Advertisement

The clinical utility of dysregulated microRNA expression in paediatric solid tumours

  • Karan R. Chadda
    Affiliations
    Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK
    Search for articles by this author
  • Ellen E. Blakey
    Affiliations
    Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK
    Search for articles by this author
  • Nicholas Coleman
    Correspondence
    Corresponding authors: Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK.
    Affiliations
    Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK

    Department of Paediatric Histopathology, Cambridge University Hospitals NHS Foundation Trust, Hills Road, Cambridge, CB2 0QQ, UK
    Search for articles by this author
  • Matthew J. Murray
    Correspondence
    Corresponding authors: Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK.
    Affiliations
    Department of Pathology, University of Cambridge, Tennis Court Road, Cambridge, CB2 1QP, UK

    Department of Paediatric Haematology and Oncology, Cambridge University Hospitals NHS Foundation Trust, Hills Road, Cambridge, CB2 0QQ, UK
    Search for articles by this author
Open AccessPublished:October 07, 2022DOI:https://doi.org/10.1016/j.ejca.2022.09.010

      Highlights

      • MicroRNAs (miRNAs) are short, non-protein-coding genes that regulate coding gene expression.
      • MiRNAs are dysregulated in cancer, including paediatric solid tumours.
      • Dysregulated miRNAs offer opportunities for diagnosis and risk stratification.
      • Dysregulated miRNAs also offer the possibility for future therapeutic intervention.
      • Further work will be required for routine adoption into clinical practice.

      Abstract

      MicroRNAs (miRNAs) are short, non-protein-coding genes that regulate the expression of numerous protein-coding genes. Their expression is dysregulated in cancer, where they may function as oncogenes or tumour suppressor genes. As miRNAs are highly resistant to degradation, they are ideal biomarker candidates to improve the diagnosis and clinical management of cancer, including prognostication. Furthermore, miRNAs dysregulated in malignancy represent potential therapeutic targets. The use of miRNAs for these purposes is a particularly attractive option to explore for paediatric malignancies, where the mutational burden is typically low, in contrast to cancers affecting adult patients. As childhood cancers are rare, it has taken time to accumulate the necessary body of evidence showing the potential for miRNAs to improve clinical management across this group of tumours. Here, we review the current literature regarding the potential clinical utility of miRNAs in paediatric solid tumours, which is now both timely and justified. Exploring such avenues is warranted to improve the management and outcomes of children affected by cancer.

      Keywords

      1. Introduction

      Only a very small proportion, estimated to be <5%, of the genome encodes protein-coding messenger RNA (mRNA). The dogma that the remaining ∼95% merely constitutes ‘junk’ DNA has long been dispelled, with the demonstration that the genome is pervasively transcribed, producing both protein-coding mRNAs and non-coding RNAs (ncRNAs) [
      • Kapranov P.
      • et al.
      RNA maps reveal new RNA classes and a possible function for pervasive transcription.
      ]. MicroRNAs (miRNAs) are small ncRNAs, typically 18–23 nucleotides (18–23 nt) in length, that regulate the expression of numerous protein-coding genes [
      • He L.
      • Hannon G.J.
      MicroRNAs: small RNAs with a big role in gene regulation.
      ]. The first miRNA, lin-4, discovered in 1993 in the nematode Caenorhabditis elegans, regulates development via the protein lin-14 [
      • Lee R.C.
      • Feinbaum R.L.
      • Ambros V.
      The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14.
      ]. Since then, miRNAs have been shown to have crucial roles in diverse biological processes, such as proliferation, differentiation and development, and accordingly, their temporal and spatial expression is normally tightly regulated [
      • Kloosterman W.P.
      • et al.
      In situ detection of miRNAs in animal embryos using LNA-modified oligonucleotide probes.
      ]. Consequently, miRNA dysregulation has been implicated in many disease processes, including tumorigenesis. Much recent research effort has, therefore, focused on the potential use of miRNAs as biomarkers in multiple diseases, including adult cancer, as reviewed elsewhere [
      • Condrat C.E.
      • et al.
      miRNAs as biomarkers in disease: latest findings regarding their role in diagnosis and prognosis.
      ]. Studying dysregulated miRNA expression is of particular relevance to childhood cancers as they typically have a low mutational burden compared with their adult counterparts [
      • Alexandrov L.B.
      • et al.
      Signatures of mutational processes in human cancer.
      ]. As a result, identifying DNA mutations and circulating tumour DNA (ctDNA) is likely to be of more limited use for this patient group.
      As childhood cancers are rare, it has taken time to accumulate the necessary body of evidence showing the potential for miRNAs to improve clinical management across this group of tumours; a comprehensive review is now both possible and justified. This review, therefore, outlines our current understanding of miRNA biogenesis and describes their function. We emphasise the potential for elucidating the roles of specific miRNAs in paediatric solid tumours that are likely to impact on diagnosis, risk stratification and clinical management. Finally, we discuss opportunities to select dysregulated miRNAs for future therapeutic interventions.

      2. MiRNA biogenesis

      An overview of miRNA biogenesis is shown in Fig. 1. MiRNAs are transcribed from both intragenic and intergenic regions of the genome [
      • O'Brien J.
      • et al.
      Overview of MicroRNA biogenesis, mechanisms of actions, and circulation.
      ]. RNA polymerase II, or occasionally RNA polymerase III, transcribes DNA to produce pri-miRNAs, which are approximately one kilobase (1 Kb) in length. MiRNAs can be transcribed polycistronically to contain many miRNAs, termed a ‘cluster’, e.g., miR-17–92 and miR-371–373. The vast majority (∼99%) of pri-miRNAs are processed via the canonical pathway, where pri-miRNAs are processed into pre-miRNAs by Drosha (a ribonuclease III enzyme) and DGCR8 (an RNA-binding protein [RBP]) [
      • Denli A.M.
      • et al.
      Processing of primary microRNAs by the Microprocessor complex.
      ]. The pre-miRNAs are exported from the nucleus into the cytoplasm by Exportin-5 [
      • O'Brien J.
      • et al.
      Overview of MicroRNA biogenesis, mechanisms of actions, and circulation.
      ] and further processed by Dicer, a ribonuclease (RNase) III enzyme, to form a 22 nt mature miRNA duplex [
      • Denli A.M.
      • et al.
      Processing of primary microRNAs by the Microprocessor complex.
      ]. Either miRNA strand can be incorporated into the Argonaute protein-based RNA-induced-silencing-complex (RISC) to become the guide strand [
      • Ha M.
      • Kim V.N.
      Regulation of microRNA biogenesis.
      ], then termed miRISC [
      • O'Brien J.
      • et al.
      Overview of MicroRNA biogenesis, mechanisms of actions, and circulation.
      ], and the non-incorporated strand is degraded by AGO2 [
      • Yoda M.
      • et al.
      ATP-dependent human RISC assembly pathways.
      ]. In addition, multiple other ‘non-canonical’ pathways for miRNA biogenesis are being discovered and are reviewed elsewhere [
      • Ha M.
      • Kim V.N.
      Regulation of microRNA biogenesis.
      ].
      Fig. 1
      Fig. 1Overview of microRNA biogenesis. Adapted from ‘microRNA in cancer’, by BioRender.com (2022). Retrieved from https://app.biorender.com/biorender-templates.

      3. MiRNA function

      MiRNAs control the activity of ∼50% of all protein-coding genes [
      • Krol J.
      • Loedige I.
      • Filipowicz W.
      The widespread regulation of microRNA biogenesis, function and decay.
      ], predominantly through miRNA-mediated gene silencing via miRISC. This interaction occurs via target sequences in the 3′ untranslated region (3′UTR) of mRNA strands, known as miRNA response elements (MREs). Functional interaction occurs via the 5’ ‘seed’ region of the miRNA, comprising consecutive nucleotides at positions 2–8. Nucleotides 2–7 (2-7 nt) are most critical for determining binding specificity to the 3′UTR on target mRNA transcripts [
      • Lewis B.P.
      • Burge C.B.
      • Bartel D.P.
      Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets.
      ]. Once bound, the regulation of gene expression occurs via translational repression and/or mRNA degradation, with intra-cytoplasmic structures called P-bodies playing an important role [
      • Nilsen T.W.
      Mechanisms of microRNA-mediated gene regulation in animal cells.
      ]. Physiologically, miRNA-dependent gene regulation acts on the proteome to fine-tune protein output [
      • Baek D.
      • et al.
      The impact of microRNAs on protein output.
      ]. In humans, most miRNA gene regulation is now believed to be mediated through mRNA degradation [
      • Guo H.
      • et al.
      Mammalian microRNAs predominantly act to decrease target mRNA levels.
      ,
      • Lim L.P.
      • et al.
      Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs.
      ], as detected from changes in mRNA expression [
      • Farh K.K.
      • et al.
      The widespread impact of mammalian MicroRNAs on mRNA repression and evolution.
      ]. It has also been proposed that miRNAs could potentially regulate genes by other mechanisms, which has again been reviewed elsewhere [
      • O'Brien J.
      • et al.
      Overview of MicroRNA biogenesis, mechanisms of actions, and circulation.
      ].

