ACC) is of paramount relevance. ACC is a rare (incidence: 1-2 cases per million people/year) and aggressive tumor with poor prognosis, as its 5-year survival rate is below 30%
(Fassnacht et al. 2018, Fassnacht et al. 2023). Diagnosis and follow-up of ACC both rely heavily on imaging techniques. Blood-borne biomarkers would be highly welcome as markers of malignancy and follow-up.
MicroRNAs (miRNAs) are a class of 21-23 nucleotide long, short, non-coding RNAs regulating translation via forming complex molecular networks of miRNA and mRNA (Bereczki et al. 2025). Their tissue-specific expression profiles and propensity for being excreted into the extracellular space, coupled with their stability, render them promising biomarkers (Igaz & Igaz 2014, Anfossi et al. 2018). Their potential also earned Victor Ambrose and Gary Ruvkun, the scientists who discovered miRNA, the 2024 Nobel Prize in Medicine. Circulating miRNA-based biomarker research has many pitfalls, including the lack of standardization as one of the most significant (Butz 2022), and there is a lack of studies exploring the reproducibility and interchangeability of differing methods.
It has been known since the 1970s that, in laboratory measurements, even as a little difference as the type of the EDTA (ethylenediaminetetraacetic acid) anticoagulant used in the blood collection tube can cause significant differences in results, even in a routine blood count (Goossens et al. 1991, Lima-Oliveira et al. 2013, Vrtaric et al. 2016). EDTA is most commonly used in two salt forms, either di- or tripotassium salts, K2- or K3-EDTA. The presence of one more potassium ion makes K3-EDTA more water-soluble, while also causing it to have higher osmolarity. For circulating miRNA studies, the consensus is to use EDTA-containing collection tubes. Heparin is known to inhibit polymerase chain reactions (PCR), and citrate generally induces greater platelet activation compared to EDTA; therefore, EDTA is considered the anticoagulant of choice for nucleic acid-based studies (Lam et al. 2004, Djordjevic et al. 2006), but different centers use either K2- or K3-EDTA. Whether the difference of one potassium ion has a significant effect on miRNA measurements is unknown. Real-time quantitative PCR (qPCR) is the gold standard method in miRNA experiments, but digital PCR (dPCR) is a newer and emerging technique that also shows promise in this field, particularly for the detection of small amounts of nucleic acids and for the diagnosis and monitoring of minimal residual disease (Conserva et al. 2021).
Our aim in this study was to investigate the interchangeability of dPCR and qPCR utilizing samples taken from the same individuals and to check whether the preference for K2- or K3-EDTA tubes could affect the results of either technique.
Materials and methods
Patient samples
Blood was taken from 20 individuals simultaneously during routine blood draws in both
K2- (BD Vacutainer, BD Plymouth, UK) and K3-EDTA (Vacuette, Greiner-Bio-One, Germany) collection tubes. Both patients followed up at the endocrine outpatient consultations of the Department of Internal Medicine and Oncology (Semmelweis University, Budapest) or otherwise healthy individuals were included. Samples were taken from 13 women and seven men (median age 55.5, IQR: 16). The 20 individuals included 13 healthy individuals and seven outpatients with various endocrine disorders, including but not limited to primary aldosteronism, idiopathic hypoparathyroidism, nonfunctional pituitary adenoma, and hypothyroidism. The study was endorsed by the Ethical Committee of the Hungarian Science Council (ETT, BM/17716-1/2024), and all participants gave informed consent.
RNA isolation
Samples were immediately put on ice, and plasma was separated by centrifugation (1,861 g, 10 min at 4℃, on MPW-380R centrifuge with a swing-out rotor, MPW MED. Instruments, Poland). MicroRNA isolation was carried out utilizing the miRNeasy Serum/Plasma Kit (QIAGEN, Germany) containing QiaZOL, according to the manufacturer’s protocol. 2 uL of cel-miR-39 (miRNA Spike-in Kit, QIAGEN, Germany) was added to each sample during isolation to serve as an external control. The isolated RNA and plasma samples were stored at -80°C. RNA concentrations were measured with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA).
