Sage Journals: Discover world-class research

Abstract

Typical and atypical carcinoid tumors belong to the neuroendocrine lung tumors. They have low recurrence and proliferation rate, lymph node, and distant metastases. Nevertheless, these tumors have shown a more aggressive behavior. In the last years, microRNAs were screened as new tumor markers for their potential diagnostic and therapeutic relevance. The expression of hsa-let-7b-5p, hsa-let-7f-5p, hsa-miR-222-3p, and their targets HMGA2 (high-mobility group A2) and CDKN1B (cyclin-dependent kynase inhibitor 1B, p27^kip1) was evaluated in this rare small group of patients. We analyzed the clinical data of all typical and atypical carcinoid tumors of patients who underwent surgical operation at Marburg University Hospital (n = 18) from 2000. Quantitative reverse transcription polymerase chain reaction was performed in formalin-fixed paraffin-embedded tumor tissue versus four tumor-free lung tissue samples. HMGA2 was stable or downregulated; only one patient showed a significant overexpression. CDKN1B showed a significant overexpression or a stable level; it was downregulated in two samples only. Hsa-miR-222-3p resulted almost stable or overexpressed except for two samples (significantly downregulated). Hsa-let-7f-5p was stable or overexpressed in the majority of analyzed samples, whereas hsa-let-7b-5p was significantly downregulated. HMGA2 and CDKN1B are differently expressed between atypical and typical carcinoid tumors, thus representing valid biomarkers for the classification of the two tumor groups. Hsa-let-7f-5p and HMGA2 are inversely correlated. Hsa-miR-222-3p does not correlate with its predicted target CDKN1B.

Keywords

microRNAs Let-7 neuroendocrine lung tumors typical carcinoid atypical carcinoid diagnostic markers CDKN1B HMGA2

Introduction

Neuroendocrine tumors of the lung represent 25% of all neuroendocrine cancer disease. Despite of their low percentage of all lung tumors (2%), its incidence is rising up to 6% each year.¹ The neuroendocrine lung tumors are divided into four groups: typical carcinoid (TC) tumor, atypical carcinoid (AC) tumor, large-cell neuroendocrine (LCNEC) tumors, and small-cell lung cancer (SCLC). In large surgical series, TC (80%) and AC (8%–13%) represent the most common neuroendocrine lung tumors.² TC tumors are characterized by low proliferation rate and small mitosis rate (<2 per 2 mm²) without necrotic areas. The proliferation index Ki-67 is lower than 5%.³ AC tumors show a high proliferation rate, a high mitosis rate (>2 per 2 mm²), and a focal cell necrosis. The proliferation marker Ki-67 is usually lower than 20%.³ The prognosis of patients affected by carcinoid tumors is better if compared with the commonly bad prognosis of LCNEC tumors and SCLC. However, lymph node metastases can occur within the TC tumors.⁴ Unfortunately, despite histological and immune-histological diagnosis, the aggressiveness of the carcinoid tumors can be hardly foreseen.

In recent years, the role of microRNAs (miRNAs) in various malignant tumors has been intensively studied.⁵ Several observations suggest for miRNAs a diagnostic, prognostic, and therapeutic potential.⁶ Further studies highlighted the occurrence of miRNAs’ cluster mutations as a main cause of cancer development.⁷ Up to now, few studies about miRNAs focus on neuroendocrine lung tumors. Dossing et al.⁸ described the downregulation of the miR-129-5p and the let-7 family miRNAs in neuroendocrine lung tumors that leads to an upregulation of their targets Egr1, G3bp1, HMGA2, and BACH1. Altered expression of miR-21, miR-155, and hsa-let-7a was observed in neuroendocrine tumors of the lung. In this study, there were no differences in the expression profile between typical and AC tumors. MiR-21 and miR-155 were differentially expressed according to the histological subtypes of pulmonary neuroendocrine tumors.^9,10 Mairinger et al. screened 763 miRNAs in 12 different neuroendocrine lung tumors (TC, AC, LCNEC, and SCLC). Four showed a positive correlation with the histological types of neuroendocrine tumors: miR-18, miR-15b, miR-335, and miR-1201.¹¹

HMGA2 belongs to a family of small chromatin-associated non-histone proteins and acts as architectural transcription factor.^12,13 HMGA2 is normally expressed during embryogenesis and tissue development and suppressed in normal differentiated tissue. Its overexpression has been shown in several benign and malignant tumors and it is inversely correlated with the expression of Let-7 family miRNAs.^14,15 Let-7 family miRNAs have been shown to be highly expressed in normal thyroid tissue.¹⁶ Up to now, the correlation between HMGA2 and let-7 miRNAs in carcinoid tumors has not been described yet.