      4. MiRNAs in cancer

      MiRNAs play a key role in cancer development, both as oncogenes (‘oncomiRs’) and as tumour suppressor genes (TSGs) [
      • Calin G.A.
      • Croce C.M.
      MicroRNA signatures in human cancers.
      ,
      • Cho W.C.
      OncomiRs: the discovery and progress of microRNAs in cancers.
      ,
      • Esquela-Kerscher A.
      • Slack F.J.
      Oncomirs - microRNAs with a role in cancer.
      ,
      • Farazi T.A.
      • et al.
      miRNAs in human cancer.
      ,
      • Kent O.A.
      • Mendell J.T.
      A small piece in the cancer puzzle: microRNAs as tumor suppressors and oncogenes.
      ,
      • Zhang B.
      • et al.
      microRNAs as oncogenes and tumor suppressors.
      ]. The earliest discovery of miRNA involvement in human cancer was in B-cell chronic lymphocytic leukaemia, where the frequently deleted chromosome 13q14 region contains miR-15a and miR-16-1 [
      • Calin G.A.
      • et al.
      Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia.
      ], which normally act as TSGs by repressing Bcl-2, inducing apoptosis [
      • Cimmino A.
      • et al.
      miR-15 and miR-16 induce apoptosis by targeting BCL2.
      ]. Genomic amplification, for example, results in the overexpression of the miR-17–92 cluster in B-cell lymphoma [
      • Tagawa H.
      • Seto M.
      A microRNA cluster as a target of genomic amplification in malignant lymphoma.
      ]. Furthermore, genome-wide studies have shown that miRNA genes are located in ‘cancer-associated genomic regions’, with 50% of miRNAs in common fragile sites [
      • Mishra S.
      • Yadav T.
      • Rani V.
      Exploring miRNA based approaches in cancer diagnostics and therapeutics.
      ], suggesting that abnormal miRNA expression due to defects at the genome level promotes tumorigenesis. Such defects include translocations or specific DNA copy number gains/losses [
      • Calin G.A.
      • Croce C.M.
      MicroRNAs and chromosomal abnormalities in cancer cells.
      ,
      • Calin G.A.
      • et al.
      Human microRNA genes are frequently located at fragile sites and genomic regions involved in cancers.
      ].
      Epigenetic changes (heritable changes in gene expression that do not involve a change in DNA sequence) [
      • Fabbri M.
      • Calin G.A.
      Epigenetics and miRNAs in human cancer.
      ], where miRNA transcription is altered due to promoter DNA methylation [
      • Stumpel D.J.
      • et al.
      Hypermethylation of specific microRNA genes in MLL-rearranged infant acute lymphoblastic leukemia: major matters at a micro scale.
      ] or chromatin remodelling through histone modifications [
      • Li X.
      • et al.
      Gene silencing of MIR22 in acute lymphoblastic leukaemia involves histone modifications independent of promoter DNA methylation.
      ], may lead to cancer. For example, the TSG miR-127 downregulates the proto-oncogene BCL6, but in malignant cells, it is within a CpG (methylation) island, and methylation results in reduced miR-127 expression. Inhibition of DNA methylation using a histone deacetylase inhibitor replenished levels of miR-127, with concomitant BCL6 downregulation [
      • Saito Y.
      • et al.
      Specific activation of microRNA-127 with downregulation of the proto-oncogene BCL6 by chromatin-modifying drugs in human cancer cells.
      ].
      MiRNAs are also controlled by transcription factors, so the dysregulation of these alters miRNA expression. For example, the transcription factor c-Myc, which is upregulated in many cancers, has been shown to activate transcription of oncomiRs, such as the miR-17–92 cluster, and downregulate the expression of TSG miRNAs, such as miR-15a, miR-26, miR-29, miR-30, and let-7 [
      • Peng Y.
      • Croce C.M.
      The role of MicroRNAs in human cancer.
      ]. Alterations in the levels of miRNA biogenesis processing machinery may also result in cancer [
      • Arrate M.P.
      • et al.
      MicroRNA biogenesis is required for Myc-induced B-cell lymphoma development and survival.
      ,
      • Kumar M.S.
      • et al.
      Impaired microRNA processing enhances cellular transformation and tumorigenesis.
      ,
      • Thomson J.M.
      • et al.
      Extensive post-transcriptional regulation of microRNAs and its implications for cancer.
      ] or extensive post-transcriptional regulation of miRNAs [
      • Thomson J.M.
      • et al.
      Extensive post-transcriptional regulation of microRNAs and its implications for cancer.
      ,
      • Kawahara Y.
      • et al.
      Redirection of silencing targets by adenosine-to-inosine editing of miRNAs.
      ]. For example, the upregulation of Drosha in cervical cancer, through gain of chromosome 5p, results in altered miRNA profiles [
      • Muralidhar B.
      • et al.
      Global microRNA profiles in cervical squamous cell carcinoma depend on Drosha expression levels.
      ], with the miRNAs most significantly associated with Drosha overexpression implicated in carcinogenesis [
      • Muralidhar B.
      • et al.
      Functional evidence that Drosha overexpression in cervical squamous cell carcinoma affects cell phenotype and microRNA profiles.
      ]. Additionally, RBPs play an important role in miRNA biogenesis, through the binding to the stem-loop of miRNA precursors. For example, KSRP binds pre-miRNAs, promoting biogenesis [
      • Trabucchi M.
      • et al.
      The RNA-binding protein KSRP promotes the biogenesis of a subset of microRNAs.
      ], whilst LIN28 specifically binds let-7 pri- and pre-miRNAs, and blocks processing by Drosha and Dicer, respectively [
      • Viswanathan S.R.
      • Daley G.Q.
      • Gregory R.I.
      Selective blockade of microRNA processing by Lin28.
      ]. As let-7 is a known miRNA tumour suppressor [
      • Johnson C.D.
      • et al.
      The let-7 microRNA represses cell proliferation pathways in human cells.
      ], the overexpression of LIN28, which occurs in ∼15% of all human malignancies, results in reduced levels of mature let-7 and tumorigenesis [
      • Viswanathan S.R.
      • Daley G.Q.
      • Lin28
      A microRNA regulator with a macro role.
      ,
      • Viswanathan S.R.
      • et al.
      Lin28 promotes transformation and is associated with advanced human malignancies.
      ].
      Alterations in the quantitative or qualitative nature of mRNA target 3′UTRs may also alter miRNA regulation. For example, proliferating cells express mRNAs with shortened 3′ UTRs and thus fewer miRNA-binding sites [
      • Sandberg R.
      • et al.
      Proliferating cells express mRNAs with shortened 3' untranslated regions and fewer microRNA target sites.
      ], and this phenomenon also occurs in cancer cells [
      • Mayr C.
      • Bartel D.P.
      Widespread shortening of 3'UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells.
      ]. Additionally, polymorphisms or variants within 3′UTRs have been shown to increase cancer risk [
      • Godshalk S.E.
      • et al.
      A Variant in a MicroRNA complementary site in the 3' UTR of the KIT oncogene increases risk of acral melanoma.
      ] and predict therapy response [
      • Zhang W.
      • et al.
      A let-7 microRNA-binding site polymorphism in 3'-untranslated region of KRAS gene predicts response in wild-type KRAS patients with metastatic colorectal cancer treated with cetuximab monotherapy.
      ]. Finally, RBPs, such as DND1, bind to the 3′UTR of mRNAs and prevent access of, and subsequent regulation by, certain miRNAs [
      • Kedde M.
      • et al.
      RNA-binding protein Dnd1 inhibits microRNA access to target mRNA.
      ], suggesting the dysregulated expression of such proteins has the potential to result in cancer formation.
      The following section of the review specifically uses a biology-based thematic analysis to demonstrate the clinical utility of dysregulated miRNA expression in paediatric malignancies pertaining to potential biomarkers that may be used for diagnosis, disease monitoring, prognostication, and as candidate novel therapeutic targets.

