Articulos Cientificos Ginecologia Obstetricia

Clinical Science (2015) 128, 153180 (Printed in Great Britain) doi: 10.1042/CS20140087

Role of microRNAs in cancers of the femalereproductive tract: insights from recent clinicaland experimental discovery studiesMonica Logan and Shannon M. Hawkins

Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, TX, U.S.A.

AbstractmicroRNAs (miRNAs) are small RNA molecules that represent the top of the pyramid of many tumorigenesiscascade pathways as they have the ability to affect multiple, intricate, and still undiscovered downstream targets.Understanding how miRNA molecules serve as master regulators in these important networks involved in cancerinitiation and progression open up significant innovative areas for therapy and diagnosis that have been sadlylacking for deadly female reproductive tract cancers. This review will highlight the recent advances in the field ofmiRNAs in epithelial ovarian cancer, endometrioid endometrial cancer and squamous-cell cervical carcinomafocusing on studies associated with actual clinical information in humans. Importantly, recent miRNA profilingstudies have included well-characterized clinical specimens of female reproductive tract cancers, allowing forstudies correlating miRNA expression with clinical outcomes. This review will summarize the current thoughts on therole of miRNA processing in unique miRNA species present in these cancers. In addition, this review will focus oncurrent data regarding miRNA molecules as unique biomarkers associated with clinically significant outcomes suchas overall survival and chemotherapy resistance. We will also discuss why specific miRNA molecules are notrecapitulated across multiple studies of the same cancer type. Although the mechanistic contributions of miRNAmolecules to these clinical phenomena have been confirmed using in vitro and pre-clinical mouse model systems,these studies are truly only the beginning of our understanding of the roles miRNAs play in cancers of the femalereproductive tract. This review will also highlight useful areas for future research regarding miRNAs as therapeutictargets in cancers of the female reproductive tract.

Key words: endometrial cancer, epithelial ovarian cancer, human discovery studies, microRNA (miRNA), squamous-cell cervical carcinoma

INTRODUCTION

According to the World Health Organization (WHO) Interna-tional Agency for Research on Cancer (GLOBOCAN 2012),over 1 million women worldwide were diagnosed with femalereproductive tract cancers including ovarian, cervical and endo-metrial cancers in 2012 [1,2]. Squamous-cell cervical carcinoma(SCCC) accounts for over half of these diagnoses, whereasovarian (20 %) and endometrial (30 %) cancers account forthe remaining half together. More than half of these 1 million wo-men will die from their disease, with SCCC accounting for a dis-

Abbreviations: bp, base pair; CCND1, cyclin D1; CIN, cervical intraepithelial neoplasia; EEC, endometrioid endometrial cancer; EMT, epithelial to mesenchymal transition; EOC,epithelial ovarian cancer; HPV, human papilloma virus; miRNA, microRNA; NGS, next-generation sequencing; PI3K, phosphoinositide 3-kinase; RISC, RNA-induced silencing complex;SCCC, squamous-cell cervical carcinoma; SNP, single nucleotide polymorphism; STAT3,signal transducer and activator of transcription 3; STIC, serous tubal intraepithelial carcinoma;TCGA, The Cancer Genome Atlas; WHO, World Health Organization.

Correspondence: Dr Shannon M. Hawkins (email [email protected]).

proportionate number of deaths among female reproductive tractcancers worldwide. This SCCC cancer death burden is seen pre-dominantly in less-developed nations with limited resources forscreening and disease prevention. Furthermore, SCCC deaths aretwice as prevalent in younger women (65 years) whereas ovarian and endometrial cancer deathsare nearly equally distributed in younger (65 years). In developed nations, ovarian cancer is themost deadly female reproductive tract cancer due to its lack ofscreening modalities, poor understanding of initiating lesionsand frequent metastatic disease at time of diagnosis [1,2]. These

C The Authors Journal compilation C 2015 Biochemical Society 153

Clin

ical

Sci

ence

w

ww

.clin

sci.o

rg

M. Logan and S. M. Hawkins

health disparities between less-developed and developed nationshighlight the need for research into the discovery of inexpensiveearly diagnostic tools, biomarkers, prognostic factors and initiat-ing lesions to understand the pathogenesis of these diseases andimprove therapy for these deadly cancers in both developed andless-developed nations.

An area of translational interest on these fronts has beenmicroRNAs (miRNAs). miRNAs are single-stranded RNA mo-lecules that are 1824 nucleotides in length that potentially regu-late large gene networks [35]. Over the last 20 years since thesesmall RNA molecules were originally described [6], the numberof unique miRNAs has exponentially increased with the latestversion of miRBase (www.miRBase.org), the online resourcededicated to the miRNA world [7], describing the sequence ofover 30 000 mature miRNA species. To date (miRBase v20), 2578of these are expressed in human [7]. miRNAs represent the topof the pyramid of many tumorigenesis cascade pathways as theyhave the ability to affect multiple, intricate and yet undiscovereddownstream targets [8]. Understanding how miRNA moleculesserve as master regulators in these important networks involvedin cancer initiation and progression open up significant innovat-ive areas for therapy and diagnosis that have been sadly lackingfor deadly female reproductive tract cancers. Although still inthe discovery phase, miRNA molecules as diagnostic markersor therapeutic applications have potential for significant clinicalimpact in the field of female reproductive tract cancers.

Multiple studies have profiled miRNAs in ovarian, cervical,and endometrial cancer tissues and cell lines using either microar-ray technology or next-generation sequencing (NGS). NGS of-fers depth of sequencing giving novel miRNA molecules, uniqueisomiRs, and both miR-5p and miR-3p forms [915], whereasmicroarrays are good for characterizing expression of knownmiRNAs usually with less expense. Importantly, many investig-ators have studied individual miRNAs in depth to push those mo-lecules forward as biomarkers or therapeutic modulators basedon in vitro studies or clinical association studies. Additionally,bioinformatics approaches integrating multiple platforms anddatasets have allowed investigators to ask clinically importantquestions and get biologically plausible answers. This reviewwill highlight the recent advances in the field of miRNAs inepithelial ovarian cancer (EOC), SCCC and endometrioid endo-metrial cancer (EEC), focusing on studies associated with clinicalinformation in humans since our last reviews [3,16]. This reviewwill summarize the current thoughts on the role of miRNA pro-cessing in unique miRNA species present in these cancers. Thisreview will then focus on current data regarding miRNA mo-lecules in early diagnosis, initiating lesions, pathogenesis, pro-gnostic factors and therapy in each of these cancers. Additionally,potential reasons for why specific miRNA molecules are not rep-licated across specific cancer types will be included. Although themechanistic contributions of miRNA molecules to these clinicalphenomena have been confirmed using in vitro and pre-clinicalmouse model systems, these studies are truly only the beginningof our understanding of the role miRNAs play in cancers of thefemale reproductive tract. This review will also contain hopefulareas for future research regarding miRNAs as therapeutic targetsin cancers of the female reproductive tract.

OVERVIEW OF miRNA PROCESSING AND ITSIMPORTANCE IN FEMALE REPRODUCTIVETRACT CANCERS

The role of DICER processing (Figure 1) in regulation of miRNAexpression is only beginning to be understood within the com-plexities of novel miRNAs, isomiRs and expression of both com-plementary miRNA forms (i.e., miR-3p forms). NGS has allowedfor discovery of uniquely processed miRNAs, called isomiRs,which differ from mature miRNA sequence being 12 nucle-otides longer [915]. Differential excision by DICER may leadto these variations in the 5 and 3 ends of mature miRNAs,altering affinity to Argonaute proteins, and changing target spe-cificity [1720]. Moreover, additions of adenosine residues maybe catalysed by nucleotidyltransferases [14,18,21]. The uniquefunctions of these isomiRs such as adenylated forms that maybe degraded faster [22,23] or have alternate functions are not yetunderstood in female reproductive tract cancers. From a trans-lational standpoint, the biological and clinical association of al-ternately processed miRNAs such as isomiRs and less abundantmature miR-3p forms needs to be fully characterized in well-characterized clinical specimens of female reproductive cancersusing NGS to identify miRNAs that are associated with thesedeadly cancers.

From a basic mechanistic standpoint, the role of DICER as acritical miRNA-processing enzyme in the formation and stabil-ity of these isomiRs and miR-3p forms is not fully understood.Mutations in one allele of DICER were discovered in 30 % ofnon-EOCs [24]. Further studies showed that this haploinsuffi-cient mutation in one of the two RNAse III domains led to sig-nificant changes in the processing of precursor miRNAs, givingaltered processing of miR-5p and miR-3p forms (Figure 1B).In vitro experiments with DICER haploinsufficiency showed lossof processing of the more abundant mature miR-5p forms butnormal processing of the less abundant miR-3p form [25]. Otherin vitro experiments showed that mutations in the RNAse IIIAdomain disrupted miR-3p processing completely and decreasedmiR-5p processing whereas mutations in RNAse IIIB domaindisrupted miR-5p processing of let-7 [26]. In additional studies,mutations in the two Mg2+ binding sites in the RNAse IIIB do-main ofDICER have recently been reported in endometrial cancer[27]. Mg2+ ions help DICER bind and cleave RNA, which is es-sential for miRNA biogenesis [28]. Typically, the miR-5p andmiR-3p forms of the same mature miRNA have unique seed se-quences, belong to different miRNA families and target differentgenes. Given that mature miR-5p forms are the most abundantand the most widely studied, the switch to mature miR-3p forms(Figure 1B) being more abundant may have drastic downstreameffects on a large number of pathways important in cancer. How-ever, the relative expression of the miR-5p and miR-3p speciesis still unknown in well-characterized clinical specimens. Futurestudies will be important to determine the expression patterns ofthese miRNA forms as they may have significant clinical impactinto disease prognosis.