CDKN1B (p27^kip1) is responsible for inhibiting the cyclin-dependent kinases and arrest the cell cycle in adverse event of genotoxic stress. Its aberrant expression could exert a pro-oncogenic role.¹⁷ Its mutation is linked with multiple endocrine neoplasia (MEN) syndromes.¹⁸ CDKN1B has been recently identified as reference gene in neuroendocrine lung cancer.¹⁹ It is a target of hsa-miR-222-3p as it has been observed in several tumors.^20–22

We aimed to analyze the expression of hsa-let-7b-5p, hsa-let-7f-5p, and their related target HMGA2 in a panel of typical and AC tumor samples. Furthermore, we wanted to investigate the expression of hsa-miR-222-3p and its target CDKN1B (p27^kip1) in the same group of samples.

Subjects and methods

Basic clinical data collection

In all, 18 patients with typical and AC tumors in a time interval between 2000 and 2014 underwent surgical operation at Marburg University Hospital. On the basis of a complete prospective documentation, we gathered the following data for each patient: sex, age at operation time (years), smoker status at operation time (yes/no), tumor side and localization, pre-operative findings of the positron emission tomography–computed tomography (PET-CT) and the DOTATOC PET-CT (positive/negative), Ki-67 index (%), number of lymph nodes tumor-stricken taken during the operation, as well as the follow-up (between 3 and 15 years). For the parameters, the median and range were determined in each case. Statistical analysis was performed with GraphPad Prism 6.0 (GraphPad Prism, La Jolla, CA, USA).

Sample collection

Tumor tissue was collected when patients, diagnosed for typical and AC tumors, underwent surgical resection of a lung tumor at the University Hospital of Marburg. A total of 8 typical and 10 AC tissue samples were collected between 2000 and 2014 and were all included in the study. The study was conducted under the approval of the Ethics Committee of Marburg University Hospital (No. 68/14), and all patients signed out the inform consent. Four samples of tumor-free lung tissues were collected from formalin-fixed paraffin-embedded blocks obtained from patients who underwent surgical resection of lung adenocarcinoma and metastasis of colon carcinoma. The tissue was found tumor free after pathological examination and was used as control for the following experiments.

RNA isolation

Total RNA, including short RNA, was isolated from typical and atypical pulmonary neuroendocrine tumors; 10 µm slices were cut from formalin-fixed paraffin-embedded blocks and processed by FFPE XS Kit (MACHEREY-NAGEL GmbH & Co. KG, Dueren, Germany) following the manufacturer instructions. Total miRNAs-enriched RNA amount was measured with NanoDrop Lite (Thermo Fisher Scientific, Darmstadt, Germany) and its quality was assessed by the 260/280 absorbance ratio.

In silico analysis

In order to reveal potential targets of these miRNAs, we performed in silico analysis in four independent databases: miRanda (http://www.microrna.org), miRDB (http://mirdb.org/miRDB/), RNA22-HSA (https://cm.jefferson.edu/rna22/), and TargetScanHuman (http://www.targetscan.org).

Quantitative reverse transcription polymerase chain reaction

MiRNA-enriched RNA lysates were reverse transcribed with miScript II RT Kit (Qiagen, Hilden, Germany). Complementary DNA (cDNA) was amplified with miScript SYBR Green PCR Kit using commercially available hsa-let-7b-5p (MS00003122), hsa-let-7f-5p (MS00006489), and miR-222-3p (MS00007609) miScript Primer Assays (Qiagen). RNU6B (MS00029204) was amplified as reference miRNA. For the amplification of HMGA2 and CDKN1B, Sso Fast Eva Green from BioRad (Munich, Germany) was used. HMGA2 (QT01157674), CDKN1B (QT00998445), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH; QT01192646) were purchased from Qiagen; quantitative polymerase chain reaction (qPCR) was run on CFX96 (BioRad). GAPDH was amplified as reference gene transcript and used for normalization of target gene quantification cycles. Tumor-free lung tissue was also processed for miRNAs and target expression.