      5. Review methodology

      It was not feasible to undertake a formal systematic review as the subject was too broad. Accordingly, after appropriate expert advice (Dr Bob Philips, Senior Clinical Academic and Honorary Consultant Paediatric Oncologist, Centre for Reviews and Dissemination, University of York, York, UK) and after journal approval, a hybrid review methodology was undertaken. This involved two phases. The first phase was an initial review approach, with defined inclusion/exclusion criteria, to identify potentially relevant papers. The second phase then involved selecting key papers, based on a ‘best evidence’ fashion involving impact and relevance, that exemplified fundamental points in the review regarding clinical utility. This selection of representative papers from phase two was then used in the construction of a thematically driven synthesis, using a biology-based rather than disease-based approach.
      The first review phase was undertaken in January 2022. For this, a detailed search strategy was developed using search terms that would capture relevant publications in the subject area of both paediatric malignancy and miRNAs. These search terms were specifically as follows: ((((((((cancer∗[tiab]) OR (neoplasm∗[tiab])) OR (tumor∗[tiab])) OR (tumour∗[tiab])) OR (malignanc∗[tiab])) OR (“Neoplasms" [Mesh]))) AND ((((“MicroRNAs" [Mesh]) OR (microRNA∗[tiab])) OR (miRNA∗[tiab]))))) AND (((((((“Pediatrics" [Mesh])) OR (child∗[tiab])) OR (infant∗[tiab])) OR (adolescen∗[tiab])) OR (pediatric∗[tiab])) OR (paediatric∗[tiab]))). These terms were searched in MEDLINE, and the ‘humans’ filter was applied (n = 1205) (Fig. 2). Two independent reviewers (KC/EB) subsequently screened the manuscripts using the inclusion/exclusion criteria. We included all primary literature to allow us to detect discovery studies in small cohorts, given the rarity of some paediatric cancer subtypes. We excluded manuscripts that involved leukaemia or lymphoma (n = 244), as our focus was on paediatric solid tumours, those that were not relevant to the topic of the clinical utility of miRNAs (n = 155), those that were not primary literature (n = 139) or in English (n = 29), those that were not on cancer (n = 112) or paediatric cases (n = 27), and those that were subsequently retracted (n = 2). The first phase included 497 manuscripts, of which 101 were selected for phase 2 based on impact and relevance, to highlight key principles and examples of the clinical utility of miRNAs in paediatric solid tumours.
      Fig. 2
      Fig. 2PRISMA flow diagram of the literature search for the Review.

      6. MiRNAs in cancer diagnosis and treatment monitoring

      It is likely that for each malignancy, a ‘signature’ comprising a small panel of key miRNAs will distinguish malignant tumours from benign tumours and normal tissue, as well as between different malignant tumours. Tumour tissue may be examined by in situ hybridisation (ISH) or polymerase chain reaction (qRT-PCR) to detect this differential expression, which may become routinely available for miRNAs in the future. However, the storage and preparation of fresh tissue is expensive and labour-intensive. Furthermore, a non-invasive method of detecting differential miRNA expression in malignancy, for example, from routine blood samples, would have advantages, such as potentially obviating the need for diagnostic surgical biopsy prior to definitive therapy (Fig. 3).
      Fig. 3
      Fig. 3A flow diagram illustrating the typical conventional patient journey (lower panel) through diagnosis, prognostication, therapy, and disease monitoring for a child with a solid tumour, versus a potential new approach (upper panel) with the use of miRNAs for these purposes. CSF, cerebrospinal fluid. Created with BioRender.com.
      To be of use as non-invasive blood-based markers, miRNAs released from solid tumours into the bloodstream need to be protected from endogenous RNase activity that would result in their degradation and be robustly detectable in the cell-free fraction (either plasma [containing clotting factors] or serum [without]) [
      • Murray M.J.
      • et al.
      "Future-proofing" blood processing for measurement of circulating miRNAs in samples from biobanks and prospective clinical trials.
      ]. MiRNAs released from tumour cells appear to be protected from RNase degradation by packaging within membrane-bound exosome particles, which are shed into the extracellular space from the cell membrane and then released into the bloodstream [
      • Caby M.P.
      • et al.
      Exosomal-like vesicles are present in human blood plasma.
      ,
      • Valadi H.
      • et al.
      Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells.
      ]. Furthermore, miRNA levels have been shown to be stable, even in serum samples subjected to multiple freeze–thaw cycles [
      • Chen X.
      • et al.
      Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and other diseases.
      ,
      • Mitchell P.S.
      • et al.
      Circulating microRNAs as stable blood-based markers for cancer detection.
      ] or those left at room temperature for 24 h prior to processing [
      • Mitchell P.S.
      • et al.
      Circulating microRNAs as stable blood-based markers for cancer detection.
      ]. These properties are critical when considering their potential use in routine clinical practice, where such variations in sample handling will occur. Additionally, serum miRNA levels in healthy individuals are stable and similar to that of circulating blood cells [
      • Chen X.
      • et al.
      Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and other diseases.
      ] and a good correlation exists between individual miRNA levels obtained from serum and plasma samples from the same patients [
      • Mitchell P.S.
      • et al.
      Circulating microRNAs as stable blood-based markers for cancer detection.
      ].
      It should be noted that miRNAs are not just released from tumour cells but also by cells of the tumour microenvironment and peripheral immune cells [
      • Rupaimoole R.
      • et al.
      miRNA deregulation in cancer cells and the tumor microenvironment.
      ,
      • Seijo L.M.
      • et al.
      Biomarkers in lung cancer screening: achievements, promises, and challenges.
      ]. Furthermore, the use of miRNAs is not restricted to blood-based detection, with miRNAs reliably detected in a wide range of body fluids [
      • Weber J.A.
      • et al.
      The microRNA spectrum in 12 body fluids.
      ] (Fig. 3); for example, miRNAs are used for forensic examinations. Urinary levels of miR-126 and miR-152 have been used to identify cases of bladder cancer [
      • Hanke M.
      • et al.
      A robust methodology to study urine microRNA as tumor marker: microRNA-126 and microRNA-182 are related to urinary bladder cancer.
      ], and a recent study has highlighted the potential use of miR-204-5p in urinary exosomes, as an early diagnostic biomarker for Xp11.2 translocation-positive renal cell carcinoma [
      • Kurahashi R.
      • et al.
      MicroRNA-204-5p: a novel candidate urinary biomarker of Xp11.2 translocation renal cell carcinoma.
      ]. Furthermore, miRNAs are detected within the cerebrospinal fluid (CSF) [
      • Baraniskin A.
      • et al.
      Identification of microRNAs in the cerebrospinal fluid as biomarker for the diagnosis of glioma.
      ], offering the potential for facilitating the diagnosis and monitoring of central nervous system (CNS) malignancies [
      • Murray M.J.
      • et al.
      A pipeline to quantify serum and cerebrospinal fluid microRNAs for diagnosis and detection of relapse in paediatric malignant germ-cell tumours.
      ].
      Certain technical issues still need to be resolved in blood-based miRNA detection before they can be used as a routine clinical tool, including clarification regarding the most appropriate quality control and/or normalisation methods. Whilst studies have investigated robust diagnostic serum miRNA profiling method [
      • Murray M.J.
      • et al.
      Solid tumors of childhood display specific serum microRNA profiles.
      ], there is currently no universally accepted gold standard, and such issues are reviewed in more detail elsewhere [
      • Brase J.C.
      • et al.
      Serum microRNAs as non-invasive biomarkers for cancer.
      ]. However, they hold the substantial potential to revolutionise the diagnostic (D), prognostic (P), and therapeutic (T) approach for children with cancer, and many potential biomarkers have already been identified across various solid tumours (Table 1). Examples of where miRNAs may be used in patients’ cancer pathways are shown in Fig. 3.
      Table 1Dysregulated expression and clinical role of miRNAs in paediatric solid tumours. Ordered by ascending miRNA number and then by primary clinical use (D = diagnosis, P = prognostication, T = therapy).
      CancerMiRNAKey finding(s)Conclusion(s)Primary clinical useOther clinical use(s)Ref
      Embryonal

