**6. Future challenges: new reliable miRNAs for target therapies**

Epithelial-to-mesenchymal transition (EMT) allows tumor cells to enter the metastatic cascade by changing their morphological and molecular characteristics, and it represents a wide spectrum that cancer cells keep transiting. The EMT program relies on an intricate network of signaling pathways that dictate a series of phenotypic changes in epithelial cells. One clearly visible aspect is the loss of apical-basal polarity and cell-to-cell interaction caused by the destabilization of tight junctions, decreased claudin, occludin, and E-cadherin repression by SNAIL, SLUG, ZEB, TWIST, and SMAD, and its interchange with N-cadherin, a process known as the "cadherin switch" [90, 91]. Moreover, the overexpression of vimentin, a cytoskeleton intermediate-filament protein of mesenchymal origin that connects the nucleus to the plasma membrane, enables filopodia formation and a fibroblastic spindle-like morphology. Vimentin was proved to be induced through Slug and Ras signaling, and it promotes cell movement and migration [92]. The next step is the degradation of extracellular matrix components under the activity of matrix metalloproteinases and the invasion of the surrounding stroma (**Figure 3**).

The EMT is coordinated by a series of signaling pathways triggered by either the tumor microenvironment or the intrinsic factors, and it falls under the incidence of regulatory noncoding RNAs, especially micro-RNAs [93].

This is the reason why finding potential microRNA targets for blocking EMT could support the efforts already made in blocking metastasis and improving cancer therapy strategies. Right below, we are succinctly reviewing the results of several studies that investigated the role of microRNAs in EMT and metastasis and their potential in becoming microRNA therapy targets (**Table 2**).

The influence of cancer-associated fibroblasts (CAFs) secreted exosomes over endometrial cancer progression was questioned in a study by Li et al. [94]. CAFs secreted exosomes contained significantly lower levels of miR-148b than normal fibroblasts, and miR-148b expression was lower in endometrial cancer specimens than in normal adjacent tissues. miR-148b was correlated with improved prognosis, *in vitro* and *in vivo* studies suggesting its role as an EMT inhibitor. Downregulation of DNMT1 oncogene was the mechanism proposed for miR-148b-mediated suppression of endometrial cancer progression. Also, relating to the tumor microenvironment, emerging evidence shows the prometastatic effect of a hypoxic

#### **Figure 3.** *The EMT-MET plasticity of tumor cells, their migration, and invasion as tumor circulating cells (CTCs).*

microenvironment over tumor cells. Acidic conditions were shown to promote miR-210 overexpression by activating HIF1 [95, 96]. In prostate cancer, high levels of miR-210 were detected in bone metastases, and they were correlated with poor prognosis of prostate cancer patients. Exogenously overexpression of miR-210-3p in cancer cell lines enhanced cell motility and migration as well as bone metastasis in mouse model by inducing NF-κB signaling and EMT. Moreover, miR-210-3p inhibition reversed EMT and impaired the metastatic potential of cancer cells [95]. Another study confirmed the hypoxia-induced EMT activation and metastasis by HIF1-miR-210 axis in breast cancer. Tang et al. [96] identified high expression of miR-210 in hypoxia grown breast cancer stem cells and in isolated human breast cancer stem cells. The overexpression of MiR-210 in poorly metastatic MCF7 cell lines leads to their invasiveness and migration *in vitro* as well as increasing metastatic potential *in vivo*. One suggested mechanism is related to the direct binding of miR-210 to the Open Reading Frame (ORF) of the E-cadherin mRNA and its post-transcriptional inhibition in breast cancer cells.

miR-652 downregulation in the acidic microenvironment of pancreatic cancer promoted EMT by ZEB1 activation, and it was correlated with a progressive stage, lymphatic invasion, vascular infiltration, and distant metastasis. *In vitro* experiments showed that miR-652 overexpression has an antimetastatic effect by inducing MET in PANC-1-A miR-652-mimic transfected cells, and it reduced their migration and invasion. Also, *in vivo* studies confirmed the *in vitro* results with lower tumor mass, fewer metastases, and overexpression of E-cadherin over vimentin/N-cadherin in mouse xenografts PANC-1-A miR-652-mimic transfected cells vs. miR-652 inhibitor [97].

Partially due to cancer tumors heterogeneity in construction and behavior and the relative novelty of noncoding RNAs as potential targets for cancer therapy, there are still missing bricks in understanding the mechanism that triggers the regulation of cancer-related microRNAs. However, as shown above, tumor microenvironment has a great impact over cancer progression, and probably understanding its role in modulating microRNA expression in tumor cells might bring light to new potential targets for improved therapy.