For final representation, PCR data (fold change compared to tumor-free lung tissue) were classified as follows:

↑ >10, ↑↑ >100, ↑↑↑ >1000, ↑↑↑↑ > 10,000 fold and ↓ <0.5, ↓↓ <0.1, and ↓↓↓ <0.01 fold

Statistical analysis

Statistical analysis of PCR results was performed using CFX Manager (BioRad) and Rest 2008. Significance was calculated using the t-test for paired samples; p < 0.05 was regarded as significant. Each tumor probe was normalized against the mean result of four tumor-free lung tissue samples.

Results

Demographic patient data collection and tumor stage

Median age was 60.5 years (range: 39–71 years) for the TC tumors and about 59 years (range: 31–72 years) for the atypical ones. The gender distribution for TC was 3 men versus 5 women and for AC 5 men versus 5 women. The ratio between the smoker and the non-smokers in both groups was equal: 62.5% of the patients were non-smokers. In the group of the TC tumors, six tumors were detected in the right lobe and two in the left lobe of the lung. Nine of the AC tumors were localized in the right side and only one in the left side. DOTATOC-PET-CT scan was positive in 7 of the TC tumors, whereas 6 of the AC tumors were positive. The fludeoxyglucose (FDG)-PET-CT scan showed a low sensitivity in both kinds of tumors—TC: 7 negative; AC: 5 negative. All patients received a complete resection of the tumors including a radical lymphadenectomy. The TC tumors were classified as G1 tumors. Seven AC tumors were classified as G1, two G2, and one G3. All resected tumors were classified as T1 with a size smaller than 2 cm. The median of resected lymph nodes in TC tumors was 18.5 (range: 4–34) and in AC tumors was 19 (range: 10–39). The lymph nodes were tumor free in 7 of the TC tumors after histological examination. Only in one case, a N1 lymph node was estimated histologically as tumor-stricken. The AC tumors showed one N2 and one N3 tumor infiltrating nodules (for details, see Table 1). After 3 years, all patients were alive without any local recurrence or new distant metastases of the tumor. One patient was disease free after 15 years.

Table 1.

Clinical basic demographic data of the patients (n = 18; comparison between the typical carcinoid (TC; n = 8) and atypical carcinoid (AC; n = 10) groups).

∑n = 18	TC (n = 8)	AC (n = 10)
Gender	3 males/5 females	5 males/5 females
Age (years)	Median: 60.5 (range: 39–71)	Median: 59 (range: 31–83)
Smoker	3 yes/5 no	3 yes/7 no
Side of the tumor	6 right/2 left	9 right/1 left
Tumor localization lung lobe	2 UL/1 ML/5 LL	2 UL/3 ML/5 LL
FDG-PET	7 negative/1 positive	5 negative/5 positive
DOTATOC-PET	1 negative/7 positive	4 negative/6 positive
Grading	8 G1	5 G1/4 G2/1 G3
T	8 T1	8 T1/1 T2/1 T3
N	7 N0/1 N1	8 N0/1 N2/1 N3
Resected lymph nodes	Median: 18.5 (range: 4–34)	19 (range: 10–49)

FDG: fludeoxyglucose; PET: positron emission tomography; UL: upper lobe; ML: middle lobe; LL: lower lobe.

HMGA2 and CDKN1B expression in typical and AC tumors

HMGA2 is a non-transcription factor involved in mediating the transcriptional activity of E2F, thus promoting the proliferation and tumor invasiveness.²³ For this reason, it was found useful to evaluate its expression in neuroendocrine tumors of the lung.

As shown in Figure 1 (lower panel), HMGA2 was detectable in 16 out of 18 samples collected from patients.

Figure 1.

Let-7 miRNAs and HMGA2 expression. RT-qPCR of neuroendocrine tumor resection samples. The graphs show in the upper panels the expression of hsa-let-7f-5p and hsa-let-7b-5p. The samples were normalized to tumor-free lung tissue and to RNU6B as control miRNA. The lower panel shows the expression of HMGA2 transcript. The samples were normalized to tumor-free lung tissue and GAPDH as housekeeping transcript. Shown are means of experiments performed in triplicates ± SEM.

The majority of samples (n = 10) showed a significant downregulation (<0.01-fold) of HMGA2 transcript in comparison to tumor-free lung tissue. Five samples showed an unvaried expression and one sample only showed a significant upregulation (>10-fold; Table 2).

Table 2.

Schematic presentation of miRNAs and predicted target expression in atypical and typical carcinoid tumors.