      CNS tumours
      C19MCCNS tumours with C19MC amplification and/or LIN28 expression span various histologies but comprise a single molecular diseaseIdentifies LIN28/let-7/PI3K/mTOR axis and DNMT3B as promising therapeutics for this distinct entityTD[
      • Spence T.
      • et al.
      CNS-PNETs with C19MC amplification and/or LIN28 expression comprise a distinct histogenetic diagnostic and therapeutic entity.
      ]
      ETMRC19MCGenomic and epigenomic alterations of C19MC drive multiple feedforward loops to drive C19MC-LIN28A-MYCN oncogenic circuit, which can be abrogated by bromodomain inhibitorsHighlights C19MC as a critical oncogene in ETMRs and offers therapeutic insights using bromodomain inhibitorsT[
      • Sin-Chan P.
      • et al.
      A C19MC-LIN28A-MYCN oncogenic circuit driven by hijacked super-enhancers is a distinct therapeutic vulnerability in ETMRs: a lethal brain tumor.
      ]
      ETMRC19MC, miR-17–92Possible that tumours with C19MC or miR-17–92 amplification oversaturate the miRNA processing machineryOversaturation of the miRNA machinery could explain why all ETMRs, regardless of amplification status, show many structural aberrationsTD[
      • Lambo S.
      • et al.
      The molecular landscape of ETMR at diagnosis and relapse.
      ]
      ETMRLet-7OE of LIN28A augments SHH and Wnt signalling in precursor cells via downregulated let-7-miRNALIN28A/let-7a interaction with the SHH pathway was detected at the level of Gli mRNATD[
      • Neumann J.E.
      • et al.
      A mouse model for embryonal tumors with multilayered rosettes uncovers the therapeutic potential of Sonic-hedgehog inhibitors.
      ]
      GCTLet-7LIN28, the negative-regulator of let-7 biogenesis, was abundant in GCTs, regardless of age, site or histologyLIN28 depletion in GCT cells restored let-7 levels and repressed several oncogenic let-7 mRNA targetsTD[
      • Murray M.J.
      • et al.
      LIN28 Expression in malignant germ cell tumors downregulates let-7 and increases oncogene levels.
      ]
      WTLet-7Withdrawal of LIN28 expression reverts tumorigenesis; tumour formation is suppressed by enforced expression of let-7LIN28/let-7 pathway implicated in tumorigenesisT[
      • Urbach A.
      • et al.
      Lin28 sustains early renal progenitors and induces Wilms tumor.
      ]
      RMSMiR-7, miR-3245pMiR-7 and miR-324-5p OE reduce tumour growth in RMS models and miR-7 impaired metastatic lung colonisationMiR-7 and miR-324-5p show anti-oncogenic and anti-metastatic potentialT[
      • Molist C.
      • et al.
      miRNA-7 and miRNA-324-5p regulate alpha9-Integrin expression and exert anti-oncogenic effects in rhabdomyosarcoma.
      ]
      RMSMiR-9-5pMiR-9-5p reduction inhibited RMS cell migrationMiR-9-5p levels correlated with poor outcome and were higher in metastatic vs non-metastatic diseasePD[
      • Missiaglia E.
      • et al.
      MicroRNA and gene co-expression networks characterize biological and clinical behavior of rhabdomyosarcomas.
      ]
      NBMiR-14q32UE of 14q32 miRNAs in tumours associated with poor prognostic factors was confirmed in 226 primary NBsIdentification of miRNAs involved in the process of surviving treatment and gaining resistance in NBPD[
      • Roth S.A.
      • et al.
      Next generation sequencing of microRNAs from isogenic neuroblastoma cell lines isolated before and after treatment.
      ]
      NBMiR-15a, miR-15b, miR-16Induced expression of miR-15a, miR-15b and miR-16 reduced the proliferation, migration, and invasion of NB cells and repressed tumour formation in vivoMiR-15a, miR-15b and miR-16 exert a TSG function in NB by targeting MYCNT[
      • Chava S.
      • et al.
      miR-15a-5p, miR-15b-5p, and miR-16-5p inhibit tumor progression by directly targeting MYCN in neuroblastoma.
      ]
      OSMiR-16-1-3p, miR-16-2-3pEctopic expression of these miRNAs affected tumour growth, metastasis, chemoresistance, and invasivenessMiR-16-1-3p and miR-16-2-3p ‘passenger’ strands and ‘lead’ miR-16-5p strand act as TSGTD[
      • Maximov V.V.
      • et al.
      MiR-16-1-3p and miR-16-2-3p possess strong tumor suppressive and antimetastatic properties in osteosarcoma.
      ]
      MBMiR-17–92MiR-17–92 is highly up-regulated in MB. SHH treatment of primary cerebellar GNPs increase miR-17–92 expressionMiR-17–92 is a positive effector of SHH-mediated proliferation; aberrant expression/amplification of this miRNA confers a growth advantageD[
      • Uziel T.
      • et al.
      The miR-17∼92 cluster collaborates with the Sonic Hedgehog pathway in medulloblastoma.
      ],

      [
      • Northcott P.A.
      • et al.
      The miR-17/92 polycistron is up-regulated in sonic hedgehog-driven medulloblastomas and induced by N-myc in sonic hedgehog-treated cerebellar neural precursors.
      ]
      NB, RMSMiR-17–92MiR-17–92 cluster host gene, MIRHG1, is correlated with tumours with MYCN amplification, higher stage, and poor prognosisMiR-17–92 correlated with aggressive form of NBPD[
      • Wei J.S.
      • et al.
      microRNA profiling identifies cancer-specific and prognostic signatures in pediatric malignancies.
      ]
      MBMiR-17–92Inhibition of miR-17 and miR-19a seed families by anti-miR-17 and anti-miR-19 diminished cell proliferation in vitro and reduced tumour growth in vivoInhibition of the miR-17–92 cluster has therapeutic implications for SHH MBsT[
      • Murphy B.L.
      • et al.
      Silencing of the miR-17∼92 cluster family inhibits medulloblastoma progression.
      ]
      MBMiR-22Forced expression of miR-22 reduced cell proliferation and induced apoptosis; knockdown of miR-22 increased proliferationDownregulated miR-22 expression is associated with cell proliferation in MBs, potentially via PAPST1T[
      • Xu Q.F.
      • et al.
      MiR-22 is frequently downregulated in medulloblastomas and inhibits cell proliferation via the novel target PAPST1.
      ]
      OSMiR-22MiR-22 inhibited cell proliferation; miR-22 corroborated the effect of cisplatinMiR-22 inhibited cell proliferative activity and decreased cisplatin resistanceT[
      • Meng C.Y.
      • et al.
      MicroRNA22 mediates the cisplatin resistance of osteosarcoma cells by inhibiting autophagy via the PI3K/Akt/mTOR pathway.
      ]
      RMSMiR-22MiR-22 decreased cell proliferation, anchorage-independent growth and invasiveness, and promoted apoptosisRestoring miR-22 expression blocked tumour growth and prevented dissemination in vivoTD[
      • Bersani F.
      • et al.
      Deep sequencing reveals a novel miR-22 regulatory network with therapeutic potential in rhabdomyosarcoma.
      ]
      NBMiR-26a-5p, miR-26b-5pMYCN regulates LIN28B expression in NB via two distinct parallel mechanisms: MYCN-miR-26a-5p-LIN28B and a direct MYCN-LIN28B regulatory axisMYCN-regulated miRNAs have a role in the MYCN-driven oncogenic processT[
      • Beckers A.
      • et al.
      MYCN-driven regulatory mechanisms controlling LIN28B in neuroblastoma.
      ]
      RMSMiR-27aRe-expression of miR-27a led to PAX3:FOXO1 mRNA destabilisation and chemotherapy sensitisation in alveolar RMS cellsImplicates a HDAC3-SMARCA4-miR-27a-PAX3:FOXO1 circuit as a driver of chemotherapy-resistant alveolar RMST[
      • Bharathy N.
      • et al.
      The HDAC3-SMARCA4-miR-27a axis promotes expression of the PAX3:FOXO1 fusion oncogene in rhabdomyosarcoma.
      ]
      OSMiR-27a-3pTransfection with miR-27a-3p inhibitor decreased proliferative abilityMiR-27a-3p inhibition suppresses proliferation and invasion of OS cellsT[
      • Liu J.
      • et al.
      miR-27a-3p promotes the malignant phenotypes of osteosarcoma by targeting ten-eleven translocation 1.
      ]
      OSMiR-29bCDK6 downregulated by miR-29b in OS cells; inverse correlation between miR-29b and CDK6 protein levelsMiR-29b acts as a TSG of OS by targeting CDK6 in proliferation and migration processesT[
      • Zhu K.
      • et al.
      MiR-29b suppresses the proliferation and migration of osteosarcoma cells by targeting CDK6.
      ]
      OSMiR-29b-1MiR-29b-1 OE causes proliferation, self-renewal and chemosensitivity, associated with downregulation of stem cell, cell cycle-related, and anti-apoptotic markersMiR-29b-1 suppresses stemness properties of OS cells and is a potential therapeutic targetT[
      • Di Fiore R.
      • et al.
      MicroRNA-29b-1 impairs in vitro cell proliferation, selfrenewal and chemoresistance of human osteosarcoma 3AB-OS cancer stem cells.
      ]
      Embryonal CNS tumoursMiR-34aMiR-34a up-regulated in both MB and AT/RTIndicated tissue-specific roles rather than global TSG properties of miR-34aDP[
      • Braoudaki M.
      • et al.
      Microrna expression signatures predict patient progression and disease outcome in pediatric embryonal central nervous system neoplasms.
      ]
      MBMiR-34aTumour incidence increased and formation accelerated in mice transgenic for SmoA1 and lacking miR-34aMiR-34a is dispensable for normal development, but its loss accelerates tumour developmentD[
      • Thor T.
      • et al.
      MiR-34a deficiency accelerates medulloblastoma formation in vivo.
      ]
      ESMiR-34aWhen miR-34a expression was enforced, cells were less proliferative and sensitised to doxorubicin and vincristineRestoration of miR-34a activity may decrease malignancy and increase tumour sensitivity to current drugsTP[
      • Nakatani F.
      • et al.
      miR-34a predicts survival of Ewing's sarcoma patients and directly influences cell chemo-sensitivity and malignancy.
      ]
      NBMiR-34aMiR-34a and let-7b NB-targeted nanoparticles, individually and synergistically, reduce cell division, proliferation, neoangiogenesis and tumour growthMiR-34a may act as a TSG; miR-34a and let-7b combined replacement shows therapeutic efficacyTD[
      • Welch C.
      • Chen Y.
      • Stallings R.L.
      MicroRNA-34a functions as a potential tumor suppressor by inducing apoptosis in neuroblastoma cells.
      ],