From a different perspective, there are many well-known pathways underlying cancer development and progression. Wnt signaling pathway represents one of the fundamental pathways involved in cell proliferation and specialization, as well as cell movement during both embryonic development and tissue homeostasis. The canonical Wnt signaling functions by regulating the amount of the transcriptional coactivator β-catenin, a molecule that controls key developmental gene expression programs [98]. Wnt signaling aberrations have been shown to regulate various processes that are important for cancer progression, including tumor initiation, tumor growth, cell senescence, cell death, differentiation, and metastasis. Wnt signaling molecules and downstream effectors can promote transcriptional changes in order to induce EMT in cancer cells while also being further activated by EMT in a continuous feedback loop [99].

Several microRNAs were proven to modulate Wnt signaling in EMT and cancer metastasis. In endometrial cancer, Wnt signaling can be activated by miR-652, which targets and inhibits retinoid orphan nuclear receptor alpha (RORA) gene. RORA represents a tumor-suppressor gene that represses the Wnt/β-catenin pathway through attenuating β-catenin transcriptional activity. Expression of miR-652 is frequently increased in human endometrial cancer tissues, its high expression being correlated with poor tumor differentiation, shorter overall survival, and recurrence. Overexpression of miR-652 in endometrial cancer cell promotes their proliferation and migration *in vitro* and *in vivo* [100].

Another metastasis enhancer, miR-374a, acts by activating the Wnt/β-catenin cascade and promoting EMT. miR-374a maintains constitutively activated

**45**

*MiRNA-Based Therapeutics in Oncology, Realities, and Challenges*

Wnt/β-catenin signaling by suppressing multiple negative regulators including WIF1, PTEN, and WNT5A. miR-374a was upregulated in primary tumor samples from breast cancer patients with distant metastases, and it was associated with poor metastasis-free survival. miR-374a transfection into poorly metastatic MCF7 cell line promoted its motility and invasiveness *in vitro* and lung metastasis forming abilities in BALB/c mice, while miR-374a knockdown in highly invasive MDA-MB-231 cells decreased their motility and metastatic potential [101].

On the other hand, microRNA-590-5p was found to function as a tumor-suppressor in breast cancer, inhibiting EMT, cell migration, and invasion by downregulating the Wnt/β-catenin pathway. Quantitative RT-PCR analysis on breast tumor tissues and paired adjacent normal tissues showed that miR-590-5p was downregulated in breast cancer together with E-cadherin, while its target PITX2, b-catenin, Wnt-1, N-cadherin, and vimentin were upregulated. *In vitro* experiments and mouse xenografts showed that miR-590-5p upregulation or PITX2 silencing inhibits the activation of Wnt/β-catenin signaling pathway and suppresses the EMT of breast cancer [102]. Another metastasis inhibitor, miR-625, was characterized in different pathologies, and surprisingly, it maintained its function, even though it was involved in different pathways. miR-625 was reported as downregulated in hepatocellular carcinoma, gastric cancer, and colorectal cancer, and its low expression was associated with local invasion, lymph node, and distant metastasis. Ectopic expression of miR-625 induces suppression of migration and invasion of hepatocellular carcinoma cells by post-transcriptionally inhibiting IGF2BP1. Loss of IGF2BP1 suppressed F-actin polymerization, inhibiting the formation of cell protrusions, required for cell migration. The same effect was observed *in vivo*, where miR-625 overexpression decreased intrahepatic and lung metastasis [103]. Moreover, when it was ectopically induced, miR-625 suppressed the migration and invasion of gastric cancer cells as well as metastasis in nude mice by inhibiting ILK protein synthesis [98]. While the exact signaling pathway was not fully elucidated in colorectal cancer, ectopic miR-625 expression inhibited cell migration and invasion and suppressed colorectal cancer cell metastasis in nude mice [99]. The aforementioned findings highlight miR-625 as an interesting candidate for further *in vivo* studies in order to test its potential for

developing a therapeutic microRNA for blocking invasion and metastasis.