From a mouse model standpoint, genetically engineeredmouse models support a role of Dicer haploinsufficiency in non-female reproductive tract cancers [2932]. However, complete

154 C The Authors Journal compilation C 2015 Biochemical Society

microRNAs in female reproductive tract cancer

Figure 1 Schema for miRNA genesis(A) Primary miRNA (pri-miRNA) molecules are transcribed from DNAloci by RNA polymerase II. The pri-miRNA transcript, roughly 200 bp inlength, has a characteristic stem-loop structure. The pri-miRNA tran-script is further processed by DROSHA, an RNAse, removing the armsand leaving the looped structure, roughly 70 bp in length precursormiRNA (pre-miRNA). Exportin 5 transports the pre-miRNA to the cyto-plasm where DICER, another RNAse, processes it into two reversecomplementary single-stranded mature miRNA molecules, each roughly22 bp in length. These mature miRNA molecules are incorporated intoand Argonaut protein in the RNA induced silencing complex (RISC)[196,197]. The seed sequence, nucleotides 28 of the mature miRNA,determines the complementary targets in the 3UTR of target genes.Within the RISC, the miRNA binds to complementary target sites in the3UTR mRNA, leading to mRNA degradation if perfect complementarybinding or translational repression if imperfect complementary binding[3,198200]. In general, the mRNA degradation and subsequent de-crease in protein levels are the main post-transcriptional functions ofmiRNAs. Thus, an overexpressed miRNA will lead to decreased expres-sion of its target genes. Loss of miRNA will lead to overexpression of itstarget genes [198200]. Furthermore, each miRNA can target multiplegenes. Thus, miRNAs regulate large networks of genes in a coordinatedfashion, serving as master regulators and, as such, some have beentermed oncomiRs or tumour suppressors [201]. (B) In the cytoplasm,DICER processes pre-miRNA molecules to two molecules, the miR-5pstrand and the miR-3p strand (also known as the form). Each of thesemolecules can be loaded independently into RISC. Although miRNAswith the same seed sequence have similar targets and are groupedinto miRNA families [198200], miR-3p and miR-5p molecules may nottarget the same 3UTRs as they have different seed sequence. Themore abundant form is typically the miR-5p form that is likely the mostactive. The miR-3p form (also known as the miRNA form) is typicallyless abundant due to degradation with little known activity. Currently,the function of many miR-3p forms on gene expression is not yet known

deletion of Dicer alone in female reproductive tract of mice doesnot lead to cancer [3,16,3336], and cancer studies in female re-productive tract withDicer haploinsufficiency have not been pub-lished. In both endometrial cancer and EOC, low DICER expres-sion is associated with poor prognosis [3739]. However, the spe-cific mechanism for reduced DICER transcript in these cancers isnot clear. Several miRNAs reported to be up-regulated in endo-metrial cancer have been predicted to target DICER by in silicoalgorithms and include miR-103, miR-107, miR-200a, miR-141,miR-9 and let-7c [38,40]. However, none of these miRNAs have

been confirmed to directly regulate DICER expression in vitro.On the other hand, miR-130b has been shown to directly targetDICER in human endometrial cancer cells, which in turn leadsto abnormal expression of epithelial to mesenchymal transition(EMT)-related genes [41]. Critical studies are needed to determ-ine if DICER haploinsufficiency or loss of DICER expressionis related to miRNA processing (i.e., miR-5p to miR-3p switch-ing or isomiR production) or other mechanisms in these femalereproductive tract cancers.

CLINICAL ASSOCIATIONS OF miRNAMOLECULES IN EPITHELIAL OVARIANCANCER

Brief overview of miRNAs in epithelial ovariancancerUsing 2012 data, 238,719 women worldwide were diagnosedwith ovarian cancer and over 151,905 women died from thisdisease [42], mostly from disease recurrence or progression ontherapy. Although 70 % of women with advanced-stage EOC ini-tially respond to surgical debulking and standard chemotherapy,the 5-year survival rate is still poor, probably due to a subset ofcells being resistant to chemotherapy [43]. These data are notsurprising considering that the survival rates for EOC have notimproved significantly in 40 years due to lack of early diagnosticmarkers [44]. In stark contrast with breast cancer, which usespersonalized therapy based on molecular markers (i.e., ER, PR,Her2-neu) [4548], EOC treatment does not yet use molecularmarkers to personalize therapy. Just as oestrogen affects multiplegene networks in breast tissue [49], individual miRNAs affectlarge network of genes in ovarian cancer [16,5052]. Multiplestudies have profiled miRNAs in EOC compared with normaltissues or cell lines to generate lists of differentially expressedmiRNAs [12,5260]. Many of these miRNA molecules from pro-filing studies have in vitro or in vivo xenograft data to supportthe biological plausibility of the miRNA action (Table 1) [3,16].Many in vitro studies in EOC have focused on the role of miRNAsin chemotherapy sensitivity, tumour growth or invasion that areall clinically important cellular properties. Thus, miRNAs havethe potential to play significant roles in early diagnosis, initiatinglesion formation, pathogenesis, prognosis and therapy.

Role of miRNAs in predisposition and earlydiagnosis of epithelial ovarian cancerFor EOC, two genome-wide association studies suggest thatsingle nucleotide polymorphisms (SNPs) in miRNAs or 3UTRare uncommon in EOC [61,62]. On the other hand, a large cohortstudy using SNPs in miRNAs suggests that 17q21.31 contains asusceptibility locus for EOC, although the candidate genes at thislocus have not yet been studied for biological plausibility [63].A candidate gene approach examined a polymorphism in a let-7binding site in the 3UTR of KRAS, finding no association withEOC in one population but association with poor outcome in an-other population [64,65] (Table 2). Given the multiple genomicchanges noted in EOC by The Cancer Genome Atlas (TCGA)



Table 1 In vitro studies of miRNA molecules in ovarian cancer

miRNA Effect in vitro Genes, pathways regulated Reference(s)

let-7, let-7e, let-7g Growth (xenograft) EZH2, CCND1 [202204]

Chemotherapy sensitivity IMP1, MDR1

miR-100 Proliferation PLK1 [205]

miR-101 Growth (xenograft) EZH2 [206]

Apoptosis

Migration

miR-106a Chemotherapy sensitivity PDCD4, MCL1, BCL10, Caspase-7, RBL2 [207210]

Apoptosis

Proliferation

Putative stem cell population

Growth (xenograft)

miR-124, miR-124a Migration, invasion SPHK1 [211,212]

Proliferation

Cell cycle

miR-128 Motility and adhesion CSF1 (affects large networks of genesassociated with cell cycle control)

[213,214]

miR-130a,b Chemotherapy sensitivity XIAP, CSF1, MDR1 [215217]

Apoptosis

miR-138 Invasion (xenograft) SOX4, HIF1 [218]

miR-141 Chemotherapy sensitivity KEAP1, p38 [219,220]

Oxidative stress response

Growth (xenograft)

miR-145 Growth (xenograft) P70S6K1 [221,222]

Invasion MUC1

Angiogenesis

miR-148a Proliferation [223]

miR-152 Chemotherapy sensitivity DNMT1 [213,223,224]

Apoptosis CSF1

Metastasis (xenograft)

Adhesion and motility

Proliferation

miR-155 Ovarian cancer initiating cells (CD44+CD117+) CLDN1 [225]

Proliferation

Invasion (xenograft)

miR-182 Chemotherapy sensitivity PDCD4 [226]

Growth and invasion

miR-182 + miR-96 Leptin-mediated proliferation FOXO3, STAT5 [227]miR-185 Chemotherapy sensitivity DNMT1 [224]

Apoptosis

miR-187 Migration DAB2 [228]

EMT

miR-192 Proliferation [212]

Cell cycle

miR-193b, miR-193a Chemotherapy sensitivity ARHGAP19, CCND1, ERBB4, KRAS,MCL1

[212,229]

Proliferation

Cell cycle

Apoptosis

miR-199a Chemotherapy sensitivity (xenograft) mTOR, CD44+ [230,231]

Apoptosis

Cancer stem cell



Table 1 Continued


miR-199a + miR-125b Angiogenesis VEGF [232234]Regulated by reactive oxygen species Akt/p70S6K1/HIF1

Chemotherapy sensitivity HER2, HER3

Apoptosis ERBB2, ERBB3, Bcl-2

miR-200a Chemotherapy sensitivity p38 [220]

Oxidative stress response

Growth (xenograft)

miR-200c Putative stem cell population PTEN [235237]

EMT TrkB


Proliferation

Anoikus sensitivity

Chemotherapy sensitivity (xenograft)

miR-20a Putative stem cell population PTEN [235]

miR-21 Proliferation, apoptosis PCDC4 [238]

Migration, Invasion

miR-214 Ovarian cancer stem cells P53, NANOG [239]

miR-222 Proliferation p27KIP1 [240]

miR-25 Apoptosis BIM [241]

miR-27a/b Proliferation/DGCR8 Sprout2 [242,243]

Genistein sensitivity

miR-29 Chemotherapy sensitivity COL1A1, ERK1/2, GSK3B [244]

Apoptosis

miR-30-5p Chemotherapy sensitivity [245]

miR-302b Chemotherapy sensitivity (xenograft) HDAC4 [246]

miR-30c-2 Proliferation BCL9 [247]

miR-30d Proliferation CASP3 [248]

Apoptosis

Cellular senescence

Growth (xenograft)

miR-31 Chemotherapy sensitivity MET [249]

miR-335 Migration, invasion BCL2L2 [250]

miR-367 Chemotherapy sensitivity [245]



miR-375 Chemotherapy modulation to RAWQ01 (xenograft) [251]

miR-520d-3p Proliferation EPHA2 [183]

Migration, invasion


miR-591 Chemotherapy sensitivity ZEB1 [209]

Apoptosis

miR-7 Proliferation Affects large networks of genesassociated with EMT and proliferation

[212,214]

miR-9 Proliferation TLN1, RAB34 [83,252]

Migration, invasion

miR-92a Peritoneal dissemination (xenograft) ITGA5 [253]

miR-93 Chemotherapy sensitivity PTEN [254]

Apoptosis

miR-498 Growth (xenograft) Telomerase [255]



Table 2 Clinically associated miRNA molecules in ovarian cancer

miRNA Clinical indication Reference(s)

let-7 Genomic deletion in 44 % of EOC [64,65,202]

KRAS let-7 binding site (LCS6):

Not associated with EOC outcomes

Associated with EOC outcomes, chemotherapy sensitivity

let-7a Chemotherapy sensitivity to paclitaxel [256]

let-7b Master regulator of networks of proteins associated with poor survival [257]

let-7f (plasma) Low expression correlates with poor progression free survival [68]

miR-100 Associated with overall survival [205]

miR-138 Low miR-138/high SOX4 associated with poor prognostic features [218]

miR-148 Expression not correlated with clinical features [223,258]

miR-181d High expression associated with decrease disease free interval [88]

miR-187 High expression associated with improved overall survival [228]

miR-200a Biphasic expression with lowest expression associated with high grade and stage [220,259]

High miR-200a with low p38 associated with stress response and better prognosis

miR-200b + miR-200c High expression in high-grade serous EOC [260]miR-200c Low expression associated with relapse in stage I EOC [261]

Low expression associated with poor overall survival and progression free survival instage I EOC

miR-205 (plasma), let-7f (plasma) High specificity as diagnostic marker [68]

miR-21, miR-214 High expression in ascites compared with metastatic lesions [262]

miR-221 (serum) Biomarker [263]

High expression correlates with poor prognosis

miR-27a High expression in cancer stem cells (ALD+) [264]

Associated with metastatic disease

miR-30a Low expression associated with poor survival in elderly EOC [85]

miR-30c High expression associated with decreased disease-free interval [88]

miR-30d High expression associated with decreased overall survival [88,248]

High expression associated with decreased disease-free interval

Amplified in 50 % of EOC

miR-30e Low expression associated with poor survival in elderly EOC [85,88]

High expression associated with decrease disease free interval

miR-31 Low expression correlates with poor survival [249]

miR-484 Associated with tumour vessel density (IHC) [84]

miR-502 Mutation in miR-503 binding site in SET8 correlates with EOC [265]

miR-503 High expression in cancer stem cells (ALD+) [264]

Associated with clinical stage

miR-509-5p Low expression associated with poor overall survival [266]

miR-510 Low expression associated with poor overall survival [266]

miR-520d-3p High expression associated with improved overall survival [183]

miR-9 High expression associated with overall survival [83]

miR-92 Biomarker [267]

miR-505 Low expression associated with poor survival in elderly EOC [85]

data and others [66,67], small polymorphisms may not play asignificant role in advanced stage disease but may play a role inpredisposition of normal tissues to malignant transformation.