      [
      • Cole K.A.
      • et al.
      A functional screen identifies miR-34a as a candidate neuroblastoma tumor suppressor gene.
      ,
      • Li Z.
      • Chen H.
      miR-34a inhibits proliferation, migration and invasion of paediatric neuroblastoma cells via targeting HNF4alpha.
      ,
      • Di Paolo D.
      • et al.
      Combined replenishment of miR-34a and let-7b by targeted nanoparticles inhibits tumor growth in neuroblastoma preclinical models.
      ]
      OSMiR-34aOS xenograft tumour growth was inhibited in vivo when miR-34a prodrug and doxorubicin co-administeredCombination of doxorubicin and miR-34a replacement causes antiproliferative effectsT[
      • Zhao Y.
      • et al.
      Combination therapy with bioengineered miR-34a prodrug and doxorubicin synergistically suppresses osteosarcoma growth.
      ,
      • Zhao Y.
      • et al.
      Genetically engineered pre-microRNA-34a prodrug suppresses orthotopic osteosarcoma xenograft tumor growth via the induction of apoptosis and cell cycle arrest.
      ]
      NBMiR-34bDLL1 ligand identified as the Notch pathway component OE in MYCN-amplified NB cellsMiR-34b downregulated DLL1 mRNA expression levels to arrest cell proliferation and induce neuronal differentiation in NB cellsT[
      • Bettinsoli P.
      • et al.
      Notch ligand Delta-like 1 as a novel molecular target in childhood neuroblastoma.
      ]
      RBMiR-34b/cMir-34b/c rs4938723T > C was sequenced; age at diagnosis lower in CC carriers than TT genotype in hereditary RBMir-34b/c rs4938723T > C may represent a candidate biomarker for hereditary RBD[
      • Carvalho I.N.
      • et al.
      A polymorphism in mir-34b/c as a potential biomarker for early onset of hereditary retinoblastoma.
      ]
      OSMiR-34c-5pMiR-34c-5p was UE in OS tissues and cells and inhibited proliferation, migration, and invasionMiR-34c-5p inhibited proliferation, migration and invasion, potentially via targeting FLOT2T[
      • Wang Y.
      • et al.
      The study of mechanism of miR-34c-5p targeting FLOT2 to regulate proliferation, migration and invasion of osteosarcoma cells.
      ]
      OSMiR-107MiR-107 OE increases cell viability, migration, and invasion and inhibits apoptosisMiR-107 inhibitor transfection abolished these effectsT[
      • Jiang R.
      • et al.
      MicroRNA-107 promotes proliferation, migration, and invasion of osteosarcoma cells by targeting tropomyosin 1.
      ]
      MBMiR-124CDK6 is present in approximately one-third of MBs and is an independent poor prognostic markerMiR-124 OE inhibits the proliferation of MB through inhibiting CDK6, in vitro and in vivoT[
      • Silber J.
      • et al.
      Expression of miR-124 inhibits growth of medulloblastoma cells.
      ]
      pLGGMiR-125bMiR-125b OE resulted in decreased growth and invasion, as well as apoptosisMiR-125 is frequently UE in pLGG; OE decreases cell growth and induces apoptosisTD[
      • Yuan M.
      • et al.
      MicroRNA (miR) 125b regulates cell growth and invasion in pediatric low grade glioma.
      ]
      ESMiR-130bSmall molecule inhibition of PAK1 blocked miR-130b activation of JNK and downstream AP-1 target genesMiR-130b induces proliferation, invasion and migration in vitro and increased metastatic potential in vivoT[
      • Satterfield L.
      • et al.
      miR-130b directly targets ARHGAP1 to drive activation of a metastatic CDC42-PAK1-AP1 positive feedback loop in Ewing sarcoma.
      ]
      NBMiR-137Low miR-137 in NB is associated with poor prognosis; re-expressing miR-137 in vitro increased apoptosis; miR-137 downregulated KDM1A mRNAMiR-137 acts as a TSG through targeting KDM1A mRNA; re-expression of miR-137 could be a therapeutic strategyTP[
      • Althoff K.
      • et al.
      MiR-137 functions as a tumor suppressor in neuroblastoma by downregulating KDM1A.
      ]
      pLGGMiR-139-5pOE of miR-139-5p inhibited cell proliferationMiR-139-5p downregulation drives cell proliferation by derepressing PI3K/AKT signallingTD[
      • Catanzaro G.
      • et al.
      The miR-139-5p regulates proliferation of supratentorial paediatric low-grade gliomas by targeting the PI3K/AKT/mTORC1 signalling.
      ]
      WTMiR-140-5pMiR-140-5p was UE in WTHigh levels of miR-140-5p inhibited cellular proliferation and metastasisT[
      • Liu Z.
      • et al.
      miR-140-5p could suppress tumor proliferation and progression by targeting TGFBRI/SMAD2/3 and IGF-1R/AKT signaling pathways in Wilms' tumor.
      ]
      ESMiR-143, miR-145TARBP2 restoration and systemic delivery of miR-143 or miR-145 inhibited EWS CSC clonogenicity and tumour growth in vivoCSC self-renewal and tumour maintenance may depend on the deregulation of TARBP2-dependent miRNA expressionT[
      • De Vito C.
      • et al.
      A TARBP2-dependent miRNA expression profile underlies cancer stem cell properties and provides candidate therapeutic reagents in Ewing sarcoma.
      ]
      OSMiR-143-3pMiR-143-3p expression was lower in OS tissues compared with normal and associated with poor prognosisMiR-143-3p inhibited OS cell proliferation and metastasis whilst promoting apoptosisTD, P[
      • Sun X.
      • et al.
      miR-143-3p inhibits the proliferation, migration and invasion in osteosarcoma by targeting FOSL2.
      ]
      MBMiR-183∼96∼182MiR-183∼96∼182 expression is associated with lower rates of EFS and overall survivalIdentifies a previously unrecognised molecular subgroup with poor clinical outcomeP[
      • Cho Y.J.
      • et al.
      Integrative genomic analysis of medulloblastoma identifies a molecular subgroup that drives poor clinical outcome.
      ]
      NBMiR-184MiR-184 levels increase with ATRA treatment and MYCN knockdown in vitro; pro-apoptotic when OE in vitroMiRNA profiles distinguish subtypes of NB; miR-184 UE important in NB pathogenesisTD, P[
      • Chen Y.
      • Stallings R.L.
      Differential patterns of microRNA expression in neuroblastoma are correlated with prognosis, differentiation, and apoptosis.
      ]
      NBMiR-184Pro-apoptotic effects of miR-184 ectopic OE in cells reproduced by the siRNA inhibition of AKT2MYCN may contribute to tumorigenesis miR-184 repression, increasing AKT2; miR-184 could have therapeutic implications for MYCN-amplified NBT[
      • Foley N.H.
      • et al.
      MicroRNA-184 inhibits neuroblastoma cell survival through targeting the serine/threonine kinase AKT2.
      ]
      WTMiR-185MiR-185 UE in WT, allowing the de-repression of the oncogene SIX1MiR-185 reduces growth and cell migration in vitro and tumour growth in vivoTD[
      • Imam J.S.
      • et al.
      MicroRNA-185 suppresses tumor growth and progression by targeting the Six1 oncogene in human cancers.
      ]
      MBMiR-193aMiR-193a expression in MYC amplified group 3 MB cells inhibited growth and tumorigenicity, as well as increased radiation sensitivityMiR-193a has therapeutic potential in the treatment of group 3 MBs and other MYC overexpressing cancersT[
      • Bharambe H.S.
      • et al.
      Restoration of miR-193a expression is tumor-suppressive in MYC amplified Group 3 medulloblastoma.
      ]