RT-qPCR
Reverse transcription (RT) was performed individually by first diluting each sample to a 2 ng/uL concentration to achieve equal starting amounts of RNA, then the TaqMan MicroRNA Reverse Transcription Kit with specific TaqMan MicroRNA assay primers (Thermo Fisher Scientific, USA) was used according to the manufacturer’s protocol. The following assays were used: cel-mir-39-3p (ID: 000200), hsa-miR-16-5p (ID: 000391), hsa-miR-21-5p (ID: 000397), hsa-miR-210-3p (ID: 000512), and hsa-miR-483-5p (ID: 002338). Quantitative measurements were performed on a QuantStudio 7 Flex real-time PCR, using TaqMan Fast Advanced Master Mix 2X and specific TaqMan probes from the assay kits above (Thermo Fisher Scientific, USA), according to the manufacturer’s protocol, utilizing three parallel wells for each sample on 384-well plates.
Digital PCR
For the dPCR measurements, the same undiluted cDNA samples were used. The two procedures could be easily coupled by utilizing the same FAM-labeled TaqMan probes as reporters for fluorescent detection, which are native dyes to both qPCR and dPCR systems. For these
measurements, we utilized the same TaqMan assays as listed above, with QIAcuity Probe PCR Master Mix, on a two-channel QIAcuity One digital PCR platform (QIAGEN, Germany). 8.5k partitioned 24- and 96-well QIAcuity NanoPlates (QIAGEN, Germany) were applied with the following protocol: each well contained 2 uL of cDNA, 1.2 uL of TaqMan assay, and 3 uL of 4x Master Mix in a 12 uL volume. The thermal cycling protocol contained one activation/denaturation cycle at 95℃ for 2 min and 40 amplification cycles where the denaturation phase was at 95℃ for 15 s, and the annealing/extension phase was at 60℃ for 30 s. Imaging happened immediately after the PCR ended, for 350 ms.
Statistical analysis
All datasets underwent Shapiro-Wilk as well as Kolmogorov-Smirnov tests to check whether their distributions were normal. Differences between K2- and K3-EDTA samples for qPCR results were analyzed by paired t-test with Welch’s correction, as all of the raw Ct as well as ACt values showed normal distribution, utilizing the GraphPad Prism 10.3 software (Dotmatics Inc., USA). P values under 0.05 were considered significant. We compared both the raw Ct values measured during qPCR and the ACt values calculated by normalizing the raw Ct values of a given miRNA to the raw Ct values of the external control, cel-mir-39 (Livak & Schmittgen 2001). As dPCR is an absolute quantification method, no such normalization was needed in its case. Since copy numbers obtained through dPCR are on a linear scale, while Ct values are on a logarithmic scale, copy numbers were transformed to a 2-based logarithm so that the two measurements could be compared. The results of dPCR do not show normal distribution, but after logarithmic conversion, four out of six miRNAs do show Gaussian distribution, while two show a non-normal one. Because of this, we used the Wilcoxon matched-pairs signed-rank test to compare the K2 and K3 groups. Correlation analyses were performed between the raw as well as the ACt values of K2 and K3 groups and the corresponding dPCR results utilizing Spearman’s rank correlation. Bland-Altman analysis was carried out between K2 and K3 qPCR and dPCR results. This method is useful in comparing the results of two different
techniques used on the same sample. It is done through plotting the average of the two measurements on the x-axis and the difference of the two measurements on the y-axis. This analysis is particularly useful for identifying systematic errors and biases between methods (Bland & Altman 1999). Both aforementioned analyses were carried out in the GraphPad Prism software.
Results
Comparison of qPCR and dPCR
Correlation analyses between raw Ct values and copy numbers in the K2 and K3 groups showed that there is
statistically significant correlation between qPCR and dPCR measurements for four of the five miRNAs tested (cel-miR-39-3p, hsa-miR-16-5p, hsa-miR-21-5p, hsa-miR-210-3p, and hsa-miR-483-5p), with hsa-miR-483- 5p being the only miRNA not showing significant correlation. However, qPCR measurements rarely use raw expression data and are usually normalized to at least an external control, often to cel-miR-39-3p.