A dual and somewhat contradictory behavior can be observed in the case of miR-409. While low miR-409 expression in breast cancer and nonsmall cell lung carcinoma (NSCLC) was associated with poorer prognosis and its ectopic uptake decreased the invasiveness of cancer cells [104, 105], it also seems to exert the negative effect in the case of prostate cancer, where it promotes tumorigenesis and EMT [106, 107]. Qi Song et al. [104] showed that miR-409 inhibits NSCLC cell migration, growth, and proliferation abilities by inhibiting SPIN1 translation. miR-409 downregulates PI3K/AKT pathway in NSCLC and inhibits its downstream targets such as CREB1, BCL2, and Cyclin D. Overexpression of miR-409 led to fewer lung metastases in nude mice, confirming its antimetastatic potential. Interestingly, miR-409 targeted the same pathway and suppressed cell growth and invasion in breast cancer. miR-409-3p inhibits the proliferation, migration, and invasion of breast cancer by targeting and suppressing the AKT expression. miR-409-3p was downregulated in several human tumors compared to their corresponding nontumor tissues [105]. On the other hand, miR-409-entrapped exosomes secreted by cancerassociated fibroblasts (CAFs) and promoted EMT and prostate tumorigenesis. In their study, Josson et al. demonstrated that miR-409-3p was highly expressed in CAFs derived from human patients, and it was correlated with higher Gleason score in prostatic tissues. Moreover, the ectopic expression of miR-409 in normal prostate stromal fibroblasts conferred them a CAF phenotype *in vitro*. Exosome-mediated transport of miR-409 into normal prostate stromal cells induced cell growth and

*DOI: http://dx.doi.org/10.5772/intechopen.81847*

#### *MiRNA-Based Therapeutics in Oncology, Realities, and Challenges DOI: http://dx.doi.org/10.5772/intechopen.81847*

*Antisense Therapy*

microenvironment over tumor cells. Acidic conditions were shown to promote miR-210 overexpression by activating HIF1 [95, 96]. In prostate cancer, high levels of miR-210 were detected in bone metastases, and they were correlated with poor prognosis of prostate cancer patients. Exogenously overexpression of miR-210-3p in cancer cell lines enhanced cell motility and migration as well as bone metastasis in mouse model by inducing NF-κB signaling and EMT. Moreover, miR-210-3p inhibition reversed EMT and impaired the metastatic potential of cancer cells [95]. Another study confirmed the hypoxia-induced EMT activation and metastasis by HIF1-miR-210 axis in breast cancer. Tang et al. [96] identified high expression of miR-210 in hypoxia grown breast cancer stem cells and in isolated human breast cancer stem cells. The overexpression of MiR-210 in poorly metastatic MCF7 cell lines leads to their invasiveness and migration *in vitro* as well as increasing metastatic potential *in vivo*. One suggested mechanism is related to the direct binding of miR-210 to the Open Reading Frame (ORF) of the E-cadherin mRNA and its

miR-652 downregulation in the acidic microenvironment of pancreatic cancer promoted EMT by ZEB1 activation, and it was correlated with a progressive stage, lymphatic invasion, vascular infiltration, and distant metastasis. *In vitro* experiments showed that miR-652 overexpression has an antimetastatic effect by inducing MET in PANC-1-A miR-652-mimic transfected cells, and it reduced their migration and invasion. Also, *in vivo* studies confirmed the *in vitro* results with lower tumor mass, fewer metastases, and overexpression of E-cadherin over vimentin/N-cadherin in mouse xenografts PANC-1-A miR-652-mimic transfected cells vs. miR-652 inhibitor [97]. Partially due to cancer tumors heterogeneity in construction and behavior and the relative novelty of noncoding RNAs as potential targets for cancer therapy, there are still missing bricks in understanding the mechanism that triggers the regulation of cancer-related microRNAs. However, as shown above, tumor microenvironment has a great impact over cancer progression, and probably understanding its role in modulating microRNA expression in tumor cells might bring light to new potential

From a different perspective, there are many well-known pathways underlying cancer development and progression. Wnt signaling pathway represents one of the fundamental pathways involved in cell proliferation and specialization, as well as cell movement during both embryonic development and tissue homeostasis. The canonical Wnt signaling functions by regulating the amount of the transcriptional coactivator β-catenin, a molecule that controls key developmental gene expression programs [98]. Wnt signaling aberrations have been shown to regulate various processes that are important for cancer progression, including tumor initiation, tumor growth, cell senescence, cell death, differentiation, and metastasis. Wnt signaling molecules and downstream effectors can promote transcriptional changes in order to induce EMT in cancer cells while also being further activated by EMT in