      HBMiR-193a-5pMiR-193a-5p/DPEP1 axis participated in HB progression via regulating the PI3K/Akt/mTORMiR-193a-5p was UE in HB tissues and associated with a poor clinical prognosisPT[
      • Cui X.
      • et al.
      DPEP1 is a direct target of miR-193a-5p and promotes hepatoblastoma progression by PI3K/Akt/mTOR pathway.
      ]
      WTMiR-195Forced expression of miR-195 impaired tumour survival and metastasis; this could be restored by LINC00473Loss-of-function of LINC00473 in vivo caused regression of WT via miR-195/IKKα-mediated growth inhibitionT[
      • Zhu S.
      • et al.
      LINC00473 antagonizes the tumour suppressor miR-195 to mediate the pathogenesis of Wilms tumour via IKKalpha.
      ]
      OSMiR-199a-3pMiR-199a-3p is decreased in OS cells; transfection of miR-199a-3p increased drug sensitivity via the downregulation of CD44CD44-miR-199a-3p axis has a role in metastasis, recurrence, and drug resistance of OSTD, P[
      • Gao Y.
      • et al.
      CD44 is a direct target of miR-199a-3p and contributes to aggressive progression in osteosarcoma.
      ]
      OSMiR-199a-5pMiR-199a-5p OE in OS; miR-199a-5p inhibition decreased cell proliferation and growthMiR-199a-5p reduction by stable antisense oligonucleotides of miR-199a-5p inhibited the OS tumour growth in nude miceTD[
      • Wang C.
      • et al.
      MicroRNA-199a-5p promotes tumour growth by dual-targeting PIAS3 and p27 in human osteosarcoma.
      ]
      OSMiR-200cMiR-200c OE in lung metastases, implicating an inhibitory feedback loop to PI3K-AKTIdentified a new role for miR-200c as a mediator of lung metastasis in OSPT[
      • Berlanga P.
      • et al.
      miR-200c and phospho-AKT as prognostic factors and mediators of osteosarcoma progression and lung metastasis.
      ]
      NBMiR-204MiR-204 expression was predictive of patient EFS and overall survival; ectopic miR-204 expression increased sensitivity to cisplatin and etoposide in vitroMiR-204 is a prognostic marker in NB, functioning, partly by increasing sensitivity to cisplatin through anti-apoptotic BCL2 downregulationPT[
      • Ryan J.
      • et al.
      MicroRNA-204 increases sensitivity of neuroblastoma cells to cisplatin and is associated with a favourable clinical outcome.
      ]
      NBMiR-204MiR-204 directly bound MYCN mRNA and repressed MYCN expression; miR-204 OE inhibited NB cell proliferation in vitro and tumorigenesis in vivoIdentifies miR-204 as a TSG and negative regulator of MYCNT[
      • Ooi C.Y.
      • et al.
      Network modeling of microRNA-mRNA interactions in neuroblastoma tumorigenesis identifies miR-204 as a direct inhibitor of MYCN.
      ]
      RCCMiR-204-5pMiR-204-5p was OE in cell lines from Tg mouse tumours and from tissue from 2 Xp11 tRCC patientsMiR-204-5p in urinary exosomes is a potential biomarker for early diagnosis of Xp11 tRCCD[
      • Kurahashi R.
      • et al.
      MicroRNA-204-5p: a novel candidate urinary biomarker of Xp11.2 translocation renal cell carcinoma.
      ]
      RMSMiR-206Muscle-specific miR-1, miR-133a/b and miR-206 are lower in RMS compared to normal skeletal muscle; miR-206 OE inhibited cell growth and migration and induced apoptosis in vitroLow miR-206 expression is associated with poor prognosis, high tumour stage and presence of metastases at diagnosisPD, T[
      • Missiaglia E.
      • et al.
      MicroRNA-206 expression levels correlate with clinical behaviour of rhabdomyosarcomas.
      ]
      RMSMiR-206Both in vitro and in vivo miR-206 acts as a TSG in fusion negative RMS, potentially through the downregulation of PAX7MiR-206 relieves the differentiation arrest in fusion negative RMS and miR-206 replacement could be a potential therapeutic strategyT[
      • Hanna J.A.
      • et al.
      PAX7 is a required target for microRNA-206-induced differentiation of fusion-negative rhabdomyosarcoma.
      ]
      GCTMiR-214-3pShowed 27 miRNA candidates with differential expression between germinomas and nongerminomatous malignant GCTsMiR-214-3p expression contributed to cisplatin resistance by targeting the pro-apoptotic protein BCL2L11.TD[
      • Hsieh T.H.
      • et al.
      Global DNA methylation analysis reveals miR-214-3p contributes to cisplatin resistance in pediatric intracranial nongerminomatous malignant germ cell tumors.
      ]
      OSMiR-216aMiR-216a expression predicted improved outcomes; miR-216a OE suppressed proliferation, migration, and invasion in vivo and in vitro by CDK14 inhibitionSuggested possibility that miR-216a activation and CDK14 inhibition may be therapeutic strategies in OSTD, P[
      • Ji Q.
      • et al.
      miR-216a inhibits osteosarcoma cell proliferation, invasion and metastasis by targeting CDK14.
      ]
      OSMiR-216a-5pMiR-216a-5p OE exerted inhibition effects via downregulating SOX5 expression; DANCR-regulated SOX5 expression by sponging to miR-216a-5pLncRNA DANCR silence inhibits SOX5-medicated progression and autophagy in OS via miR-216a-5pT[
      • Pan Z.
      • et al.
      LncRNA DANCR silence inhibits SOX5-medicated progression and autophagy in osteosarcoma via regulating miR-216a-5p.
      ]
      OSMiR-221/222Identified and validated 29 deregulated miRNAs in OSMiR-221/miR-222 were associated with time to metastasisPD[
      • Andersen G.B.
      • et al.
      miRNA profiling identifies deregulated miRNAs associated with osteosarcoma development and time to metastasis in two large cohorts.
      ]
      GCTMiR-302/367Genes downregulated by miR-302 family involved in key biological processes, e.g., apoptosis regulatorsMiRNA profiles distinguish the most common pure malignant GCT subtypes, YST and germinomaD[
      • Murray M.J.
      • et al.
      The two most common histological subtypes of malignant germ cell tumour are distinguished by global microRNA profiles, associated with differential transcription factor expression.
      ]
      GCTMiR-302/367, miR-371-373Genes downregulated by these miRNA clusters are involved in key biological processes, e.g., signalling pathwaysMiR-371-373 and miR-302/367 clusters segregate malignant GCTs from benign GCTs and controls, regardless of patient age, anatomical site or histological subtypeD[
      • Palmer R.D.
      • et al.
      Malignant germ cell tumors display common microRNA profiles resulting in global changes in expression of messenger RNA targets.
      ]
      NBMiR-323a-5p, miR-342-5pMiR-323a-5p and miR-342-5p reduced cell proliferation in vitro and in vivoNew vulnerabilities of high-risk NB through the combined inhibition of targets such as CCND1, CHAF1A, INCENP and BCL-XLT[
      • Soriano A.
      • et al.
      Functional high-throughput screening reveals miR-323a-5p and miR-342-5p as new tumor-suppressive microRNA for neuroblastoma.
      ]
      NBMiR-340Identified 67 epigenetically regulated miRNA; 42% of these were associated with poor survival when UE; miR-340 induced either differentiation or apoptosis in a cell context-dependent manner and represses SOX2Extensive epigenetic silencing of miRNAs that target a large repertoire of genes that are OE in unfavourable NBPT[
      • Das S.
      • et al.
      Modulation of neuroblastoma disease pathogenesis by an extensive network of epigenetically regulated microRNAs.