The correlation between the ACt values and copy numbers in the K2 group showed statistical significance for all four miRNAs tested (hsa-miR-16-5p, hsa-miR-21-5p, hsa-miR-210-3p, and hsa-miR-483-5p). In the K3 group, only hsa-miR-210-3p showed correlation between ACt values and copy numbers. The correlation coefficients and P values for the comparison between the ACt values and copy numbers are shown in Table 1, while the results of the correlation of the raw Ct values and copy numbers are presented in Table 2.
The Bland-Altman analyses between raw Ct, as well as ACt values and copy numbers, revealed that qPCR and dPCR are generally agreeable with each other in most of our miRNA measurements. There was, however, a proportional bias present. These biases can be inferred from the Bland-Altman plots if the points in a given interval diverge more from the central horizontal line,
| ACt K2 | miR-16-5p | miR-21-5p | miR-210-3p | miR-483-5p |
|---|---|---|---|---|
| r | -0.6075 | -0.6054 | -0.5278 | -0.5879 |
| r2 | 0.3691 | 0.3665 | 0.2786 | 0.3456 |
| P | 0.0045 | 0.0377 | 0.0168 | 0.0064 |
| ACt K3 | miR-16-5p | miR-21-5p | miR-210-3p | miR-483-5p |
|---|---|---|---|---|
| r | 0.06165 | -0.1038 | -0.4857 | 0.2498 |
| r2 | 0.000038 | 0.01077 | 0.2359 | 0.0624 |
| P | 0.7962 | 0.6633 | 0.0299 | 0.2881 |
| K2 | cel-miR-39-3p | miR-16-5p | miR-21-5p | miR-210-3p | miR-483-5p |
|---|---|---|---|---|---|
| r | -0.9338 | -0.9805 | -0.9368 | -0.6105 | -0.03485 |
| r2 | 0.871982 | 0.96138 | 0.877594 | 0.37271 | 0.001215 |
| p | 8.72E-11 | 8.55E-15 | 7.08E-12 | 0.0042 | 0.884 |
| K3 | cel-miR-39-3p | miR-16-5p | miR-21-5p | miR-210-3p | miR-483-5p |
| r | -0.6586 | -0.9805 | -0.9489 | -0.9579 | -0.3114 |
| r2 | 0.433754 | 0.96138 | 0.900411 | 0.917572 | 0.09697 |
| p | 0.0016 | 1.22E-11 | 3.05E-13 | 5.02E-11 | 0.1814 |
A
K2
K3
B
K2
K3
-8
-10-
2-
0
-10
Difference
:
-12
0
-2-
-12-
Difference
Difference
5.5
6.0
6.5
7.0
Difference
-14-
-4-
-14-
-2.
-6
-16
-16.
-4
-8
-18
-18
4.0
4.5
5.0
5.5
6.0
3.5
4.0
4.5
5.0
5.5
6.0
-6
-10
Average
4
5
6
7
Average
Average
Average
C
K2
K3
D
K2
K3
10
15-
10
10
8
8
8
Difference
6
Difference
10
Difference
Difference
6
6
4
5
4.
4.
2
2
2-
0
4.5
5.0
5.5
6.0
6.5
0
4.0
4.5
5.0
5.5
6.0
6.5
0
0
3
4
5
6
7
3
4
5
6
7
8
Average
Average
Average
Average
which represents the average of the plot, than in other intervals. The dotted lines represent the average plus- minus two times the standard deviation. The K2 groups show somewhat less bias compared to K3 in most miRNAs measured, and these biases are stronger between the raw Ct values and copy numbers compared to ACt values. This proportional bias is strongest in the beginnings and ends of the plots, which means the two techniques produce generally similar results (if brought to a comparable scale), but for low or high expression values in a given miRNA, the two methods diverge more than expected. Namely, qPCR measures higher expression in the low miRNA range compared to dPCR, but dPCR measures more miRNA in the high expression range compared to qPCR. This is prominent in the case of miR-210-3p and miR-483-5p (the two most useful miRNA markers of adrenocortical malignancy) when comparing raw Ct values and copy numbers but is still non-negligible when ACt values were used. The Bland-Altman plots of the comparison between ACt and copy number measurements are presented in Fig. 1, while the plots of the raw Ct data and copy numbers are presented in Fig. 2.