PTC	Probes	HMGA2	CDKN1B	Hsa-miR-222-3p	Hsa-let7f-5p	Hsa-let7b-5p
Positive		16/18	16/18	18/18	18/18	18/18
1	81741_09	↑	↑↑	±	±	↓↓
2	39301_12	±	↑↑	↑	↓	↓↓
3	09296_12	↓	±	±	±	↓↓
4	19496_10	↓↓	↑↑	±	±	↓↓↓
5	3451_13	↓	↑↑	↓	↑↑	↓↓↓
6	6497_13	∅	↑	↑↑↑	↑↑↑	↓↓↓
7	11578_00	±	∅	↑↑↑↑	↑↑↑↑	↑↑↑↑
8	10444_04	↓↓↓	↓↓	±	↑	↓↓
9	12335_10	↓↓↓	±	±	↓↓↓	↓↓
10	01195_11	↓	±	±	↓↓↓	±
11	6809_11	↓↓↓	↑	±	±	↓↓
12	8377_02	±	↓↓↓	↑↑	±	↓↓
13	1797_10	↓↓	±	↓	↑	↓↓↓
14	3049_12	↓↓↓	±	↑↑	±	↓↓↓
15	752_13	±	↑↑	↓	↓↓	±
16	6389_13	±	↑	↑↑	↑↑	↑
17	2430_03	∅	±	↑↑↑↑	↓↓↓	↑
18	57741_09	↓	∅	↓	±	↓↓

↑ > 10; ↑↑ > 100; ↑↑↑ > 1000; ↑↑↑↑ > 10,000 fold; ↓ <0.5; ↓↓ < 0.1; ↓↓↓ < 0.01 fold.

CDKN1B is responsible, together with CDKN1A (p21^Cip1), for cell-cycle block by inhibiting the cyclin-dependent kinases.²⁴ Its expression could represent a valid biomarker for the slow proliferating carcinoid tumors. Analysis of expression confirmed that CDKN1B was detectable in 16 out of 18 collected samples. Its expression (Figure 2 (lower panel)) was generally stable in six samples. Eight samples showed a significant overexpression (>100-fold) of CDKN1B. Two samples only showed a significant downregulation (<0.1 <0.01-fold) of CDKN1B transcript in comparison to tumor-free lung tissue (Table 2).

Figure 2.

Hsa-miR-222-3p and CDKN1B expression. RT-qPCR of neuroendocrine tumors resection samples. The graph shows in the upper panel the expression of hsa-miR-222-3p. The samples were normalized to tumor-free lung tissue and to RNU6B as control miRNA. The lower panel shows the expression of CDKN1B transcript. The samples were normalized to tumor-free lung tissue and GAPDH as housekeeping transcript. Shown are means of experiments performed in triplicates ± SEM.

HMGA2 is generally downregulated in neuroendocrine tumors and support their low proliferation rate. Furthermore, the stable/upregulated expression of CDKN1B could strongly confirm the apparent steady state of carcinoid tumor and it could be used as valid biomarker for this malignant disease.

Expression of hsa-let-7f-5p and hsa-let-7b-5p in typical and AC tumors

Let-7 miRNAs exert a tumor-suppressor role^7,14 and their detection could represent a valuable diagnostic marker. Let-7 miRNAs were detectable in all samples of patients affected by carcinoid tumors (n = 18) included in the study. Hsa-let-7f-5p was almost stable (n = 7) or overexpressed (n = 6) in carcinoid tumor samples (>100-fold). Its expression was downregulated in five samples only (<0.01-fold; Figure 1 (upper left panel)).

As shown in Figure 1 (upper right panel), hsa-let-7b-5p, another member of Let-7 family, resulted downregulated in the majority of samples (n = 13; <0.01-fold). Only three samples showed a significant upregulation of it (>100-fold). Two samples showed a stable expression in comparison to tumor-free lung tissue used as control.

Figure 2 (upper panel) shows the expression of hsa-miR-222-3p, a miRNA with oncogenic property. The level of hsa-miR-222-3p was detectable in all patient samples (n = 18). Its expression was found stable or upregulated in the majority of samples (n = 14; >100-fold). Four samples showed a downregulation of the above-mentioned miRNA (<0.1-fold; Table 2).

The general stable/overexpression of hsa-let-7f-5p could represent a valid biomarker for carcinoid tumor and could aim to become a positive prognostic factor. Moreover, its significant upregulation could be inversely correlated with the significant low expression of its validated target HMGA2 (Figure 1 (lower panel)).