      ]
      OSMiR-340Low miR-340 and high ROCK-1 were associated with metastasis, poor response to pre-operative chemotherapy and the shortest overall and progression-free survivalMiR-340 downregulation and ROCK1 upregulation may be associated with poor prognosis in OSP[
      • Cai H.
      • et al.
      Combined microRNA-340 and ROCK1 mRNA profiling predicts tumor progression and prognosis in pediatric osteosarcoma.
      ]
      GCTMiR-371-373MiR-371-373 cluster highly expressed in seminomas, embryonal carcinoma, and yolk sac tumourPrevious miRNA 371–373 cluster finding was confirmed.D[
      • Gillis A.J.
      • et al.
      High-throughput microRNAome analysis in human germ cell tumours.
      ]
      OSMiR-377MiR-377 OE or HAT1 silencing inhibited tumour growth and reduced size in vivoMiR-377 may promote OS cell apoptosis through the inactivation of HAT1-mediated Wnt signallingT[
      • Xia P.
      • et al.
      MicroRNA-377 exerts a potent suppressive role in osteosarcoma through the involvement of the histone acetyltransferase 1-mediated Wnt axis.
      ]
      NBMiR-410, miR-487bFifteen miRNAs of the 14q32.31 cluster discriminated high-risk from low-risk NB; miR-487b and miR-410 expression was associated with disease-free survival of the non-MYCN-amplified favourable NBMiR-487b and miR-410 are potential biomarkers of relapse in favourable NBP[
      • Gattolliat C.H.
      • et al.
      Expression of miR-487b and miR-410 encoded by 14q32.31 locus is a prognostic marker in neuroblastoma.
      ]
      HBMiR-483MiR-483-5p and miR-483-3p inclusion into the four-miR signature predicted poor outcome associated with large tumours and vessel invasive growthExpansion of the four-miR signature by miR-483 serves as a prognostic biomarker in HBP[
      • Weiss J.B.W.
      • et al.
      High expression of IGF2-derived intronic miR-483 predicts outcome in hepatoblastoma.
      ]
      WTMiR-483-5pIGF2 mRNA is transcriptionally up-regulated by miR-483-5p, embedded within the IGF2 gene; ectopic expression of miR-483-5p in IGF2-dependent sarcoma cells increases tumorigenesis in vivoFunctional positive feedback loop of an intronic miR-483-5p on transcription of IGF2T[
      • Liu M.
      • et al.
      The IGF2 intronic miR-483 selectively enhances transcription from IGF2 fetal promoters and enhances tumorigenesis.
      ]
      OSMiR-486Analysis of 40 OS tissues showed miR-486 UE; low miR-486 was associated with shorter survivalMiR-486 inhibited the proliferation and migration of OS cellsTD, P[
      • He M.
      • et al.
      miR-486 suppresses the development of osteosarcoma by regulating PKC-delta pathway.
      ]
      RMSMiR-486-5pMiR-486-5p is OE in both RMS cells and exosomes; RMS serum samples showed miR-486-5p is enriched in exosomes and follow-up after chemotherapy showed a reduction to control valuesIdentified miR-486-5p in exosome-mediated oncogenic paracrine effects of RMS and its use as a potential biomarkerD[
      • Ghamloush F.
      • et al.
      The PAX3-FOXO1 oncogene alters exosome miRNA content and leads to paracrine effects mediated by exosomal miR-486.
      ]
      RMSMiR-486-5pMiR-486-5p is activated by PAX3-FOXO1 and promotes proliferation, invasion and growthInhibition of miR-486-5p in xenografts decreased tumour growthT[
      • Hanna J.A.
      • et al.
      PAX3-FOXO1 drives miR-486-5p and represses miR-221 contributing to pathogenesis of alveolar rhabdomyosarcoma.
      ]
      NBMiR-490-5pDecreased miR-490-5p levels correlated with stage, metastasis and poor survival in NB patients; miR-490-5p OE suppressed cell proliferation, migration, invasion, and induced cell apoptosisMiR-490-5p functions as a TSG in NB by targeting MYEOV; low levels are associated with a poor prognosisPT[
      • Wang J.
      • et al.
      MiR-490-5p functions as tumor suppressor in childhood neuroblastoma by targeting MYEOV.
      ]
      MBMiR-495MiR-495 expression is repressed in MB samples and an independent predictor of overall survivalMiR-495 may be a prognostic marker in MBP[
      • Wang C.
      • et al.
      MiR-495 is a predictive biomarker that downregulates GFI1 expression in medulloblastoma.
      ]
      NBMiR-497Low miR-497 expression is associated with worse EFS and overall survival; miR-497 OE reduced cell viabilityMiR-497 acts as a TSG through direct targeting of WEE1TP[
      • Creevey L.
      • et al.
      MicroRNA-497 increases apoptosis in MYCN amplified neuroblastoma cells by targeting the key cell cycle regulator WEE1.
      ]
      OSMiR-509-3pMiR-509-3p OE inhibited migration of OS cells and sensitised cells to cisplatin; AXL, which plays a role in cisplatin resistance, was downregulated upon miR-509-3p treatmentMiR-509-3p/AXL and miR-509-3p/ARHGAP1 axes have the potential therapeutic implications for resistant metastatic OST[
      • Patil S.L.
      • et al.
      MicroRNA-509-3p inhibits cellular migration, invasion, and proliferation, and sensitizes osteosarcoma to cisplatin.
      ]
      NBMiR-542-5pThirty-seven miRNAs correlated with TrkA expression, and 6 miRNAs further analysed in vitro were regulated upon TrkA transfection, suggesting a functional relationshipMiR-542-5p discriminated local vs metastatic disease and was inversely correlated with MYCN amplification and EFSP[
      • Schulte J.H.
      • et al.
      Accurate prediction of neuroblastoma outcome based on miRNA expression profiles.
      ]
      NBMiR-542-5pMiR-542-5p ectopic OE decreased the invasive potential of NB cell lines in vitro; miR-542-3p-loaded nanoparticles decreased proliferation and induced apoptosis in vivoFunctional evidence for miR-542-5p as a TSG potentially through targeting SurvivinT[
      • Bray I.
      • et al.
      MicroRNA-542-5p as a novel tumor suppressor in neuroblastoma.
      ,
      • Althoff K.
      • et al.
      miR-542-3p exerts tumor suppressive functions in neuroblastoma by downregulating Survivin.
      ]
      NBMiR-558Knockdown of miR-558 decreased growth, invasion, metastasis and angiogenesis of NB cells in vitro and in vivoMiR-558 induces the transcriptional activation of heparanase facilitating the tumorigenesis and aggressiveness of NBT[
      • Qu H.
      • et al.
      miRNA-558 promotes tumorigenesis and aggressiveness of neuroblastoma cells through activating the transcription of heparanase.
      ]
      OSMiR-590-3pMiR-590-3p was decreased OS tissues and cell linesMiR-590-3p inhibits proliferation and metastasis in OS cells via SOX9TD[
      • Wang W.T.
      • et al.
      miR-590-3p is a novel microRNA which suppresses osteosarcoma progression by targeting SOX9.
      ]
      OSMiR-598MiR-598 was UE in OS tissues, serum and cell lines; OE suppressed proliferation, migration, and invasion of cellsInhibitory role of miR-598 in OS progression in vivo, by targeting PDGFB and METTD[
      • Liu K.
      • et al.
      MiR-598: a tumor suppressor with biomarker significance in osteosarcoma.
      ]
      ESMiR-708EWS/FLI1 represses miR-708, resulting in EYA3 OE, causing chemoresistance by decreased DNA repairEYA3 inhibitors and/or re-introduction of miR-708 could sensitise EWS to chemotherapeuticsTP[
      • Robin T.P.