Effect of K2 or K3 EDTA
qPCR
K3 EDTA samples showed larger standard deviations, most prominently in cel-miR-39. These standard deviation values are presented in Table 3. In the case of raw Ct values, miR-483-5p showed a statistically significant difference between K2 and K3 samples. However, after normalization to the aforementioned
external control, cel-miR-39, the ACt values showed significant differences in all of the miRNAs measured except for hsa-miR-483-5p. The paired line plots presenting the differences observed within the ACt values are shown in Fig. 3, while the paired line plots of the comparisons between raw Ct values are presented in Fig. 4.
dPCR
Wilcoxon matched-pairs signed rank tests showed no statistically significant differences between the results of samples taken from the same patients into K2 and K3 EDTA collection tubes in the case of copy numbers obtained through dPCR. To find out whether the same effect can be observed here as with the qPCR results, we tested the copy numbers again after normalizing them to cel-miR-39. This, however, did not result in any statistically significantly different results either. The paired line plots presenting these analyses are presented in Fig. 5.
Discussion
MicroRNA expression profiles have great potential as biomarkers for the diagnosis of disease (e.g., malignancy), treatment follow-up, or grading of tumors (Barbarotto et al. 2008, Valihrach et al. 2020, Butz et al. 2024). However, as of today, few diagnostic panels based on tissue miRNA expression profiling are available in clinical practice, and even fewer, if any, based on circulating miRNA. This is related to many biological and technical challenges in this field, with the issue of
A
K2
K3
B
K2
K3
20
22
0
20
18
20
Difference
Difference
-5
16
18-
Difference
15
Difference
16
-10-
10
14-
14-
12.
5
12
-15
10
10
-20
0
17.7
17.8
17.9
18.0
18.1
18.2
15
16
17
18
19
20
17.4
17.5
17.6
17.7
17.8
17.9
18.0
17.0
17.2
17.4
17.6
17.8
18.0
Average
Average
Average
Average
C
K2
K3
D
K2
K3
28
30
36
40
26
34
Difference
24
Difference
25
Difference
32
Difference
35
22
20
30
30
20
28
18
15
26
25
18.1
18.2
18.3
18.4
18.5
18.6
18.7
18.0
18.2
18.4
18.6
18.8
19.0
17
18
19
20
17.5
18.0
18.5
19.0
Average
Average
Average
Average
E
K3
K3
36
36
34
34
Difference
32-
Difference
32
30
30
28
28
26
15
16
17
18
19
20
26
16
17
18
19
20
Average
Average
normalization being one of the most significant (Donati et al. 2019, Butz 2022, Butz et al. 2024). Circulating miRNA-based methods would prove especially useful in diseases where differential diagnoses involve difficult, invasive procedures that usually incur a heavy cost on the healthcare system as well, for example in primary aldosteronism (Decmann et al. 2019, Vékony et al. 2024), or where accurate differentiation is of paramount importance for the patient, such as differentiating benign and malignant tumors (e.g., adrenocortical adenoma from carcinoma) (Turai et al. 2022, Fassnacht et al. 2023). Studies on rare diseases involving circulating microRNAs in large cohorts usually require international collaboration, which again underlines the relevance of standardization. As far as we are aware, there are no data on whether dPCR and qPCR would give equivalent results on circulating miRNA expression from the same sample.
According to our results, digital PCR and real-time PCR are not easily interchangeable with each other if the detection should cover a wide dynamic range of expression values. The given proportional bias in the case of miRNAs with very low or very high expression values means some targets require additional validation and calibration. This also incurs that studies done by utilizing dPCR or qPCR cannot directly be validated with the other method without great care in equalizing the two methodologies. This complication would be especially difficult to overcome in studies focusing on miRNA expression patterns, measuring multiple miRNAs covering a wide range of expressions simultaneously. Analyzing combinations of miRNAs compared to individual miRNAs seems to provide better-performing, clinically more relevant biomarkers (Turai et al. 2022, Nyirő et al. 2024, Vékony et al. 2024) that could further exacerbate this problem.