Unfortunately, most EOC are diagnosed after significant dis-ease spread, as there are no highly specific or sensitive methodsof early diagnosis [44]. Several studies have focused on circu-lating miRNAs in blood as a means of non-invasive diagnosisand early detection in EOC (Table 3). One study has discovered

miRNAs that are potential diagnostic markers in the blood suchas miR-205/let-7f [68], although the study did not have con-crete data on the function of these miRNAs in EOC. Studiesin prostate cancer showed that miR-205 was critical for activa-tion of the RAS oncogene [69], which is a potential player inEOC [70]. The present study was well-designed having trainingand independent validation sets for calculation of sensitivity andspecificity [68]. Recently, the miRNA profiles of pre-operative



Table 3 miRNAs in blood for use as diagnostic markers of female reproductive tract cancersNC, not calculated.

miRNACancer compared with comparison group(sample studied) ROC Sensitivity Specificity Reference(s)

let-7f Ovarian cancer compared with controls(plasma)

0.780 66.9% 84.2% [68]

miR-205 Ovarian cancer compared with controls(plasma)

0.609 30.1% 94.2% [68]

let-7f + miR-205 Independent validation set of ovarian cancercompared with controls (plasma)

0.813 71.3% 81.2% [68]

miR-106b, miR-126, miR-158,miR-17, miR-20a, miR-92a

Ovarian cancer compared with benign masscompared with control (plasma)

NC NC NC [71]

miR-1290 Ovarian cancer compared with benign masscompared with control (plasma)

NC NC NC [71]

miR-21, miR-141, miR-200a,miR-200c, miR-200b, miR-203,miR-205, miR-214

Ovarian cancer compared with benigndisease (circulating exosomes)

NC NC NC [74]

miR-16, miR-191, miR-21 Endometrioid or clear cell ovarian cancercompared with healthy controls (plasma)

0.96 86% 85% [75]

miR-16, miR-191, miR-4284 Serous ovarian cancer compared with healthycontrols (plasma)

0.90 90% 55% [75]

miR-342-3p Relapsed ovarian cancer compared withhealthy controls (whole blood)

0.86 NC NC [72]

miR-221 Ovarian cancer compared with healthycontrols (serum)

NC NC NC [263]

miR-92 Ovarian cancer compared with healthycontrols (serum)

0.803 80.7% 75% [267]

miR-203 Stage IIIA SCCC lymph node metastasis+compared with lymph node metastasis-(serum)

0.658 65% 62.5% [128]

miR-20a Stage IIIA SCCC lymph node metastasis+compared with lymph node metastasis-(serum)

0.734 75% 72.5% [128]

miR-1246, miR-20a, miR-2392,miR-3147, miR-3162-5p,miR-4484

Stage IIIA SCCC lymph node metastasis+compared with lymph node metastasis-(serum)

0.992 85.6% 85% [129]

miR-218 SCCC compared with healthy controls(germline DNA)

NC NC NC [111,268]

miR-27a SCCC compared with healthy controls(germline DNA)

NC NC NC [269]

miR-222, miR-223, miR-186,miR-204

Endometrial cancer to healthy controls(serum)

0.927 87.5% 91.7% [157]

miR-9, miR-1228 Endometrial cancer to healthy controls(plasma)

0.909 73% 100% [138]

miR-9, miR92a Endometrial cancer to healthy controls(plasma)

0.913 79% 100% [138]

miR-200b, miR-200c, miR-203,miR-449a

Endometrial cancer with myometrial invasion>0.5 compared with myometrial invasion


Profiling from circulating exosomes showed significant correla-tion in miRNA expression between tumour and circulating exo-somes. However, a signature of eight exosomal-miRNAs was notassociated with stage of disease [74]. Finally, a critical studyexamining miRNAs found in blood compared control women towomen with endometriosis to women with EOC. They found thatthe number of miRNAs detected in the blood increased from con-trol to endometriosis to cancer. Importantly, a minimal panel ofmiRNAs had reasonable sensitivity and specificity between con-trol, endometriosis, clear-cell, endometrioid and serous EOC.Excitingly, these same miRNAs were tested in a mouse model ofendometriosis-associated EOC and were found to closely replic-ate the human studies [75]. The use of miRNA signatures fromblood as a means of non-invasive diagnosis seems promising butstandardization regarding RNA isolation (i.e., whole blood, cellu-lar components, plasma or exosomes) is needed. Additionally, themechanism by which tumours release miRNAs (i.e., exosomes,secretion, cell lysis or tumour macrophages) still needs to be de-ciphered. Prudent studies would use pre-clinical mouse models ofovarian cancer for optimization of miRNA isolation from bloodand determination of minimal signature of miRNA for diagnosticpurposes.

The role of miRNAs in initiating lesions in epithelialovarian cancerThe initiating lesion is hotly debated as to whether high-gradeserous EOC, the most common histotype of EOC, begins in thefimbria of the fallopian tube or the ovarian surface epithelium[76,77]. Serous tubal intraepithelial carcinoma (STIC) lesions infimbria may be the precursor lesions for high-grade serous EOC.Studies in humans have shown increased expression of miR-182in STIC lesions over normal fallopian tube. Forced overexpres-sion of miR-182 in vitro lead to increased cellular transformation,likely through dysregulation of DNA repair pathways involvingBRCA1, MTSS1 and HMGA2 [78]. Additionally studies sup-port the role of overexpression of miR-182a in suppression ofFOXO3A in fallopian tube cells leading to high-grade serousEOC [79]. Recently, a genetically engineered mouse model, con-taining conditional deletions of Dicer and Pten in the mesen-chymal cells of the female reproductive tract (Amhr2cre; Ptenf/f;Dicerf/f), showed that high-grade serous EOC originates in the fal-lopian tube [80]. A different genetically engineered mouse model,containing an oncogenic Kras with Pten deletion in the mesen-chymal cells of the female reproductive tract (Amhr2cre; Ptenf/f;KrasG12D/+), suggested that low-grade serous EOC originates inthe ovarian surface epithelium with dysregulation of p53 andmiR-34c [81]. These few studies highlight the use of miRNAs as futurebiomarkers for the origins of EOC. Further studies are needed todecipher the role of miRNAs in cancer initiation in female re-productive tract cancers including the endometriosis-associatedhistotypes of EOC, specifically clear-cell and endometrioid EOC.Again, the use of pre-clinical mouse models with well-definedgenetics offers a unique advantage if validated by human tissueprofiles that are available.

miRNAs as prognostic markers in epithelial ovariancancerRecent profiling datasets have focused on well-defined clinicalcharacteristics to determine specific miRNAs or miRNA signa-tures associated with important clinical features such as EOChistotype, lesion location, prognosis or chemotherapy resistance(Table 2) [8288]. Looking at miRNA profiles, a recent studyshowed the histotypes of stage I ovarian tumours are molecularlydifferent [82]. Importantly, miR-30a and miR-30a were markersof clear-cell tumours whereas miR-192 and miR-194 were mark-ers of mucinous tumours. Each of these miRNAs had functionaltargets leading to unique networks affected in clear-cell or mucin-ous tumours at the mRNA level [82]. Previous studies have shownunique gene signatures for EOC histotypes [8992]. Therefore,miRNA play critical roles in distinguishing EOC histotypes andunderlying biological networks. Thus, miRNA play a significantrole in determining which molecular markers in tumours willrespond to specific therapies in the future.

Although individual miRNA molecules may be important, aminimal signature of miRNAs in the tumours may increase sens-itivity and specificity of prognosis or treatment effects. Focusingon miRNA profiles from metastatic lesions, one study recentlyprofiled miRNAs from primary tumours compared with omentalmetastatic lesions. The authors found thatmiR-146a andmiR-150were highly expressed in omental lesions with significant chemo-therapy sensitization effects of these 2 miRNAs in vitro [86].Thus, miRNAs may allow better differentiation by location oflesion that affects clinically important factors such as chemother-apy resistance. For prognostic studies, analysis of miRNA data-sets from early relapsing compared with late relapsing tumoursrevealed low expression of miRNAs on chromosome Xq27.3.In vitro expression of these miRNAs was associated with chemo-therapy sensitivity [93]. Similarly, miRNA profiling studies fo-cusing on serous EOC with chemotherapy resistance showed aminimal signature of miR-484, miR-642 and miR-217 associatedwith chemotherapy resistance. This resistance was attributed toregulation of angiogenic factors by miR-484 in human ovariantissue samples using vessel density and xenograft models. Thislandmark study may have identified three miRNAs that may de-termine which patients will benefit from anti-angiogenesis ther-apy in combination with standard therapy [84]. In support of this,a large-scale bioinformatics study showed an anti-angiogenesismiRNA signature that correlated with overall survival [94]. Ad-ditional translational studies like these may allow stratificationof therapy, based on need for angiogenesis modulation, to de-termine which subjects would benefit from expensive and stillexperimental anti-angiogenesis therapy.

CLINICAL ASSOCIATIONS OF miRNAs INSQUAMOUS-CELL CERVICAL CARCINOMA

Brief overview of miRNAs in squamous-cell cervicalcarcinomaAccording to GLOBOCAN 2012, worldwide 527 624 womenwere diagnosed with cancer of the uterine cervix and over 265 653



died from this disease [1,2]. Worldwide, uterine cervix cancer isthe fourth most common cancer in women. Sadly, nine out of tendeaths from cervical cancer occur in less-developed countries.Overall, less-developed areas have five times the incidence ofcervical cancer and eight times the mortality [1,2]. In more de-veloped countries, screening for early cytological abnormalitieswith Pap testing and human papilloma virus (HPV) co-testing af-fords early identification and treatment of pre-cancerous lesions,cervical intraepithelial neoplasia (CIN) [9597]. Additionally,early age immunization for HPV high-risk subtypes should beprotective [98]. However, these screening tests lead to expensivediagnostic procedures such as colposcopy that are frequently notavailable in resource poor nations [99]. Therefore, additional in-expensive means of early triage for abnormal cervical cytologyare needed to determine which women need invasive proced-ures and which do not. miRNAs may play a role in the earlyinitiating lesion of SCCC, thus making them an excellent av-enue for research. Multiple studies have profiled miRNAs incervical cancers, CIN lesions and cell lines [100106]. Manyof the miRNA molecules from profiling studies have in vitro orin vivo xenograft data to support the biological plausibility ofmiRNA function (Table 4) [3]. Thus, miRNAs may play signi-ficant roles in predisposition, initiating lesions and prognosis ofSCCC.