Several microRNAs were proven to modulate Wnt signaling in EMT and cancer metastasis. In endometrial cancer, Wnt signaling can be activated by miR-652, which targets and inhibits retinoid orphan nuclear receptor alpha (RORA) gene. RORA represents a tumor-suppressor gene that represses the Wnt/β-catenin pathway through attenuating β-catenin transcriptional activity. Expression of miR-652 is frequently increased in human endometrial cancer tissues, its high expression being correlated with poor tumor differentiation, shorter overall survival, and recurrence. Overexpression of miR-652 in endometrial cancer cell promotes their

Another metastasis enhancer, miR-374a, acts by activating the Wnt/β-catenin

cascade and promoting EMT. miR-374a maintains constitutively activated

post-transcriptional inhibition in breast cancer cells.

targets for improved therapy.

a continuous feedback loop [99].

proliferation and migration *in vitro* and *in vivo* [100].

**44**

Wnt/β-catenin signaling by suppressing multiple negative regulators including WIF1, PTEN, and WNT5A. miR-374a was upregulated in primary tumor samples from breast cancer patients with distant metastases, and it was associated with poor metastasis-free survival. miR-374a transfection into poorly metastatic MCF7 cell line promoted its motility and invasiveness *in vitro* and lung metastasis forming abilities in BALB/c mice, while miR-374a knockdown in highly invasive MDA-MB-231 cells decreased their motility and metastatic potential [101].

On the other hand, microRNA-590-5p was found to function as a tumor-suppressor in breast cancer, inhibiting EMT, cell migration, and invasion by downregulating the Wnt/β-catenin pathway. Quantitative RT-PCR analysis on breast tumor tissues and paired adjacent normal tissues showed that miR-590-5p was downregulated in breast cancer together with E-cadherin, while its target PITX2, b-catenin, Wnt-1, N-cadherin, and vimentin were upregulated. *In vitro* experiments and mouse xenografts showed that miR-590-5p upregulation or PITX2 silencing inhibits the activation of Wnt/β-catenin signaling pathway and suppresses the EMT of breast cancer [102].

Another metastasis inhibitor, miR-625, was characterized in different pathologies, and surprisingly, it maintained its function, even though it was involved in different pathways. miR-625 was reported as downregulated in hepatocellular carcinoma, gastric cancer, and colorectal cancer, and its low expression was associated with local invasion, lymph node, and distant metastasis. Ectopic expression of miR-625 induces suppression of migration and invasion of hepatocellular carcinoma cells by post-transcriptionally inhibiting IGF2BP1. Loss of IGF2BP1 suppressed F-actin polymerization, inhibiting the formation of cell protrusions, required for cell migration. The same effect was observed *in vivo*, where miR-625 overexpression decreased intrahepatic and lung metastasis [103]. Moreover, when it was ectopically induced, miR-625 suppressed the migration and invasion of gastric cancer cells as well as metastasis in nude mice by inhibiting ILK protein synthesis [98]. While the exact signaling pathway was not fully elucidated in colorectal cancer, ectopic miR-625 expression inhibited cell migration and invasion and suppressed colorectal cancer cell metastasis in nude mice [99]. The aforementioned findings highlight miR-625 as an interesting candidate for further *in vivo* studies in order to test its potential for developing a therapeutic microRNA for blocking invasion and metastasis.

A dual and somewhat contradictory behavior can be observed in the case of miR-409. While low miR-409 expression in breast cancer and nonsmall cell lung carcinoma (NSCLC) was associated with poorer prognosis and its ectopic uptake decreased the invasiveness of cancer cells [104, 105], it also seems to exert the negative effect in the case of prostate cancer, where it promotes tumorigenesis and EMT [106, 107]. Qi Song et al. [104] showed that miR-409 inhibits NSCLC cell migration, growth, and proliferation abilities by inhibiting SPIN1 translation. miR-409 downregulates PI3K/AKT pathway in NSCLC and inhibits its downstream targets such as CREB1, BCL2, and Cyclin D. Overexpression of miR-409 led to fewer lung metastases in nude mice, confirming its antimetastatic potential. Interestingly, miR-409 targeted the same pathway and suppressed cell growth and invasion in breast cancer. miR-409-3p inhibits the proliferation, migration, and invasion of breast cancer by targeting and suppressing the AKT expression. miR-409-3p was downregulated in several human tumors compared to their corresponding nontumor tissues [105]. On the other hand, miR-409-entrapped exosomes secreted by cancerassociated fibroblasts (CAFs) and promoted EMT and prostate tumorigenesis. In their study, Josson et al. demonstrated that miR-409-3p was highly expressed in CAFs derived from human patients, and it was correlated with higher Gleason score in prostatic tissues. Moreover, the ectopic expression of miR-409 in normal prostate stromal fibroblasts conferred them a CAF phenotype *in vitro*. Exosome-mediated transport of miR-409 into normal prostate stromal cells induced cell growth and