| cel-miR-39-3p | miR-16-5p | miR-21-5p | miR-210-3p | miR-483-5p | ||
|---|---|---|---|---|---|---|
| Raw Ct | K2 | 0.815516 | 1.043507 | 0.973012 | 1.261875 | 0.455473 |
| K3 | 1.283423 | 1.093131 | 1.25111 | 1.526165 | 0.478171 | |
| Copy numbers | K2 1063.763 | 3405.725 | 105.1802 | 5.910047 | 13.00134 | |
| K3 | 1356.381 | 3847.437 | 326.665 | 7.842105 Downloaded from | Bioscientifica.com at 04/02/2026 07:24:41AM 11.2899 | |
A
*
B
*
(normalized to cel-mir-39-3p)
4
p=0.0008219
(normalized to cel-mir-39-3p)
7
p=0.006142
6
5
Q
Q
ACt value
ACt value
5
3
N
1
3
1
E
2
0
1
2
K3
K2
13
C
*
D
ACt value (normalized to cel-mir-39-3p)
-6
p=0.009439
(normalized to cel-mir-39-3p)
p=0.9404
-7
℮
-6
®
G
-8
ACt value
-7
3
6
O
-9
0
-8
-10
0
-9
3
COULD
-11
-10
KO
13
K
The choice of preferred anticoagulant can also have effects on the results, as there were minor differences in measurements from K2 and K3 EDTA regarding circulating miRNAs with both techniques, but the need for standardization using internal and external control
miRNAs in the case of qPCR seems to render this random error more pronounced so it might skew the result of the analyses, resulting in significant differences as shown in our measurements. K2 EDTA shows smaller standard deviation on average and results in higher concordance
A
B
29
p=0.3119
27
p=0.2027
28
O
26
G
raw Ct value
27
®
raw Ct value
25
G
Q
C
26
24
25
23
c
24
C
5
22
€
9
5
23
21
K3
K2
13
C
D
E
*
33
p=0.2525
37
p=0.6868
35
p=0.02965
32
Q
36
€
C
0
0
raw Ct value
31
C
3
raw Ct value
35
raw Ct value
34
C
30
€
34
29
33
C
33
E
E
28
O
e
32
5
8
32
27
℮
0
31
0
26
12
43
30
31
12
K3
KO
K3
A
B
C
5000
p=0.9563
20000
p=0.5958
(normalized to cel-mir-39-3p)
15000
p=0.4304
1500
p=0.2305
2000
p=0.7562
copy number/ul
4000
copy number/ul
15000
copy number/ul
10000
copy number/ul
copy number/ul
(normalized to cel-mir-39-3p)
0
3000
1000
10000
5000
-2000
2000
500
1000
5000
0
-4000
0
0
-5000
0
-6000
K2
K3
K2
K3
K2
K3
K2
K3
K2
K3
D
E
30
p=0.2455
0
p=0.9563
50
p=0.5459
0
p=0.9653
copy number/ul
copy number/ul
(normalized to cel-mir-39-3p)
-1000
20
copy number/ul
copy number/ul
(normalized to cel-mir-39-3p)
40
-1000
-2000
30
-2000
10
-3000
20
-3000
-4000
10
-4000
0
-5000
0
-5000
K2
K3
K2
K3
K2
K3
K2
K3
with dPCR results, but a more extensive study would be needed to determine which type of EDTA would be preferable.
Three out of the five miRNAs studied have altered expression in adrenocortical carcinoma, as miR-483-5p is the most important and reliable miRNA for the differentiation of benign and malignant adrenocortical tumors (Butz et al. 2024). The expression of miR-210 and miR-21 is also different in benign and malignant adrenocortical tumors (Ozata et al. 2011, Turai et al. 2022, Butz et al. 2024). There are also data regarding the use of miR-483-5p and miR-210-3p for follow-up or monitoring treatment efficacy in ACC, mainly in xenograft models (Nagy et al. 2015, Jung et al. 2016). Should these miRNAs enter clinical use for diagnostic or monitoring purposes, it would be pivotal to use the same anticoagulant and methodology to obtain as concordant results as possible. The other two, miR-16-5p and cel-miR-39- 3p, both have potential uses as controls (Yang et al. 2022).