Role of miRNAs in predisposition to squamous-cellcervical carcinomaAlthough the cause of SCCC is derived from HPV, geneticpre-disposition studies have attempted to find a genetic linkas to which women clear HPV and other women have pro-gressive disease, looking for SNPs in miRNA target genes ormiRNAs [107]. A few SNPs have been found to be protect-ive in cervical cancer (Table 5), although the underlying bio-logical plausibility for protection has not yet been clearly de-ciphered. One of the most widely studied SNP is rs2 910 164 inmiR-146a. Several meta-analysis have reviewed these datasetsand found that this SNP is associated with decreased risk ofSCCC but mostly in Asian population [108], but other meta-analyses found no protection [109,110]. Interestingly, a variantin the pri-miR-218 was found to decrease risk of SCCC in East-ern Chinese women, possibly through alteration of the secondarystructure of the primary miRNA molecule [111]. For SCCC,studies suggest that copy number changes affect DROSHA, akey enzyme for miRNA genesis [102,112,113]. However, SNPsin miRNA genesis genes are not associated with cervical can-cer [114]. Thus, genetic predisposition to infection from HPVfrom a miRNA perspective does not seem to have a high clinicalrelevance.

The role of miRNAs in initiating lesions tosquamous-cell cervical carcinomaThe role of the persistent precancerous lesion CIN is apparentprior to SCCC [115117]. Predicting which CIN lesions regresscompared with those lesions which progress to invasive cancer isan important clinical question as treatment of CIN lesions can res-ult in significant morbidity from excisional procedures (i.e., cold

knife cone, loop electrosurgical excision procedure) that may af-fect future reproductive outcomes such as incompetent cervix andpreterm labour [118]. Studies have shown that miRNAs are en-coded by HPV genes and loss of miRNA binding sites within theHPV genome is relatively common although the clinical signific-ance remains to be determined [119,120]. Profiling of high-gradeCIN lesions found a minimal signature associated with CIN that isapparent in invasive disease. Thus, miRNAs may play a signific-ant role in the initiation of pre-cancerous lesion [121]. Recently,a study found that miRNA pairs could predict which lesionswould progress to SCCC. An expression ratio of miR-25/miR-92a and miR-22/miR-29a could be a useful diagnostic markerfor HPV infection leading to SCCC, possibly through HPV E6and E7 [122]. Importantly, one of the miRNAs that was foundto be regulated by methylation and important for higher gradelesion, miR-124 [123], has been under testing for a quantitativemethylation-specific PCR assay on Pap specimens [124]. Recentwork in the miRNA field has focused on epigenetic changes thataffect miRNA expression in the progression of CIN to invasivedisease using human tissue samples and treated in vitro cultures[125]. For example,miR-203, which affects angiogenesis throughVEGFA, is regulated by methylation of its promoter [126]. Ad-ditionally, miR-214, which affects apoptosis through BCL2L2,is regulated by methylation and acetylation [127]. More work isneeded to determine a minimal miRNA profile in CIN that pre-dicts progression to invasive disease and the epigenetic changesthat regulate those miRNAs.

miRNAs as prognostic markers in squamous-cellcervical cancerRecent miRNA profiling datasets have focused on well-definedclinical characteristics to determine specific miRNAs or miRNAsignatures associated with important clinical features of SCCCsuch as chemotherapy resistance, lymph node metastasis, invas-ive disease and overall survival (Table 5). One of the most difficultclinical features of cervical cancer to determine pre-operativelycan be the presence of lymph node metastasis. Clinical suspicionof distant disease by lymph node metastasis changes manage-ment from radical hysterectomy to leaving the uterus in situ andcombined chemotherapy with radiotherapy. However, determ-ination of lymph node metastasis typically requires biopsy foractual diagnosis. Thus, non-invasive tests to determine extent ofdisease are critical for reducing surgical risk and optimizing ther-apy. Serum miR-20a had a sensitivity and specificity greater than70 % in detecting lymph node metastasis preoperatively [128].More promisingly, a minimal serum miRNA signature contain-ing miR-1246, miR-20a, miR-2392, miR-3147, miR-3162-5p andmiR-4484 has reasonable sensitivity and specificity with similarexpression in tissues [129] (Table 3). In addition, miRNA mayplay a role in chemotherapy sensitivity. A randomized-controlledtrial suggests that neoadjuvant chemotherapy may increase thep53:miR-34c:E2F1 and p53:miR-605:MDM2 pathways in stageIIB SCCC which may be protective and improve outcomes [130].Similar to EOC, none of these miRNAs are used clinically, andfuture studies are needed.



Table 4 In vitro studies of miRNA molecules in cervical cancer


let-7 Cell survival HAS2 [270]

Invasion

miR-100 Proliferation PLK1 [271]

Apoptosis

miR-101 Proliferation Cox-2 [272]

Apoptosis

Migration, invasion

miR-10a Proliferation CHL1 [273]

Migration, invasion

miR-125b Apoptosis BAK1, PIK3CD AKT pathway [274,275]

Proliferation

miR-129-5p Proliferation E6, E7, SP1 [276]

Apoptosis

miR-133b Proliferation MST2, CDC42, RHOA AKT, ERK1 [277]

Growth and metastasis (xenograft)

miR-143 Growth (xenograft) Bcl-2 [278]

miR-145 Cortisol-mediated chemotherapy sensitivity P53 [279]

miR-15-3p Apoptosis BCL2L1 [280]

miR-155 Autophagy activity mTOR pathway: RHEB, RICTOR, RPS6K2 SMAD2, CCND1 [281,282]

EGF-mediated EMT

Chemotherapy sensitivity

miR-17-5p Apoptosis TP53INP1 [283]

miR-181a,b Proliferation AC9 PRKCD [284,285]

Apoptosis

cAMP production


Growth (xenograft)

miR-182 Growth (xenograft) Cell cycle and apoptosis pathways [286]

Apoptosis FOXO1

miR-19a/b Proliferation CUL5 [287]

Invasion

miR-203 Proliferation VEGFA [126,128]

Growth (xenograft)

Angiogenesis

miR-205 Proliferation CYR61 [288]

Migration CTGF

miR-20a Proliferation TNSK2 [128,289]

Migration, invasion

miR-21 Proliferation Migration, invasion CCL20 [290]

miR-214 Proliferation GALNT7 [127,291]

Migration, invasion BCL2L2


Apoptosis

miR-218 Proliferation LAMB3 [292,293]

Chemotherapy sensitivity mTOR/AKT

Apoptosis

Growth (xenograft)

miR-223 Proliferation FOXO1 [294]

miR-29a/c Migration, invasion HSP47, YY1, CDK6 [295,296]



Table 4 Continued


HPV-mediated proliferation and apoptosis

miR-302-367 Proliferation CCND1 [297]

AKT1

miR-361-5p Proliferation E-cadherin [298]

Apoptosis

Migration

EMT

miR-375 Chemotherapy sensitivity SP1 [299301]

Apoptosis

Growth (xenograft)

Migration, invasion

miR-424 Proliferation CHK1 [302]

Apoptosis

Migration, invasion

miR-497 Proliferation IGF1R [303]

Apoptosis

Migration, invasion

miR-590-5p Growth CHL1 [304]

Migration

miR-630 Radiotherapy sensitivity [305]

miR-7 Growth XIAP [306]

Apoptosis

miR-9 Growth [307]

Proliferation

Migration

miR-99 Proliferation TRIB2 [308]

Apoptosis

Table 5 Clinically associated miRNA molecules in cervical cancer


miR-25/miR-92a and miR-22/miR-29a Increased expression ratio associated with SCCC [122]

miR-100 Decreased expression associated with invasive disease over CIN [271]

miR-133b Expression associated with invasive disease over CIN [277]

miR-146a SNP associated with decreased cervical cancer risk [108]

miR-181a Chemotherapy resistance [309]

miR-203 Circulating levels predict lymph node metastasis [128]

miR-20a Circulating levels predict lymph node metastasis, diameter oftumour

[128,310]

miR-21 Expression associated with invasive disease over CIN [311]

miR-218 Serum levels predict stage, lymph node metastasis SNPassociated with decreased cervical cancer risk

[111,268]

miR-224 Microvascular invasion, HPV infection, overall survival, stage,lymph node metastasis

[312]

miR-27a SNP associated with decreased cervical cancer risk [269]

miR-375 Decreased expression associated with lymph node metastasis [300]

miR-424 Stage, lymph node metastasis, tumour grade, microvascularinvasion

[302]

miR-497 Low tissue levels associated with worse stage and lymph nodemetastasis

[303]



CLINICAL USES OF miRNAs INENDOMETRIAL CANCER

Brief overview of miRNAs in endometrial cancerAccording to GLOBOCAN 2012, over 319 000 women were dia-gnosed with endometrial cancer and 76 155 died from this dis-ease [1,2]. Endometrial cancers can be classified into Type Iand Type II tumours. Type I tumours are low grade and earlystage tumours that are routinely cured by surgery, are hormon-ally sensitive, and encompass mainly EECs. Type II tumoursare usually poorly differentiated, frequently recur, are hormon-ally insensitive, and encompass clear-cell, serous and high-gradeEEC [131]. Many groups have performed expression-profilingstudies to identify miRNAs aberrantly expressed in endometrialcancer [40,132143]. Many of these individual miRNAs are sup-ported by in vitro studies (Table 6). For example, both miR-302and miR-503 have been shown to inhibit tumorigenicity of endo-metrial cancer cells by targeting the cell cycle-associated onco-gene, cyclin D1 (CCND1) [144,145]. In contrast, miR-125b actsas an oncogene by directly targeting the tumour suppressor gene,TP53INP1, and promoting proliferation and migration of type IIendometrial cancer cells [146]. Given that miR-125 expression issignificantly higher in type II non-EEC than type I EEC [132],miR-125 may serve as an important prognostic biomarker for dis-ease progression. Other miRNA molecules are associated withclinical outcomes (Table 7) and discussed in further detail below.

The role of miRNAs in initiating lesions inendometrial cancerFor endometrial cancer, clearly, unopposed oestrogen, either fromexogenous hormone consumption (oestrogen-only hormone re-placement therapy), endogenous hormone production (obesity) orprogesterone resistance (polycystic ovarian syndrome), can leadto abnormal epithelial cell proliferation in the uterus. Clinically,for EEC, the classic precursor lesion is complex hyperplasia withatypia from over proliferation of the endometrium [147151].Given the number of oestrogen responsive miRNAs [152], thesesmall molecules probably play a significant role. The second ma-jor mechanism of endometrial cancer involves activation of thephosphoinositide 3-kinase (PI3K) pathway, through the loss ofthe tumour suppressor PTEN. PTEN acts as a phosphatase forAKT. Phospho-AKT is the active form, and removal of phos-phorylation by PTEN decreases AKT downstream effects. Lossof PTEN through loss of protein function or mutations leadsto loss of this repression, and subsequent activation of the PI3K-AKT pathway, and abnormal endometrial proliferation (Figure 2)[153]. Somatic mutations inPTEN have been reported in 3483 %of EEC cases along with a 5083 % frequency of loss of PTENprotein [133,154,155]. Several groups have investigated the re-lationship between dysregulated miRNAs and PTEN expressionin endometrial cancer [132,133,156]. Lee et al. [133] found thatmiR-200c expression was significantly higher in PTEN-negativeendometrial tumours compared with PTEN-positive tumours. Inaddition, miR-200c and miR-183 have both been predicted to tar-get PTEN in silico [132,133], whereas miR-21 has been shown todirectly target PTEN in endometrial cancer cells [156]. Althoughthese studies shed some light into the mechanism for reduced

Figure 2 The molecular mechanisms leading to endometrialcancerThis simplified diagram depicts the critical pathways in endometrialcancer. Arrows indicate stimulation. Block lines indicate inhibition. ERand PR are likely all nuclear effects, but ER effects could be membranebound effects.