**47**

**miRNA**

miR- 374a miR- 590-5p miR-625

Hepatocellular

Metastasis

Inhibits

PTEN and Akt signaling

downregulation,

inhibiting F-actin

polymerization

IGF2BP1

inhibitor

carcinoma

miR-625 miR-625

Colorectal

Metastasis

—

—

Reduced migration in cancer cell lines;

fewer and smaller liver metastases in

mouse xenografts

inhibitor

cancer

Gastric cancer

Metastasis

Inhibits ILK

Suppressed LIMS1-ILKparvin axis signaling

inhibitor

Breast cancer

Metastasis inhibitor

Inhibits PITX2

Blocks Wnt signaling induced by PITX2

Upregulation of miR-590 inhibits EMT genes, inhibits proliferation after 48-72 h, impairs migration and invasion of cancer cells, and promotes smaller tumor size for miR-590 mimic or PITX2 inhibition and less lung and lymphatic metastases in nude mice

Suppressed migration and invasion,

Downregulated in HCC, correlated

[103]

with aggressiveness of cancer and

tumor metastasis

Downregulated in gastric cancer,

[98]

associated with lymph-node

metastasis; no relation with tumor

localization, differentiation, and

local invasion

Downregulated in cancer tissues,

[99]

associated with lymph node and

liver metastases and lower overall

survival rate in 5 years

proliferation was not influenced in cell

lines, decreased intrahepatic and lung

metastasis in mice xenografts

Reduced migration and invasion rate *in vitro*, suppressed motility and

extravasation from vessels, fewer lung

metastatic nodes

Breast cancer

Metastasis promoter

Inhibits PTEN, WIF1, WNT5A

Wnt-b-catenin signaling, promoting the transcriptional activity of TCF/LEF

Induces EMT and enhanced motility in MCF7 transfected cells, decreases MDA cells motility in knockout, promotes lung metastases with MCF7 miR-374a transfected cells, impaired MDA-435 metastases by the administration of antagomiR

Upregulated in patients that presented metastases within the 51 months follow up; associated with low WIF1, PTEN, and WNT5A expression in tumor samples; high b-catenin low E-cadherin in miR-

**Pathology**

**Effect**

**Target**

**Mechanism**

**Effect** *in vitro* **and** *in vivo*

**Clinical associations**

**Ref.** [101]

*MiRNA-Based Therapeutics in Oncology, Realities, and Challenges*

Downregulated in breast cancer tissues vs. normal tissue, together with E-cadherin, while PITX2, b-catenin, Wnt-1, N-cadherin, and vimentin are upregulated

*DOI: http://dx.doi.org/10.5772/intechopen.81847*

[102]

374a overexpressing samples


#### *MiRNA-Based Therapeutics in Oncology, Realities, and Challenges DOI: http://dx.doi.org/10.5772/intechopen.81847*

*Antisense Therapy*

**46**

**miRNA**

miR-148b miR-210-3p miR-210

Breast cancer

Metastasis

Inhibits

Blocks E-cadherin mRNA

Upregulated in mammosphere

cells, induced by hypoxia, promotes

invasion, migration, proliferation, and

self-renewal, induces EMT by loss of

E-cadherin *in vitro* and leads to poorly

differentiated tumors, high proliferation,

more metastases and higher tumor mass

*in vivo*

Anti-EMT, reduced migration and

Downregulated in pancreatic

[97]

cancer, correlated with progressive

stage, lymphatic invasion, vascular

infiltration, distant metastasis

Upregulated in EC, associated

[100]

with poor differentiation, poor

prognosis, shorter overall survival,

and recurrence; not associated with

cancer stage, localization, tumor

size

invasion, promotes MET *in vitro*, lower

tumor mass, fewer metastases, increased

expression of E-cadherin vs. vimentin,

N-cadherin *in vivo*

Increased proliferation, and increased

metastasis potential *in vitro* and *in vivo*

miR-652

Pancreatic

Metastasis

Inhibits

Acidic

microenvironmentmiR-652 downregulation-ZEB1 upregulation > EMT

inhibitor

ZEB1

cancer

miR-652

Endometrial

Metastasis

Inhibits

Enhanced b-catenin

expression, Wnt-bcatenin signaling pathway

RORA

promoter

cancer

by binding to the ORF

E-cadherin

synthesis

region

promoter

Prostate cancer

Metastasis promoter

Inhibits TNIP1 and

SOCS1

NF-κB signaling inducing

cells in mouse

EMT

Endometrial cancer

Metastasis inhibitor

Inhibits DNMT1

Anti-EMT, increased E-cadherin over

vimentin, fibronectin,

N-cadherin

HIF1-miR-210-3p- > enhances

Promoted EMT, invasion, and migration

Overexpressed in bone metastatic

[95]