Hsa-miR-222-3p overexpression could characterize carcinoid tumor profile and represent a future marker for the diagnosis of this aggressive malignancy. No correlation was found with its putative target CDKN1B.

Distribution of miRNAs and predicted targets by atypical and typical subgroups

The expression of HMGA2, CDKN1B, hsa-miR-222-3p, hsa-let-7f-5p, and hsa-let-7b-5p was analyzed independently in the two subgroups of AC and TC tumors. The median of quantified expression was calculated after excluding oversized values. As shown in Table 3, the AC tumors showed a significant downregulation (0.285) of HMGA2, whereas the typical showed a stable expression (0.775). Additionally, CDKN1B overexpression was observed in TC tumors (2.43). The atypical tumors were characterized by a stable expression of CDKN1B (1.33).

Table 3.

Median expression of HMGA2, CDKN1B, and miRNAs in atypical and typical carcinoid tumors.

Carcinoid tumors	Probes	HMGA2
Atypical	8	0.285^a
Typical	8	0.775^b
		CDKN1B
Atypical	7	1.33^b
Typical	7	2.43^c
		Hsa-let-7b-5p
Atypical	8	0.15^a
Typical	7	0.14^a
		Hsa-miR-222-3p
Atypical	6	1.095^b
Typical	6	1.155^b
		Hsa-let-7f-5p
Atypical	8	1.53^b
Typical	6	0.655^b

MiRNA: microRNA.

Downregulated.

Unvaried.

Upregulated.

Concerning the analyzed miRNAs, hsa-miR-222-3p was stably expressed in both tumor subgroups. Interestingly, hsa-let-7f-5p was slightly downregulated (0.655) in typical tumors and stably expressed in atypical (1.53). No differences were observed for hsa-let-7b-5p that was significantly suppressed in both tumor subgroups.

To summarize, it could be observed that AC tumors could be classified for their low expression rate of HMGA2 and stable rate of CDKN1B. TC tumors could be grouped for their stable expression of HMGA2 and high rate of CDKN1B.

Discussion

Our study examined the expression of two miRNAs belonging to Let-7 family and hsa-miR-222-3p in typical and AC tumors. It has been previously shown that Let-7 family miRNAs are normally expressed in normal adult tissue and they exert a tumor-suppressor role. Their suppression is correlated with tumor progression and invasiveness. The re-expression of let-7 miRNAs lead to cell death and loss of tumor aggressiveness.^14,25

Moreover, Let-7 miRNAs are able to target not only oncogenes but also transcripts of gene expressing membrane transporters like SLC5A5. In silico analysis has shown that SLC5A5 is a target of Let-7 family miRNAs. Interestingly, it was found that SLC5A5 is inversely correlated with miRNAs belonging to Let-7 family in thyroid cancer.²⁶

Here, it was shown a stable or upregulated expression of hsa-let-7f-5p, whereas hsa-let-7b-5p resulted strongly suppressed in human carcinoid tumors. Interestingly, it was observed that HMGA2, a predicted target of hsa-let-7 miRNAs, is strongly suppressed in all analyzed samples of AC tumor included in this study. This study furthermore confirmed the inverse correlation existing between hsa-let-7f-5p and HMGA2, while no correlation was found with hsa-let-7b-5p. This finding is supported by previously published studies showing that suppression of Let-7 miRNAs restores HMGA2 and promotes tumor metastasis in neuroendocrine tumors.^8,27 Interestingly, the expression of HMGA2 could represent a marker to characterize the TC tumors because of its stable expression, whereas its level is significantly low in AC tumors.

Additionally, the expression of hsa-miR-222-3p was analyzed. The majority of patient samples showed a significant overexpression of this miRNA. Furthermore, in silico analysis showed that CDKN1B represents one of the putative targets of hsa-miR-222-3p.

Analysis of CDKN1B showed a stable and/or strong significant overexpression of its transcript. Interestingly, the expression of cyclin-dependent kinase (CDK) inhibitor p27 could block the cell cycle and keep the cells in a steady state or at low proliferation rate as previously shown.²⁸ Additionally, the high expression rate of CDKN1B could exert a CDK-independent role with oncogenic properties as previously shown¹⁷ that could also clarify its overexpression in pulmonary neuroendocrine tumors. The results showed that no correlation was found between hsa-miR-222-3p and CDKN1B, which could not be clarified in this study. It could be possible that CDKN1B is under the control of other miRNAs expressed in neuroendocrine tumors of the lung as shown in prostate cancer for miR-24.²⁹ Furthermore, CDKN1B was differently expressed in the two subgroups of carcinoid tumors; in particular, its level was significantly high in TC tumors, whereas it was found stable in atypical tumors.