PTEN protein in PTEN wild-type endometrial cancers, they lackin vitro and in vivo functional studies that describe the functionalroles of miRNAs in PTEN-deficient endometrial carcinogenesis.

Role of miRNAs in diagnosis of endometrial cancerAlthough endometrial cancer can be diagnosed using an office-based endometrial biopsy, many women do not tolerate this of-fice procedure due to stenotic cervical os, low pain thresholdsand anxiety. Thus, the idea of plasma diagnosis may allow foreasier screening, especially for Type II EEC that do not presentas frequently with postmenopausal or abnormal uterine bleeding[131]. Similar to EOC and SCCC, studies have attempted to useserum or plasma to detect endometrial cancer (Table 3). Usingserum samples, a panel containing miR-222, miR-223, miR-186and miR-204 was found to be a useful serum biomarker withsensitivity and specific over 90 %, but these data were not valid-ated in independent samples [157]. Additionally, a study profiledmiRNAs from both endometrial cancer tissues and plasma. Theyfound that miR-9 with miR-1228 and miR-9 with miR-92a weregood plasma markers for diagnosis. Additionally, a plasma sig-nature containing miR-200b, miR-200c, miR-203 and miR-449acould distinguish myometrial invasion [138]. In another study,a plasma panel containing miR-99a/miR-100/miR-199b was ableto accurate classify endometrial cancer based on mTOR status[158]. Given that these miRNAs target members of the PI3K-AKT pathway, these miRNA may be useful plasma biomarkers(Table 3). In addition, a tissue biomarker panel of four miRNAs(miR-182, miR-183, miR-200a, miR-200c) had 95 % sensitivityand 91 % specificity at distinguishing complex hyperplasia fromendometrial cancer in formalin-fixed paraffin-embedded speci-mens [133]. As many more young obese women are diagnosedwith complex hyperplasia and want to maintain child bearingpotential, the accurate diagnosis of complex hyperplasia and en-dometrial cancer is critical [148,149,159164]. miRNAs may beuseful biomarkers for this in the future.



Table 6 In vitro studies of miRNA molecules in endometrial cancer


BCM-173 Proliferation Vinculin [139]

let-7a Proliferation AURORA B [306]

miR-103 Proliferation TIMP-3 [313]

Invasion

miR-106b Invasion TWIST1 [171,314]

EMT p21

Proliferation

miR-125b Proliferation TP53INP1 [146,315]

Invasion ERRB2

Migration

Growth (xenograft)

miR-130b Proliferation DICER1 [41,316,317]

Invasion ZEB1

EMT SNAIL

Growth (xenograft) E-cadherin, KLF4, NANOG, MDR1

miR-138 Migration NGAL [318]

Growth (xenograft)

miR-145 Differentiation OCT4 [319]

Tumorigenesis (xenograft)

miR-148a Migration WNT10B [320]

miR-152 Proliferation E2F3 [165]

Apoptosis MET

Growth (xenograft) RICTOR

miR-155 Proliferation AGTR1 [321]

miR-17-5p Bortezomib cell killing p21 [322]

miR-193a-5p Tumour growth (xenograft) YY1 [323]

miR-194 Invasion EMT BMI-1 [170]

miR-199a-3p Proliferation mTOR [193]

miR-200b, miR-200c, miR-429 family Chemotherapy sensitivity AP-2 [324]

miR-200b MMP2 activity TIMP2 [325]

miR-200c Proliferation ZEB1 [326]

Apoptosis VEGFA, FLT1, IKK, KLF9, BRD7

miR-204 Migration FOXC1 [327]

Invasion

Extracellular matrix formation

miR-204-5p Proliferation TrkB [172]

Invasion

Migration

Anchorage-independent growth

Growth (xenograft)

miR-205 Proliferation ESRRG [328]

Migration

Invasion

miR-206 Proliferation ER [329]

Invasion

miR-21 Proliferation PTEN [156]

miR-25 Apoptosis BCL2L11 [314]

miR-302 Proliferation CCND1 [145]

Migration CDK1

Growth (xenograft)



Table 6 Continued


miR-30c Proliferation MTA1 [166]

Migration

Invasion

miR-34b Migration MET [166]

Invasion



miR-503 Proliferation CCND1 [144]

Cell cycle

Anchorage-independent growth

Growth (xenograft)


Cell cycle

miR-98 Proliferation PGRMC1, CYP19A1 [330]

miR-181a PGR, DDX3X, TIMP3 [330]

Table 7 Clinically associated miRNA molecules in endometrial cancer


miR-100 Down-regulation correlates with poor overall survival [158]

miR-130b Expression correlates with myometrial invasion, stage, and expression of oestrogen receptor [41,317]

Higher expression associated with longer survival

miR-145 Low expression and DNMTB3 overexpression correlates with short survival [331]

miR-143

miR-194 Increased expression associated with longer overall survival [168]

miR-199b Tissue biomarker for EEC [158]

miR-205, miR-200a Expression predicts relapse [138]

miR-205 High expression correlates with shorter overall survival [332]

miR-206 Expression lower in grade1 and 2 tumours compared with grade 3 [329]

miR-214, miR-221, miR-222 Lower expression and higher VEGFA associated with stage IB over stage IA tumours [174]

miR-503 Expressed expression associated with longer survival [144]

miR-99a, miR-199b Plasma biomarker with high sensitivity and specificity [158]

Similar to SCCC, aberrant DNA hypermethylation of CpGislands is another mechanism for transcriptional silencing of tu-mour suppressor miRNAs in endometrial cancer that may im-prove biomarker ability. Tsuruta et al. [165] identified miR-152as a tumour suppressor that is frequently inhibited by DNA hy-permethylation in endometrial cancer. This inhibition was re-versed after restoring miR-152 expression in endometrial cancercells. Similarly, Hiroki et al. [166] showed that miR-34b is down-regulated in serous endometrial carcinoma due to promoter hy-permethylation, and ectopic expression of miR-34b inhibits cellgrowth, migration and invasion of endometrial cancer cells. Sim-ilar to SCCC, the future of endometrial cancer diagnosis maylie in epigenetic modifications in endocervical fluid on office-based Pap test. Studies suggest that miRNAs may be released viaexosomes and may exhibit epithelial-stromal cross talk within theuterus [33,167], making serum, plasma or endocervical detectionexciting avenues for future diagnostic studies.

miRNAs as prognostic markers in endometrialcancerSimilar to EOC and SCCC, recent miRNA profiling datasets havefocused on well-defined clinical characteristics to determine spe-cific miRNAs or miRNA signatures associated with importantclinical features of EEC such as relapse, myometrial invasionand overall survival (Table 7). Clinically, miR-194 was found tobe significantly lower in patients with advanced stage (stage IIIand IV) type I EEC than in patients with type II endometrialcancer [168]. Furthermore, patients with higher miR-194 levelswere shown to have a better prognosis, with a median survivaltime of 85 months compared with patients with low miR-194levels who have median survival time of 14 months [168]. EMTis a key mechanism in many cancers including endometrial can-cer [169], with miR-106b and miR-194 inhibiting EMT throughTWIST1 and BMI1 [170,171]. In addition, the oncogenic neur-otrophic receptor kinase B (TrkB) has been shown to promote



Table 8 Integrated datasets with miRNA molecules

Datasets analysed Results Reference

miRNA + mRNA + transcription factors in EOC Multiple overlapping feed forward networks focusing ononcogenic networks, including cellular differentiation,EMT, cellular proliferation, cell cycle regulation,apoptosis

[333337]

miR-521 associated with overall survival

miRNA + mRNAin EOC, breast, lung, prostate Common pathways across multiple cancer types with

connectivity weight based on thermodynamics[338]

in renal, EOC, GM miR-17 and miR-221/miR-222 common across all 3cancers

[339]

miRNA + mRNA in EOC miR-506, miR-141, miR-200c target mesenchymalphenotype

[51,340,341]

miR-506 expression associated with progression freesurvival

integrated analysis allows for ease of biologically plausiblevalidation studies

miR-29a affects proliferation and chemosensitivity

miRNA + mRNA + miRNA targeting by sequence + proteininteractions in EOC

Protein interaction based miRNA molecules (TCGAdatasets)

[342]

miRNA + mRNA + DNA methylation in EOC Lower levels of miR-502-5p, miR-128, miR-215/625 withhigher levels of targets CCND1, PPARG, RUNX1 hadworse overall survival

[343]

Significant functional and transcriptional enrichment usingmultiple platforms and datasets

[344,345]

miRNA + mRNA in SCCC Intricate regulatory networks affected by miRNAs [346]miRNA + aCGH in SCCC Functionally relevant miRNAs are affected by chromosomal

aberrations[307]

EMT and anoikus resistance in endometrial cancer [172] througha novel regulatory loop involving TrkB- signal transducer and ac-tivator of transcription 3 (STAT3)-miR-204-5p [173]. The presentstudy showed that overexpression of TrkB leads to activation ofthe JAK2/STAT3 pathway, which in turn constitutively repressesmiR-205-5p in endometrial cancer cells [173]. Functionally, rees-tablishment of miR-204-5p suppressed migration and invasion ofendometrial cancer cells and tumorigenicity in vitro and in vivo.Furthermore, lower miR-205-5p expression was correlated withtumour stage and lymph node metastasis in endometrial cancerpatients [172]. Finally, miR-1228 with miR-200c and miR-429were associated with overall survival as a tissue biomarker [138].Taken together, these findings suggest that miR-194 and miR-204-5p may be prognostic biomarkers in tissues and may be usedto augment the activities of endogenous miRNAs in endometrialcancer.

Similar to studies in EOC, miRNAs that function in angiogen-esis pathways are dysregulated in endometrial cancer. In oneEEC study, miR-15b, miR-17-5p, miR-125a, miR-214, miR-221,miR-222 andmiR-424were significantly down-regulated whereasmiR-200b and miR-210 were up-regulated in endometrial can-cers. Importantly, those miRNAs that were down-regulated wereinversely correlated with VEGFA protein levels. Although therewere no clinical associations, the present study suggests thatthese miRNAs may regulate VEGFA expression and opens upavenues for personalized therapy based on angiogenesis targets

and miRNAs that target angiogenesis pathway genes [174]. Over-all, the studies on the clinical function of miRNAs in endometrialcancers suggest specific miRNAs play a clinically significantrole. Future clinical trials need to be aimed at discerning the roleof molecular players such as miRNAs in treatment failures orrecurrent disease.