tissues, correlates with high PSA

levels, Gleason grade and bone

metastasis status in prostate cancer

patients

Upregulated in BCSC CD44+/

[96]

CD24-sorted from breast cancer

tissue samples

of Pca cells and bone metastasis of Pca

**Pathology**

**Effect**

**Target**

**Mechanism**

**Effect** *in vitro* **and** *in vivo*

Decreased motility and invasion *in vitro*,

—

decreased metastasis *in vivo*

**Clinical associations**

**Ref.**

[94]



**49**

*MiRNA-Based Therapeutics in Oncology, Realities, and Challenges*

taken into consideration as a future miRNA therapy target.

EMT *in vitro* and *in vivo.* In this case, the predicted targets of miR-409 were the tumor-suppressor proteins STAG2 and RSU1, which appeared downregulated [108]. Considering these, it would be worth exploring if miR-409 has a positive or a negative impact over EMT and metastasis in more human cancers, and if it could be

**Table 2** comprises a set of microRNAs that were evaluated both *in vitro* and *in vivo*, while also being analyzed in tumor samples, in correlation with the clinical outcome of the patients. As it can be easily observed, the effect of microRNAs in modulating EMT and metastasis-related pathways varies in different cancer types. In certain situations, the same microRNA can induce completely opposite outcomes by targeting multiple signaling pathways. This would be one of the biggest chal

lenges that need to be overcome when designing new microRNA-based therapeutic compounds, and it is, at the same time, an interesting research niche worth explor

ing, especially for miR-625, that maintained its anti-EMT function in three different cancer types (hepatocellular carcinoma, gastric cancer, and colorectal cancer) and miR-409 that is able to target multiple pathways with opposing effects in NSCLC,

miRNAs represent key modulators of the human genome because of their capacity to affect up to 60% of protein-coding genes. In cancer, genetic and epi

genetic events lead to the alteration of miRNA expression and consequently their mRNA target genes. Functional studies have demonstrated that miRNA modula

tion in tumor cells causes changes in the phenotype, leading to increased apoptosis and cell death, suppression of tumor development, invasion, and metastasis by inhibiting the oncogenic miRNAs (oncomiRs) and/or substituting the deficient tumor suppressive miRNAs (TS-miRNAs). Considering the encouraging preclini

cal and clinical data, miRNA-based therapy could become a reliable tool in cancer

The work for this chapter was supported by the UEFISCDI






*DOI: http://dx.doi.org/10.5772/intechopen.81847*

breast cancer, and prostate cancer.

**7. Conclusions**

management.

**Acknowledgements**

**Conflict of interest**

Program-PN-III-P2-2.1-PED-2016-1750.

The authors declare no conflict of interest.

*MiRNA-Based Therapeutics in Oncology, Realities, and Challenges DOI: http://dx.doi.org/10.5772/intechopen.81847*

EMT *in vitro* and *in vivo.* In this case, the predicted targets of miR-409 were the tumor-suppressor proteins STAG2 and RSU1, which appeared downregulated [108]. Considering these, it would be worth exploring if miR-409 has a positive or a negative impact over EMT and metastasis in more human cancers, and if it could be taken into consideration as a future miRNA therapy target.

**Table 2** comprises a set of microRNAs that were evaluated both *in vitro* and *in vivo*, while also being analyzed in tumor samples, in correlation with the clinical outcome of the patients. As it can be easily observed, the effect of microRNAs in modulating EMT and metastasis-related pathways varies in different cancer types. In certain situations, the same microRNA can induce completely opposite outcomes by targeting multiple signaling pathways. This would be one of the biggest challenges that need to be overcome when designing new microRNA-based therapeutic compounds, and it is, at the same time, an interesting research niche worth exploring, especially for miR-625, that maintained its anti-EMT function in three different cancer types (hepatocellular carcinoma, gastric cancer, and colorectal cancer) and miR-409 that is able to target multiple pathways with opposing effects in NSCLC, breast cancer, and prostate cancer.