The main limitation of this study was sample size, but even so, we managed to show significant differences. The lack of ACC patients in our cohort could also be regarded as a limitation, but this was fully intentional, as the participants of this study were chosen so that none of the conditions they had (if they had any), according to the literature, would lead to altered expression of any of these four miRNAs. We wanted to limit potential
disrupting factors as much as possible. For example, the choice of miRNA isolation kit could also lead to potential differences due to different effectiveness and quality of the obtained miRNAs (Sriram et al. 2021). In our study, we could limit these factors by placing the blood samples on ice right after acquisition, by quick preparation of collected samples to limit hemolysis and nucleic acid degradation, and by utilizing the same RNA extraction method. Having all the measurements done in the same lab also helped prevent pre-analytical errors.
Conclusion
With this study, we wanted to demonstrate the difficulties and pitfalls of circulating miRNA-based methodology that otherwise shows great diagnostic promise in endocrine diseases (Butz et al. 2024, Vékony et al. 2024). Based on these results, it could be proposed that a unification of methodologies and protocols in circulating miRNA studies would be highly beneficial, for example, the utilization of the same anticoagulant for all the samples included in a study, to avoid as many of these potential differences as possible. A more consensus-based methodology in these experiments, where as many of the details are the same as possible, would highly benefit reproducibility and pave the way for clinical application.
Declaration of interest
Peter Igaz is an Associate Editor of Endocrine-Related Cancer. Peter Igaz was not involved in the review or editorial process for this paper, on which he is listed as an author. BV, GN, HB, BKS, VT, AP and PI have no conflict of interest to report. PF is the founder and CEO of Pharmahungary Group, a group of R&D companies developing miRNA therapeutics.
Funding
The authors acknowledge support from the Hungarian National Research, Development, and Innovation Office (NKFIH) (grants K146906 to PI) and TKP2021-EGA-24 from the National Research, Development, and Innovation Fund by the Ministry of Innovation and Technology of Hungary, financed under the TKP2021-EGA) funding scheme, and also supported by the 2024- 2.1.1 .- EKOP-2020-00004 University Research Scholarship Programme of the Ministry for Culture and Innovation from the source of the National Research, Development and Innovation Fund.
References
Anfossi S, Babayan A, Pantel K, et al. 2018 Clinical utility of circulating non- coding RNAs - an update. Nat Rev Clin Oncol 15 541-563. (https://doi.org/10.1038/s41571-018-0035-x)
Barbarotto E, Schmittgen TD & Calin GA 2008 MicroRNAs and cancer: profile, profile, profile. Int J Cancer 122 969-977. (https://doi.org/10.1002/ijc.23343)
Bereczki Z, Benczik B, Balogh OM, et al. 2025 Mitigating off-target effects of small RNAs: conventional approaches, network theory and artificial intelligence. Br J Pharmaco/ 182 340-379. (https://doi.org/10.1111/bph.17302)
Bland JM & Altman DG 1999 Measuring agreement in method comparison studies. Stat Methods Med Res 8 135-160. (https://doi.org/10.1191/096228099673819272)
Butz H 2022 Circulating noncoding RNAs in pituitary neuroendocrine tumors-two sides of the same coin. Int J Mol Sci 23 5122. (https://doi.org/10.3390/ijms23095122)
Butz H, Patócs A & Igaz P 2024 Circulating non-coding RNA biomarkers of endocrine tumours. Nat Rev Endocrinol 20 600-614. (https://doi.org/10.1038/s41574-024-01005-8)
Conserva F, Gesualdo L & Pontrelli P 2021 Analysis of miRNA expression using digital PCR and the QuantStudio™ 3D digital PCR system. Methods Mol Biol 2325 191-202. (https://doi.org/10.1007/978-1-0716-1507-2_13)