USE OF INTEGRATED ANALYSIS TODETERMINE CLINICALLY IMPORTANTNETWORKS

With the requirement for resource sharing of data by journalsand funding agencies, large amounts of data are available formulti-platform integration studies. Integrated analysis of miRNAprofiles with other datasets (i.e., mRNA, methyl-DNA, protein)allows for improved triangulation of biologically plausible net-works or characterization of specific miRNA molecules associ-ated with clinical outcomes. Additionally, examination of miRNAtarget genes allows for examination of those genes for additionalassociation with clinical outcomes. Table 8 lists clinically import-ant molecules from integrated analysis in EOC and to a lesser ex-tent SCCC. Future work should focus on integration of multipledatasets with high quality clinical data so that entire networksmodulated by miRNAs will be better understood in the context



of clinical outcomes. In the future, the resources poured intoTCGA for genomic profiling and data sharing should propel an-swers to important clinical questions in female reproductive tractcancers forward quickly.

LACK OF REPRODUCIBILITY ACROSSDIFFERENT STUDIES

From the tables above, many studies have identified uniquemiRNAs associated with the same clinical feature that do notoverlap with results from other studies. Reasons for lack of re-producibility across the miRNA studies described above are nu-merous. First, each platform uses different technology. MaturemiRNAs are formed from longer precursor species (Figure 1).Detection methods of hybridization (i.e., microarray or real-timereverse transcription PCR) may detect both the mature miRNAand the longer precursor species. Additionally, the small size ofmiRNAs and high sequence similarity sometimes differing byonly 1 nucleotide makes primer design and hybridization condi-tions tricky and can limit the number of mature miRNAs detectedby a platform [175,176]. On the other hand, NGS approaches willonly select what is prepared in the library usually based on a cut-off size for the mature miRNA species. Secondly, each patientstumour is derived from a unique individual. Although most stud-ies try to get the most homogenous population as possible basedon race, age and ethnicity, some variability will be present basedon other underlying co-morbidities, medications or unknown ge-netic admixture. Work in cervical cancer has analysed differentplatforms but found little overlap across studies, suggesting that aminimal requirement for tissue characterization and clinical fea-tures needs to be reported [177]. Finally, tumours are geneticallydiverse. TCGA datasets have shown the differences across largesample sizes for endometrial cancer and EOC [66,178,179]. Asa last note, the climate of the current peer review system doesnot give impact to studies that replicate already published studiesor studies with negative data. Thus, replicative studies may havebeen done and not published.

THE FUTURE OF miRNA MOLECULES INPERSONALIZED CANCER TREATMENT

Although in vitro, in vivo and clinical association studies suggestthat miRNAs play a role in female reproductive tract cancers,delivery of miRNA molecules to targeted cancer cells remainsa challenge. Recent advances in EOC have used aptamers thattarget MUC1, a marker of ovarian cancer, as chimeras to miR-29b or let-7i. These chimeras are cleaved by DICER, allowingmiR-29b and let-7i to act on cells, decreasing tumour volume andstem cell populations in xenograft studies and increasing apop-tosis and chemotherapy sensitivity in vitro [180182]. However,low DICER expression is associated with poor prognostic ovarianand endometrial tumours [3739]. Recent headway in therapy forEOC was made combining miRNA with small interfering RNAs(siRNAs). Nanoliposomes loaded with siRNA EPHA2 and miR-

520-5p synergistically affected in vitro migration and invasion.Importantly, the combination revealed synergistic decrease in tu-mour burden and angiogenesis features [183]. Given that nanoli-posomes loaded with siRNA EPHA2 are being evaluated for usein actual human clinical trials [184,185], these results are veryencouraging for personalized therapy options. Finally, given theangiogenesis network targeted by specific miRNAs, this signaturemay allow for selection for personalized therapy with biologicalinhibitors such as bevcizumab based on miRNA signature.

Although most cases of endometrial cancer are diagnosed atan early stage and can be cured with surgery and adjuvant therapy,patients with hormone-resistant or recurrent endometrial cancerhave a less favourable prognosis and limited effective treatmentoptions. As a result, current research is focused on developingdrugs that target specific molecular pathways in endometrial car-cinogenesis [153]. Over activation of the PI3K/AKT/mTOR path-way, through loss of PTEN, is responsible for initiating 80 % ofendometrial cancers [186]. There are several inhibitors to vari-ous components of this pathway that are currently in clinicaltrials [153]. However, targeting most signalling pathways witha single therapeutic agent can lead to compensatory activationof other molecular pathways and tumour resistance, renderingthem ineffective [187192]. Thus, combining inhibitors of thePI3K/AKT/mTOR pathway with other therapeutic options mayimprove the efficacy of the drugs and quality of life of the patient[153]. Recently miR-199a-3p was shown to regulate endometrialcancer cell proliferation by directly targeting mTOR and com-bined treatment with rapamycin showed synergistic effects [193].Furthermore, miRNAs (miR-100, miR-99 and miR-199b) target-ing mTOR kinase are significantly decreased in EEC patients[158]. Use of miRNA molecules may be a novel way to modulatemTOR and allow for innovative therapy. In other cancer types, si-lencing ofmiR-21 conferred radiosensitivity through inhibition ofthe PI3K/AKT pathway in malignant glioma cells [194], whereasoverexpression of miR-175 inhibited colorectal cancer growth bytargeting this same pathway [195]. These studies highlight theneed to identify and exploit miRNAs as therapeutic targets ofthese signalling pathways in female reproductive tract cancers.

SUMMARY OF miRNAs IN FEMALEREPRODUCTIVE TRACT CANCERS

Use of miRNA sequencing data with metadata containing clin-ical and tumour information offers opportunities for discovery ofnovel therapies, innovative prognostic molecules and clinicallyuseful biomarkers. The key to implementing this translational sci-ence is the research team. A good research team will include manypeople. The clinician is critical for obtaining clinical samples thatare well characterized over extended longitudinal time periods,incorporating many data points. The bench scientist is critical forisolating miRNAs, in vitro studies and pre-clinical in vivo studies.The bioinformatician is critical to data analysis. The clinicianscientist is critical to interpreting the data into useful informationthat can be presented to clinicians. One last opportunity for ad-vancement will be a system or app that allows busy clinicians to



access the best science of innovative therapy that is personalizedfor the one particular patient in the room at a time. Each monthhundreds of manuscripts are published regarding the use of bio-markers and innovative therapy. As the system is now, a busyclinician is lost in a sea of manuscripts. A system of informingclinicians regarding the quality of the data and implications forpersonalized therapy needs to be developed as the final compon-ent of this team.

FUNDING

Our own work was supported by The Liz Tilberis ScholarshipOvarian Cancer Research Fund through the Estate of Agatha Fort(to S.M.H.) and a Postdoctoral Training Grant from the CancerPrevention Research Institute of Texas [grant number RP140102(to M.L.)].

REFERENCES

1 Ferlay, J., Soerjomataram, I., Ervik, M., Dikshit, R., Eser, S.,Mathers, C., Rebelo, M., Parkin, D. M., Forman, D. and Bray, F.GLOBOCAN (2012) v1.0, Cancer Incidence and MortalityWorldwide: IARC CancerBase No. 11 [Internet], InternationalAgency for Research on Cancer, Lyon, France, 2013. Availablefrom: http://globocan.iarc.fr, accessed on 04/01/2014

2 Ferlay, J., Shin, H. R., Bray, F., Forman, D., Mathers, C. andParkin, D. M. (2010) Estimates of worldwide burden of cancerin 2008: GLOBOCAN 2008. Int. J. Cancer 127, 28932917CrossRef PubMed

3 Hawkins, S. M., Buchold, G. M. and Matzuk, M. M. (2011)Minireview: the roles of small RNA pathways in reproductivemedicine. Mol. Endocrinol. 25, 12571279 CrossRef PubMed

4 Bernstein, E., Caudy, A. A., Hammond, S. M. and Hannon, G. J.(2001) Role for a bidentate ribonuclease in the initiation stepof RNA interference. Nature 409, 363366 CrossRef PubMed

5 Du, T. and Zamore, P. D. (2005) microPrimer: the biogenesisand function of microRNA. Development 132, 46454652CrossRef PubMed

6 Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E.and Mello, C. C. (1998) Potent and specific genetic interferenceby double-stranded RNA in Caenorhabditis elegans. Nature391, 806811 CrossRef PubMed

7 Griffiths-Jones, S., Grocock, R. J., van Dongen, S., Bateman, A.and Enright, A. J. (2006) miRBase: microRNA sequences,targets and gene nomenclature. Nucleic Acids Res. 34,D140144 CrossRef PubMed

8 Vera, J., Lai, X., Schmitz, U. and Wolkenhauer, O. (2013)MicroRNA-regulated networks: the perfect storm for classicalmolecular biology, the ideal scenario for systems biology. Adv.Exp. Med. Biol. 774, 5576 CrossRef PubMed

9 Morin, R. D., OConnor, M. D., Griffith, M., Kuchenbauer, F.,Delaney, A., Prabhu, A. L., Zhao, Y., McDonald, H., Zeng, T.,Hirst, M. et al. (2008) Application of massively parallelsequencing to microRNA profiling and discovery in humanembryonic stem cells. Genome Res. 18, 610621CrossRef PubMed

10 Ohman, M. (2007) A-to-I editing challenger or ally to themicroRNA process. Biochimie 89, 11711176CrossRef PubMed

11 Habig, J. W., Dale, T. and Bass, B. L. (2007) miRNA editing weshould have inosine this coming. Mol. Cell 25, 792793CrossRef PubMed

12 Creighton, C. J., Benham, A. L., Zhu, H., Khan, M. F., Reid,J. G., Nagaraja, A. K., Fountain, M. D., Dziadek, O., Han, D., Ma,L. et al. (2010) Discovery of novel microRNAs in femalereproductive tract using next generation sequencing. PLoS One5, e9637 CrossRef PubMed

13 Lee, L. W., Zhang, S., Etheridge, A., Ma, L., Martin, D., Galas,D. and Wang, K. (2010) Complexity of the microRNA repertoirerevealed by next-generation sequencing. RNA 16, 21702180CrossRef PubMed

14 Landgraf, P., Rusu, M., Sheridan, R., Sewer, A., Iovino, N.,Aravin, A., Pfeffer, S., Rice, A., Kamphorst, A. O., Landthaler, M.et al. (2007) A mammalian microRNA expression atlas basedon small RNA library sequencing. Cell 129, 14011414CrossRef PubMed

15 Kuchenbauer, F., Morin, R. D., Argiropoulos, B., Petriv, O. I.,Griffith, M., Heuser, M., Yung, E., Piper, J., Delaney, A., Prabhu,A. L. et al. (2008) In-depth characterization of the microRNAtranscriptome in a leukemia progression model. Genome Res.18, 17871797 CrossRef PubMed

16 Edson, M. A., Nagaraja, A. K. and Matzuk, M. M. (2009) Themammalian ovary from genesis to revelation. Endocr. Rev. 30,624712 CrossRef PubMed

17 Starega-Roslan, J., Krol, J., Koscianska, E., Kozlowski, P.,Szlachcic, W. J., Sobczak, K. and Krzyzosiak, W. J. (2011)Structural basis of microRNA length variety. Nucleic Acids Res.39, 257268 CrossRef PubMed

18 Burroughs, A. M., Ando, Y., de Hoon, M. J., Tomaru, Y., Suzuki,H., Hayashizaki, Y. and Daub, C. O. (2011) Deep-sequencing ofhuman Argonaute-associated small RNAs provides insight intomiRNA sorting and reveals Argonaute association with RNAfragments of diverse origin. RNA Biol. 8, 158177CrossRef PubMed