Despite the small group of study sample, we could clarify the expression status of two of the miRNAs belonging to Let-7 family and hsa-miR-222-3p in carcinoid tumors. First, the general expression of hsa-let-7f-5p could be considered as valid biomarker in patient samples showing a correlated suppression of its predicted target HMGA2, as shown here for AC tumors.

Furthermore, the overexpression of hsa-miR-222-3p could be a valid diagnostic marker as shown for other disorders and tumors.^30–32 Finally, the expression of CDKN1B would represent a double-edged diagnostic factor and/or predictor of therapy response, as shown for osteosarcoma, breast cancer, and ovarian cancer,^33–35 especially for the identification of TC tumors showing higher CDKN1B expression than the atypical tumors. This issue needs to be confirmed in further studies on lung carcinoid tumors.

Footnotes

Acknowledgements

Pietro Di Fazio, C.M., and Andreas Kirschbaum analyzed the data; Moritz Maass and Silvia Roth performed the experiments; Peter Rexin and Joana Grups collected the samples; and Pietro Di Fazio, Christian Meyer, Detlef K Bartsch, and Andreas Kirschbaum wrote and critically revised the manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the Ethics Committee of Marburg University Hospital (No. 68/14) and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. All patients signed out the formal consent.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study received no funds from third party and it was entirely supported by internal university funds.

References

Caplin

Baudin

Ferolla

et al . Pulmonary neuroendocrine (carcinoid) tumors: European Neuroendocrine Tumor Society expert consensus and recommendations for best practice for typical and atypical pulmonary carcinoids. Ann Oncol 2015; 26: 1604–1620.

Cueto

Burigana

Nicolini

et al . Neuroendocrine tumors of the lung: hystological classification, diagnosis, traditional and new therapeutic approaches. Curr Med Chem 2014; 21: 1107–1116.

Hendifar

Marchevsky

Tuli

. Neuroendocrine tumors of the lung: current challenges and advances in the diagnosis and management of well-differentiated disease. J Thorac Oncol 2016; 12: 425–436.

Filosso

Ferolla

Guerrera

et al . Multidisciplinary management of advanced lung neuroendocrine tumors. J Thorac Dis 2015; 7: S163–S171.

Gottardo

Liu

Ferracin

et al . Micro-RNA profiling in kidney and bladder cancers. Urol Oncol 2007; 25: 387–392.

Croce

. Oncogenes and cancer. N Engl J Med 2008; 358: 502–511.

Calin

Croce

. MicroRNA signatures in human cancers. Nat Rev Cancer 2006; 6: 857–866.

Dossing

Binderup

Kaczkowski

et al . Down-regulation of miR-129-5p and the let-7 family in neuroendocrine tumors and metastases leads to up-regulation of their targets Egr1, G3bp1, Hmga2 and Bach1. Genes 2014; 6: 1–21.

Lee

Roh

. Altered expression of PTEN and its major regulator microRNA-21 in pulmonary neuroendocrine tumors. Korean J Pathol 2014; 48: 17–23.

10.

Lee

et al . Altered expression of microRNA miR-21, miR-155, and let-7a and their roles in pulmonary neuroendocrine tumors. Pathol Int 2012; 62: 583–591.

11.

Mairinger

Ting

Werner

et al . Different micro-RNA expression profiles distinguish subtypes of neuroendocrine tumors of the lung: results of a profiling study. Mod Pathol 2014; 27: 1632–1640.

12.

Chiappetta

Ferraro

Vuttariello

et al . HMGA2 mRNA expression correlates with the malignant phenotype in human thyroid neoplasias. Eur J Cancer 2008; 44: 1015–1021.

13.

Jin

Lloyd

Nassar

et al . HMGA2 expression analysis in cytological and paraffin-embedded tissue specimens of thyroid tumors by relative quantitative RT-PCR. Diagn Mol Pathol 2011; 20: 71–80.

14.

Di Fazio

Montalbano

Neureiter

et al . Downregulation of HMGA2 by the pan-deacetylase inhibitor panobinostat is dependent on hsa-let-7b expression in liver cancer cell lines. Exp Cell Res 2012; 318: 1832–1843.