Decmann A, Nyírö G, Darvasi O, et al. 2019 Circulating miRNA expression profiling in primary aldosteronism. Front Endocrinol 10 739. (https://doi.org/10.3389/fendo.2019.00739)
Djordjevic V, Stankovic M, Nikolic A, et al. 2006 PCR amplification on whole blood samples treated with different commonly used anticoagulants. Pediatr Hematol Oncol 23 517-521. (https://doi.org/10.1080/08880010600751900)
Donati S, Ciuffi S & Brandi ML 2019 Human circulating miRNAs real-time qRT-PCR-based analysis: an overview of endogenous reference genes used for data normalization. Int J Mol Sci 20 4353. (https://doi.org/10.3390/ijms20184353)
Fassnacht M, Dekkers OM, Else T, et al. 2018 European Society of Endocrinology Clinical Practice Guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European
Network for the Study of Adrenal Tumors. Eur J Endocrinol 179 G1-G46. (https://doi.org/10.1530/eje-18-0608)
Fassnacht M, Tsagarakis S, Terzolo M, et al. 2023 European Society of Endocrinology clinical practice guidelines on the management of adrenal incidentalomas, in collaboration with the European Network for the Study of Adrenal Tumors. Eur J Endocrinol 189 G1-G42. (https://doi.org/10.1093/ejendo/lvad066)
Goossens W, Duppen V & Verwilghen RL 1991 K2- or K3-EDTA: the anticoagulant of choice in routine haematology? Clin Lab Haemato/ 13 291-295. (https://doi.org/10.1111/j.1365-2257.1991.tb00284.x)
Igaz I & Igaz P 2014 Tumor surveillance by circulating microRNAs: a hypothesis. Cell Mol Life Sci 71 4081-4087. (https://doi.org/10.1007/s00018-014-1682-4)
Jung S, Nagy Z, Fassnacht M, et al. 2016 Preclinical progress and first translational steps for a liposomal chemotherapy protocol against adrenocortical carcinoma. Endocr Relat Cancer 23 825-837. (https://doi.org/10.1530/erc-16-0249)
Lam NY, Rainer TH, Chiu RW, et al. 2004 EDTA is a better anticoagulant than heparin or citrate for delayed blood processing for plasma DNA analysis. Clin Chem 50 256-257. (https://doi.org/10.1373/clinchem.2003.026013)
Lima-Oliveira G, Lippi G, Salvagno GL, et al. 2013 Brand of dipotassium EDTA vacuum tube as a new source of pre-analytical variability in routine haematology testing. Br J Biomed Sci 70 6-9. (https://doi.org/10.1080/09674845.2013.11669922)
Livak KJ & Schmittgen TD 2001 Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods 25 402-408. (https://doi.org/10.1006/meth.2001.1262)
Nagy Z, Baghy K, Hunyadi-Gulyás É, et al. 2015 Evaluation of 9-cis retinoic acid and mitotane as antitumoral agents in an adrenocortical xenograft model. Am J Cancer Res 5 3645-3658.
Nyirő G, Szeredás BK, Decmann Á, et al. 2024 miRNA expression profiling in G1 and G2 pancreatic neuroendocrine tumors. Cancers 16 2528. (https://doi.org/10.3390/cancers16142528)
Özata DM, Caramuta S, Velázquez-Fernández D, et al. 2011 The role of microRNA deregulation in the pathogenesis of adrenocortical carcinoma. Endocr Relat Cancer 18 643-655. (https://doi.org/10.1530/erc-11-0082)
Sriram H, Khanka T, Kedia S, et al. 2021 Improved protocol for plasma microRNA extraction and comparison of commercial kits. Biochem Med 31 030705. (https://doi.org/10.11613/BM.2021.030705)
Turai PI, Herold Z, Nyirő G, et al. 2022 Tissue miRNA combinations for the differential diagnosis of adrenocortical carcinoma and adenoma established by artificial intelligence. Cancers 14 895. (https://doi.org/10.3390/cancers14040895)
Valihrach L, Androvic P & Kubista M 2020 Circulating miRNA analysis for cancer diagnostics and therapy. Mol Aspects Med 72 100825. (https://doi.org/10.1016/j.mam.2019.10.002)
Vékony B, Nyirő G, Herold Z, et al. 2024 Circulating miRNAs and machine learning for lateralizing primary aldosteronism. Hypertension 81 2479-2488. (https://doi.org/10.1161/hypertensionaha.124.23418)
Vrtaric A, Filipi P, Hemar M, et al. 2016 K2-EDTA and K3-EDTA greiner tubes for HbA1c measurement. Lab Med 47 39-42. (https://doi.org/10.1093/labmed/lmv016)
Yang L, Yang S, Ren C, et al. 2022 Deciphering the roles of miR-16-5p in malignant solid tumors. Biomed Pharmacother 148 112703. (https://doi.org/10.1016/j.biopha.2022.112703)