19 Burroughs, A. M., Ando, Y., de Hoon, M. J., Tomaru, Y., Nishibu,T., Ukekawa, R., Funakoshi, T., Kurokawa, T., Suzuki, H.,Hayashizaki, Y. et al. (2010) A comprehensive survey of 3

animal miRNA modification events and a possible role for 3

adenylation in modulating miRNA targeting effectiveness.Genome Res. 20, 13981410 CrossRef PubMed

20 Azuma-Mukai, A., Oguri, H., Mituyama, T., Qian, Z. R., Asai, K.,Siomi, H. and Siomi, M. C. (2008) Characterization ofendogenous human Argonautes and their miRNA partners inRNA silencing. Proc. Natl. Acad. Sci. U. S. A. 105, 79647969CrossRef PubMed

21 Wyman, S. K., Knouf, E. C., Parkin, R. K., Fritz, B. R., Lin, D. W.,Dennis, L. M., Krouse, M. A., Webster, P. J. and Tewari, M.(2011) Post-transcriptional generation of miRNA variants bymultiple nucleotidyl transferases contributes to miRNAtranscriptome complexity. Genome Res. 21, 14501461CrossRef PubMed

22 Guo, L. and Lu, Z. (2010) The fate of miRNA strand throughevolutionary analysis: implication for degradation as merelycarrier strand or potential regulatory molecule? PLoS One 5,e11387 CrossRef PubMed

23 Cheng, C. P., Jaag, H. M., Jonczyk, M., Serviene, E. and Nagy, P.D. (2007) Expression of the Arabidopsis Xrn4p 5 -3

exoribonuclease facilitates degradation of tombusvirus RNAand promotes rapid emergence of viral variants in plants.Virology 368, 238248 CrossRef PubMed

24 Heravi-Moussavi, A., Anglesio, M. S., Cheng, S. W., Senz, J.,Yang, W., Prentice, L., Fejes, A. P., Chow, C., Tone, A., Kalloger,S. E. et al. (2012) Recurrent somatic DICER1 mutations innonepithelial ovarian cancers. N. Engl. J. Med. 366, 234242CrossRef PubMed

25 Anglesio, M. S., Wang, Y., Yang, W., Senz, J., Wan, A.,Heravi-Moussavi, A., Salamanca, C., Maines-Bandiera, S.,Huntsman, D. G. and Morin, G. B. (2013) Cancer-associatedsomatic DICER1 hotspot mutations cause defective miRNAprocessing and reverse-strand expression bias topredominantly mature 3p strands through loss of 5p strandcleavage. J. Pathol. 229, 400409 CrossRef PubMed



26 Gurtan, A. M., Lu, V., Bhutkar, A. and Sharp, P. A. (2012) In vivostructure-function analysis of human Dicer reveals directionalprocessing of precursor miRNAs. RNA 18, 11161122CrossRef PubMed

27 Ghersi, D. and Singh, M. (2014) Interaction-based discovery offunctionally important genes in cancers. Nucleic Acids Res. 42,e18 CrossRef PubMed

28 Takeshita, D., Zenno, S., Lee, W. C., Nagata, K., Saigo, K. andTanokura, M. (2007) Homodimeric structure and double-stranded RNA cleavage activity of the C-terminal RNase IIIdomain of human dicer. J. Mol. Biol. 374, 106120CrossRef PubMed

29 Arrate, M. P., Vincent, T., Odvody, J., Kar, R., Jones, S. N. andEischen, C. M. (2010) MicroRNA biogenesis is required forMyc-induced B-cell lymphoma development and survival. CancerRes. 70, 60836092 CrossRef PubMed

30 Kumar, M. S., Pester, R. E., Chen, C. Y., Lane, K., Chin, C., Lu,J., Kirsch, D. G., Golub, T. R. and Jacks, T. (2009) Dicer1functions as a haploinsufficient tumor suppressor. Genes Dev.23, 27002704 CrossRef PubMed

31 Lambertz, I., Nittner, D., Mestdagh, P., Denecker, G.,Vandesompele, J., Dyer, M. A. and Marine, J. C. (2010)Monoallelic but not biallelic loss of Dicer1 promotestumorigenesis in vivo. Cell Death Differ. 17, 633641CrossRef PubMed

32 Mito, J. K., Min, H. D., Ma, Y., Carter, J. E., Brigman, B. E.,Dodd, L., Dankort, D., McMahon, M. and Kirsch, D. G. (2013)Oncogene-dependent control of miRNA biogenesis andmetastatic progression in a model of undifferentiatedpleomorphic sarcoma. J. Pathol. 229, 132140CrossRef PubMed

33 Hawkins, S. M., Andreu-Vieyra, C. V., Kim, T. H., Jeong, J. W.,Hodgson, M. C., Chen, R., Creighton, C. J., Lydon, J. P.,Gunaratne, P. H., DeMayo, F. J. et al. (2012) Dysregulation ofuterine signaling pathways in progesterone receptor-Creknockout of dicer. Mol. Endocrinol. 26, 15521566CrossRef PubMed

34 Nagaraja, A. K., Andreu-Vieyra, C., Franco, H. L., Ma, L., Chen,R., Han, D. Y., Zhu, H., Agno, J. E., Gunaratne, P. H., DeMayo,F. J. et al. (2008) Deletion of Dicer in somatic cells of thefemale reproductive tract causes sterility. Mol. Endocrinol. 22,23362352 CrossRef PubMed

35 Gonzalez, G. and Behringer, R. R. (2009) Dicer is required forfemale reproductive tract development and fertility in themouse. Mol. Reprod. Dev. 76, 678688 CrossRef PubMed

36 Luense, L. J., Carletti, M. Z. and Christenson, L. K. (2009) Roleof Dicer in female fertility. Trends Endocrinol. Metab. 20,265272 CrossRef PubMed

37 Zighelboim, I., Reinhart, A. J., Gao, F., Schmidt, A. P., Mutch,D. G., Thaker, P. H. and Goodfellow, P. J. (2011) DICER1expression and outcomes in endometrioid endometrialadenocarcinoma. Cancer 117, 14461453 CrossRef PubMed

38 Torres, A., Torres, K., Paszkowski, T., Jodlowska-Jedrych, B.,Radomanski, T., Ksiazek, A. and Maciejewski, R. (2011) Majorregulators of microRNAs biogenesis Dicer and Drosha aredown-regulated in endometrial cancer. Tumour Biol. 32,769776 CrossRef PubMed

39 Merritt, W. M., Lin, Y. G., Han, L. Y., Kamat, A. A., Spannuth,W. A., Schmandt, R., Urbauer, D., Pennacchio, L. A., Cheng,J. F., Nick, A. M. et al. (2008) Dicer, Drosha, and outcomes inpatients with ovarian cancer. N. Engl. J. Med. 359, 26412650CrossRef PubMed

40 Boren, T., Xiong, Y., Hakam, A., Wenham, R., Apte, S., Wei, Z.,Kamath, S., Chen, D. T., Dressman, H. and Lancaster, J. M.(2008) MicroRNAs and their target messenger RNAsassociated with endometrial carcinogenesis. Gynecol. Oncol.110, 206215 CrossRef PubMed

41 Li, B. L., Lu, C., Lu, W., Yang, T. T., Qu, J., Hong, X. and Wan,X. P. (2013) miR-130b is an EMT-related microRNA that targetsDICER1 for aggression in endometrial cancer. Med. Oncol. 30,484 CrossRef PubMed

42 Lowe, K. A., Chia, V. M., Taylor, A., OMalley, C., Kelsh, M.,Mohamed, M., Mowat, F. S. and Goff, B. (2013) Aninternational assessment of ovarian cancer incidence andmortality. Gynecol. Oncol. 130, 107114 CrossRef PubMed

43 Shah, M. M. and Landen, C. N. (2014) Ovarian cancer stemcells: are they real and why are they important? Gynecol. Oncol.132, 483489 PubMed

44 Vaughan, S., Coward, J. I., Bast, Jr., R. C., Berchuck, A., Berek,J. S., Brenton, J. D., Coukos, G., Crum, C. C., Drapkin, R.,Etemadmoghadam, D. et al. (2011) Rethinking ovarian cancer:recommendations for improving outcomes. Nat. Rev. Cancer11, 719725 CrossRef PubMed

45 Early Breast Cancer Trialists Collaborative GroupDavies, C., Godwin, J., Gray, R., Clarke, M., Cutter, D., Darby,S., McGale, P., Pan, H. C., Taylor, C. et al. (2011) Relevance ofbreast cancer hormone receptors and other factors to theefficacy of adjuvant tamoxifen: patient-level meta-analysis ofrandomised trials. Lancet 378, 771784 CrossRef PubMed

46 Yin, W., Jiang, Y., Shen, Z., Shao, Z. and Lu, J. (2011)Trastuzumab in the adjuvant treatment of HER2-positive earlybreast cancer patients: a meta-analysis of publishedrandomized controlled trials. PLoS One 6, e21030CrossRef PubMed

47 Hammond, M. E., Hayes, D. F., Dowsett, M., Allred, D. C.,Hagerty, K. L., Badve, S., Fitzgibbons, P. L., Francis, G.,Goldstein, N. S., Hayes, M. et al. (2010) American Society ofClinical Oncology/College of American Pathologists guidelinerecommendations for immunohistochemical testing of estrogenand progesterone receptors in breast cancer (unabridgedversion). Arch. Pathol. Lab. Med. 134, e4872 PubMed

48 Wolff, A. C., Hammond, M. E., Schwartz, J. N., Hagerty, K. L.,Allred, D. C., Cote, R. J., Dowsett, M., Fitzgibbons, P. L., Hanna,W. M., Langer, A. et al. (2007) American Society of ClinicalOncology/College of American Pathologists guidelinerecommendations for human epidermal growth factor receptor2 testing in breast cancer. Arch. Pathol. Lab. Med. 131, 1843PubMed

49 Leitman, D. C., Paruthiyil, S., Vivar, O. I., Saunier, E. F., Herber,C. B., Cohen, I., Tagliaferri, M. and Speed, T. P. (2010)Regulation of specific target genes and biological responses byestrogen receptor subtype agonists. Curr. Opin. Pharmacol. 10,629636 CrossRef PubMed

50 Nagaraja, A. K., Creighton, C. J., Yu, Z., Zhu, H., Gunaratne,P. H., Reid, J. G., Olokpa, E., Itamochi, H., Ueno, N. T., Hawkins,S. M. et al. (2010) A link between mir-100 and FRAP1/mTOR inclear cell ovarian cancer. Mol. Endocrinol. 24, 447463CrossRef PubMed

51 Creighton, C. J., Hernandez-Herrera, A., Jacobsen, A., Levine,D. A., Mankoo, P., Schultz, N., Du, Y., Zhang, Y., Larsson, E.,Sheridan, R. et al. (2012) Integrated analyses of microRNAsdemonstrate their widespread influence on gene expression inhigh-grade serous ovarian carcinoma. PLoS One 7, e34546CrossRef PubMed