15.

Pallante

Sepe

Puca

et al . High mobility group A proteins as tumor markers. Front Med 2015; 2: 15.

16.

Marini

Luzi

Brandi

. MicroRNA role in thyroid cancer development. J Thyroid Res 2011; 2011: 407123.

17.

Sharma

Pledger

. The non-canonical functions of p27(Kip1) in normal and tumor biology. Cell Cycle 2016; 15: 1189–1201.

18.

Lee

Pellegata

. Multiple endocrine neoplasia syndromes associated with mutation of p27. J Endocrinol Invest 2013; 36: 781–787.

19.

Walter

Werner

Vollbrecht

et al . ACTB, CDKN1B, GAPDH, GRB2, RHOA and SDCBP were identified as reference genes in neuroendocrine lung cancer via the nCounter technology. PLoS ONE 2016; 11: e0165181.

20.

Zhang

Zhou

et al . The alteration of miR-222 and its target genes in nickel-induced tumor. Biol Trace Elem Res 2013; 152: 267–274.

21.

Liang

et al . miR-221/222 promotes S-phase entry and cellular migration in control of basal-like breast cancer. Molecules 2014; 19: 7122–7137.

22.

Moses

Evans

Slone

et al . Bone marrow microenvironment niche regulates miR-221/222 in acute lymphoblastic leukemia. Mol Cancer Res 2016; 14: 909–919.

23.

Fedele

Palmieri

Fusco

. HMGA2: a pituitary tumour subtype-specific oncogene? Mol Cell Endocrinol 2010; 326: 19–24.

24.

Cheng

Lee

Hsia

et al . Methylation of histone H3 lysine 27 associated with apoptosis in osteosarcoma cells induced by staurosporine. Histol Histopathol 2009; 24: 1105–1111.

25.

Chien

Wang

Chu

et al . Lin28B/Let-7 regulates expression of Oct4 and Sox2 and reprograms oral squamous cell carcinoma cells to a stem-like state. Cancer Res 2015; 75: 2553–2565.

26.

Damanakis

Eckhardt

Wunderlich

et al . MicroRNAs let7 expression in thyroid cancer: correlation with their deputed targets HMGA2 and SLC5A5. J Cancer Res Clin Oncol 2016; 142: 1213–1220.

27.

Rahman

Qian

Wang

et al . Frequent overexpression of HMGA1 and 2 in gastroenteropancreatic neuroendocrine tumours and its relationship to let-7 downregulation. Br J Cancer 2009; 100: 501–510.

28.

Satoh

Kaida

. Upregulation of p27 cyclin-dependent kinase inhibitor and a C-terminus truncated form of p27 contributes to G1 phase arrest. Sci Rep 2016; 6: 27829.

29.

Lynch

McKenna

Walsh

et al . MiR-24 regulates CDKN1B/p27 expression in prostate cancer. Prostate 2016; 76: 637–648.

30.

Saiselet

Gacquer

Spinette

et al . New global analysis of the microRNA transcriptome of primary tumors and lymph node metastases of papillary thyroid cancer. BMC Genomics 2015; 16: 828.

31.

Kara

Yumrutas

Ozcan

et al . Differential expressions of cancer-associated genes and their regulatory miRNAs in colorectal carcinoma. Gene 2015; 567: 81–86.

32.

Xue

Wei

Ding

et al . MicroRNA-19b/221/222 induces endothelial cell dysfunction via suppression of PGC-1α in the progression of atherosclerosis. Atherosclerosis 2015; 241: 671–681.

33.

Di Fazio

Waldegger

Jabari

et al . Autophagy-related cell death by pan-histone deacetylase inhibition in liver cancer. Oncotarget 2016; 7: 28998–29010.

34.

Kurozumi

Inoue

Takei

et al . ER, PgR, Ki67, p27(Kip1), and histological grade as predictors of pathological complete response in patients with HER2-positive breast cancer receiving neoadjuvant chemotherapy using taxanes followed by fluorouracil, epirubicin, and cyclophosphamide concomitant with trastuzumab. BMC Cancer 2015; 15: 622.

35.

Hafez

Alhoshani

Al-Hosaini

et al . SKP2/P27Kip1 pathway is associated with advanced ovarian cancer in Saudi patients. Asian Pac J Cancer Prev 2015; 16: 5807–5815.