52 Creighton, C. J., Fountain, M. D., Yu, Z., Nagaraja, A. K., Zhu,H., Khan, M., Olokpa, E., Zariff, A., Gunaratne, P. H., Matzuk,M. M. et al. (2010) Molecular profiling uncovers ap53-associated role for microRNA-31 in inhibiting theproliferation of serous ovarian carcinomas and other cancers.Cancer Res. 70, 19061915 CrossRef PubMed

53 Iorio, M. V., Visone, R., Di Leva, G., Donati, V., Petrocca, F.,Casalini, P., Taccioli, C., Volinia, S., Liu, C. G., Alder, H. et al.(2007) MicroRNA signatures in human ovarian cancer. CancerRes. 67, 86998707 CrossRef PubMed



54 Nam, E. J., Yoon, H., Kim, S. W., Kim, H., Kim, Y. T., Kim, J. H.,Kim, J. W. and Kim, S. (2008) MicroRNA expression profiles inserous ovarian carcinoma. Clin. Cancer Res. 14, 26902695CrossRef PubMed

55 Yang, H., Kong, W., He, L., Zhao, J. J., ODonnell, J. D., Wang,J., Wenham, R. M., Coppola, D., Kruk, P. A., Nicosia, S. V. et al.(2008) MicroRNA expression profiling in human ovarian cancer:miR-214 induces cell survival and cisplatin resistance bytargeting PTEN. Cancer Res. 68, 425433 CrossRef PubMed

56 Dahiya, N., Sherman-Baust, C. A., Wang, T. L., Davidson, B.,Shih Ie, M., Zhang, Y., Wood, 3rd, W., Becker, K. G. and Morin,P. J. (2008) MicroRNA expression and identification of putativemiRNA targets in ovarian cancer. PloS One 3, e2436CrossRef PubMed

57 Bearfoot, J. L., Choong, D. Y., Gorringe, K. L. and Campbell,I. G. (2008) Genetic analysis of cancer-implicated MicroRNA inovarian cancer. Clin. Cancer Res. 14, 72467250CrossRef PubMed

58 Zhang, L., Volinia, S., Bonome, T., Calin, G. A., Greshock, J.,Yang, N., Liu, C. G., Giannakakis, A., Alexiou, P., Hasegawa, K.et al. (2008) Genomic and epigenetic alterations deregulatemicroRNA expression in human epithelial ovarian cancer. Proc.Natl. Acad. Sci. U. S. A. 105, 70047009 CrossRef PubMed

59 Wyman, S. K., Parkin, R. K., Mitchell, P. S., Fritz, B. R., OBriant,K., Godwin, A. K., Urban, N., Drescher, C. W., Knudsen, B. S.and Tewari, M. (2009) Repertoire of microRNAs in epithelialovarian cancer as determined by next generation sequencing ofsmall RNA cDNA libraries. PLoS One 4, e5311CrossRef PubMed

60 Iorio, M. V. and Croce, C. M. (2013) MicroRNA profiling inovarian cancer. Methods Mol. Biol. 1049, 187197CrossRef PubMed

61 Ryland, G. L., Bearfoot, J. L., Doyle, M. A., Boyle, S. E., Choong,D. Y., Rowley, S. M., Australian Ovarian Cancer Study GroupTothill, R. W., Gorringe, K. L. and Campbell, I. G. (2012)MicroRNA genes and their target 3 -untranslated regions areinfrequently somatically mutated in ovarian cancers. PLoS One7, e35805 CrossRef PubMed

62 Permuth-Wey, J., Chen, Z., Tsai, Y. Y., Lin, H. Y., Chen, Y. A.,Barnholtz-Sloan, J., Birrer, M. J., Chanock, S. J., Cramer, D. W.,Cunningham, J. M. et al. (2011) MicroRNA processing andbinding site polymorphisms are not replicated in the OvarianCancer Association Consortium. Cancer Epidemiol. BiomarkersPrev. 20, 17931797 CrossRef PubMed

63 Permuth-Wey, J., Lawrenson, K., Shen, H. C., Velkova, A., Tyrer,J. P., Chen, Z., Lin, H. Y., Chen, Y. A., Tsai, Y. Y., Qu, X. et al.(2013) Identification and molecular characterization of a newovarian cancer susceptibility locus at 17q21.31. Nat. Commun.4, 1627 CrossRef PubMed

64 Caiola, E., Rulli, E., Fruscio, R., Buda, A., Broggini, M. andMarabese, M. (2012) KRas-LCS6 polymorphism does notimpact on outcomes in ovarian cancer. Am. J. Cancer Res. 2,298308 PubMed

65 Ratner, E. S., Keane, F. K., Lindner, R., Tassi, R. A., Paranjape,T., Glasgow, M., Nallur, S., Deng, Y., Lu, L., Steele, L. et al.(2012) A KRAS variant is a biomarker of poor outcome,platinum chemotherapy resistance and a potential target fortherapy in ovarian cancer. Oncogene 31, 45594566CrossRef PubMed

66 Cancer Genome Atlas Research Network (2011) Integratedgenomic analyses of ovarian carcinoma.. Nature 474, 609615CrossRef PubMed

67 Landen, Jr., C. N., Birrer, M. J. and Sood, A. K. (2008) Earlyevents in the pathogenesis of epithelial ovarian cancer. J. Clin.Oncol. 26, 9951005 CrossRef PubMed

68 Zheng, H., Zhang, L., Zhao, Y., Yang, D., Song, F., Wen, Y., Hao,Q., Hu, Z., Zhang, W. and Chen, K. (2013) Plasma miRNAs asdiagnostic and prognostic biomarkers for ovarian cancer. PLoSOne 8, e77853 CrossRef PubMed

69 Ngalame, N. N., Tokar, E. J., Person, R. J., Xu, Y. and Waalkes,M. P. (2014) Aberrant microRNA expression likely controls RASoncogene activation during malignant transformation of humanprostate epithelial and stem cells by arsenic. Toxicol. Sci. 138,268277 CrossRef PubMed

70 Kohn, E. C. and Hurteau, J. (2013) Ovarian cancer: making itsown rules-again. Cancer 119, 474476 CrossRef PubMed

71 Shapira, I., Oswald, M., Lovecchio, J., Khalili, H., Menzin, A.,Whyte, J., Dos Santos, L., Liang, S., Bhuiya, T., Keogh, M. et al.(2014) Circulating biomarkers for detection of ovarian cancerand predicting cancer outcomes. Br. J. Cancer 110, 976983CrossRef PubMed

72 Hausler, S. F., Keller, A., Chandran, P. A., Ziegler, K., Zipp, K.,Heuer, S., Krockenberger, M., Engel, J. B., Honig, A., Scheffler,M. et al. (2010) Whole blood-derived miRNA profiles aspotential new tools for ovarian cancer screening. Br. J. Cancer103, 693700 CrossRef PubMed

73 Lu, M., Zhang, Q., Deng, M., Miao, J., Guo, Y., Gao, W. and Cui,Q. (2008) An analysis of human microRNA and diseaseassociations. PLoS One 3, e3420 CrossRef PubMed

74 Taylor, D. D. and Gercel-Taylor, C. (2008) MicroRNA signaturesof tumor-derived exosomes as diagnostic biomarkers of ovariancancer. Gynecol. Oncol. 110, 1321 CrossRef PubMed

75 Suryawanshi, S., Vlad, A. M., Lin, H. M., Mantia-Smaldone, G.,Laskey, R., Lee, M., Lin, Y., Donnellan, N., Klein-Patel, M., Lee,T. et al. (2013) Plasma microRNAs as novel biomarkers forendometriosis and endometriosis-associated ovarian cancer.Clin. Cancer Res. 19, 12131224 CrossRef PubMed

76 Auersperg, N. (2013) Ovarian surface epithelium as a source ofovarian cancers: unwarranted speculation or evidence-basedhypothesis? Gynecol. Oncol. 130, 246251 CrossRef PubMed

77 Auersperg, N. (2013) The origin of ovarian cancers hypotheses and controversies. Front. Biosci. (Schol. Ed.) 5,709719 CrossRef

78 Liu, Z., Liu, J., Segura, M. F., Shao, C., Lee, P., Gong, Y.,Hernando, E. and Wei, J. J. (2012) MiR-182 overexpression intumourigenesis of high-grade serous ovarian carcinoma. J.Pathol. 228, 204215 CrossRef PubMed

79 Levanon, K., Sapoznik, S., Bahar-Shany, K., Brand, H.,Shapira-Frommer, R., Korach, J., Hirsch, M. S., Roh, M. H.,Miron, A., Liu, J. F. et al. (2013) FOXO3a loss is a frequent earlyevent in high-grade pelvic serous carcinogenesis. Oncogene 33,44244432 CrossRef PubMed

80 Kim, J., Coffey, D. M., Creighton, C. J., Yu, Z., Hawkins, S. M.and Matzuk, M. M. (2012) High-grade serous ovarian cancerarises from fallopian tube in a mouse model. Proc. Natl. Acad.Sci. U. S. A. 109, 39213926 CrossRef PubMed

81 Mullany, L. K., Fan, H. Y., Liu, Z., White, L. D., Marshall, A.,Gunaratne, P., Anderson, M. L., Creighton, C. J., Xin, L.,Deavers, M. et al. (2011) Molecular and functionalcharacteristics of ovarian surface epithelial cells transformedby KrasG12D and loss of Pten in a mouse model in vivo.Oncogene 30, 35223536 CrossRef PubMed

82 Calura, E., Fruscio, R., Paracchini, L., Bignotti, E., Ravaggi, A.,Martini, P., Sales, G., Beltrame, L., Clivio, L., Ceppi, L. et al.(2013) MiRNA landscape in stage I epithelial ovarian cancerdefines the histotype specificities. Clin. Cancer Res. 19,41144123 CrossRef PubMed

83 Li, X., Lu, Y., Chen, Y., Lu, W. and Xie, X. (2013) MicroRNAprofile of paclitaxel-resistant serous ovarian carcinoma basedon formalin-fixed paraffin-embedded samples. BMC Cancer 13,216 CrossRef PubMed

84 Vecchione, A., Belletti, B., Lovat, F., Volinia, S., Chiappetta, G.,Giglio, S., Sonego, M., Cirombella, R., Onesti, E. C., Pellegrini,P. et al. (2013) A microRNA signature defines chemoresistancein ovarian cancer through modulation of angiogenesis. Proc.Natl. Acad. Sci. U. S. A. 110, 98459850 CrossRef PubMed



85 Zhao, H., Ding, Y., Tie, B., Sun, Z. F., Jiang, J. Y., Zhao, J., Lin,X. and Cui, S. (2013) miRNA expression pattern associatedwith prognosis in elderly patients with advanced OPSC andOCC. Int. J. Oncol. 43, 839849 PubMed

86 Vang, S., Wu, H. T., Fischer, A., Miller, D. H., MacLaughlan, S.,Douglass, E., Steinhoff, M., Collins, C., Smith, P. J., Brard, L.et al. (2013) Identification of ovarian cancer metastaticmiRNAs. PLoS One 8, e58226 CrossRef PubMed

Articulos Cientificos Ginecologia Obstetricia

Documents