      • et al.
      EWS/FLI1 regulates EYA3 in Ewing sarcoma via modulation of miRNA-708, resulting in increased cell survival and chemoresistance.
      ]
      GBMMiR-1300Observed cytokinesis failure followed by apoptosis in miR-1300 transfected cells; ectopic expression of miR-1300 decreased tumour growthIdentified miR-1300 as a regulator of endomitosis with therapeutic potentialTD[
      • Boissinot M.
      • et al.
      Profiling cytotoxic microRNAs in pediatric and adult glioblastoma cells by high-content screening, identification, and validation of miR-1300.
      ]
      MBMiR-4521MiR-4521 transfection reduced proliferation and invasion of cell lines and induced cell death through caspase 3/7 activationMiR-4521 restoration may suppress the effects of aberrant FOXM1 expressionT[
      • Senfter D.
      • et al.
      High impact of miRNA-4521 on FOXM1 expression in medulloblastoma.
      ]
      CNS tumoursVariousMiR-17-5p and miR-20a obtained a high level of expression in MBs and EPs but not in PA samplesMiRNA expression depended on tumour grade and histologyD[
      • Gruszka R.
      • et al.
      mRNA and miRNA expression analyses of the MYC/E2F/miR-17-92 network in the most common pediatric brain tumors.
      ]
      GCTVariousMiR-371a-3p/miR-372-3p/miR-373-3p/miR-367-3p in serum and CSF samples showed high sensitivity and specificity in the diagnosis of malignant GCTs and enabled the early detection of relapseA robust pipeline for diagnosis and monitoring of extracranial and intracranial paediatric malignant GCTsD[
      • Murray M.J.
      • et al.
      A pipeline to quantify serum and cerebrospinal fluid microRNAs for diagnosis and detection of relapse in paediatric malignant germ-cell tumours.
      ]
      NBVariousMiR-124-3p/miR-9-3p/miR-218-5p/miR-490-5p/miR-1538 were highly OE in MYCN-amplified high-risk NBA pipeline for diagnostic serum miRNA profiling in childhood solid tumoursD[
      • Murray M.J.
      • et al.
      Solid tumors of childhood display specific serum microRNA profiles.
      ]
      OSVariousMiR-195-5p/miR-199a-3p/miR-320a/miR-374a-5p were increased in OS patients and decreased in plasma post-surgery; miR-195-5p and miR-199a-3p correlated with metastasis statusFour plasma miRNA signature as a non-invasive biomarker for OSDP[
      • Lian F.
      • et al.
      Identification of a plasma four-microRNA panel as potential noninvasive biomarker for osteosarcoma.
      ]
      OSVariousMiR-199a-5p targeted the highest number of genesIdentified 36 UE miRNAs and 182 OE miRNAs in OS samples compared to controlsD[
      • Ma G.
      • et al.
      Construction of microRNA-messenger networks for human osteosarcoma.
      ]
      OSVariousMiR-205-5p was decreased and miR-574-3p/miR-214/miR-335-5p were increased in OS samples; in metastatic patients at diagnosis, low levels of miR-214 were associated with better overall survivalValidated a signature profile of plasma miRNAs that distinguish OS from healthy animals in retrospective and prospective studies; translated these findings to 40 human plasma samplesDP[
      • Allen-Rhoades W.
      • et al.
      Cross-species identification of a plasma microRNA signature for detection, therapeutic monitoring, and prognosis in osteosarcoma.
      ]
      RBVarious537 detectable miRNAs in plasma and 625 in extracellular vesicles; identified plasma signature of 19 miRNAs present in all Rb casesIdentified plasma signature of 19 miRNAs that discriminate RB cases from controlsD[
      • Castro-Magdonel B.E.
      • et al.
      Circulating miRNome detection analysis reveals 537 miRNAS in plasma, 625 in extracellular vesicles and a discriminant plasma signature of 19 miRNAs in children with retinoblastoma from which 14 are also detected in corresponding primary tumors.
      ]
      EPVariousMultivariate analysis adjusted for age, sex, grade and localisation showed miR-17-5p as prognostic markerHigh expression of miR-17-5p was associated with reduced EFS and overall survivalP[
      • Zakrzewska M.
      • et al.
      Altered MicroRNA expression is associated with tumor grade, molecular background and outcome in childhood infratentorial ependymoma.
      ]
      NBVarious25-miRNA signature discriminates test patients with respect to progression-free and overall survivalEstablished a miRNA classifier to identify high-risk NB patients at greater risk for adverse outcomeP[
      • De Preter K.
      • et al.
      miRNA expression profiling enables risk stratification in archived and fresh neuroblastoma tumor samples.
      ]
      NBVariousDrosha or Dicer knockdown promotes cell growth in vitro; reduced miRNA biogenesis results in the global UE of miRNAs observed in advanced NB and is associated with poor prognosisCombination of 15 biomarkers delineates risk groups of NB and predicts clinical outcomePT[
      • Lin R.J.
      • et al.
      microRNA signature and expression of Dicer and Drosha can predict prognosis and delineate risk groups in neuroblastoma.
      ]
      OSVariousHigher expression of miR-27a and miR-181c pre-treatment correlated with metastatic disease; higher expression of miR-451 and miR-15b pre-treatment correlated with positive chemotherapy responseMiRNA signature-associated OS and pre-treatment biomarkers of metastasis and therapeutic responsePD, T[
      • Jones K.B.
      • et al.
      miRNA signatures associate with pathogenesis and progression of osteosarcoma.
      ]
      OSVariousMiR-155-5p/miR-135b-5p/miR-146a-5p were OE, and miR-199b-5p/miR-100-3p were UE in highly aggressive cell linesMiR-135b-5p and miR-146a-5p were predictive of metastatic capacityP[
      • Lauvrak S.U.
      • et al.
      Functional characterisation of osteosarcoma cell lines and identification of mRNAs and miRNAs associated with aggressive cancer phenotypes.
      ]
      Abbreviations: ATRA, all-trans-retinoic acid; AT/RT, atypical teratoid/rhabdoid tumour; CNS, central nervous system; CSC, cancer stem cell; DANCR, differentiation antagonising non-protein-coding RNA; EFS, event-free survival; EP, ependymoma; ETMR, embryonal tumour with multilayered rosettes; ES, Ewing sarcoma; GBM, glioblastoma multiforme; GCT, germ cell tumour; GNP, granule neuron precursor; HB, hepatoblastoma; MB, medulloblastoma; NB, neuroblastoma; OE, over-expressed; OS, osteosarcoma; pLGG, paediatric low-grade glioma; RB, retinoblastoma; RCC, renal cell carcinoma; RMS, rhabdomyosarcoma; SHH, Sonic Hedgehog; TSG, tumour suppressor gene; UE, under-expressed; WT, Wilms tumour; YST, yolk sac tumour.
      For medulloblastoma (MB), the most common paediatric brain malignancy, the miR-17–92 polycistron was highly upregulated, particularly in the MB subgroup associated with Sonic Hedgehog (SHH) signal pathway activation [
      • Northcott P.A.
      • et al.
      The miR-17/92 polycistron is up-regulated in sonic hedgehog-driven medulloblastomas and induced by N-myc in sonic hedgehog-treated cerebellar neural precursors.
      ]. This increased miR-17–92 expression leads to SHH-mediated cerebellar granule neuron precursor (GNP) cell proliferation, suggesting that the amplification of this miRNA cluster imparts a selective growth advantage to MB tumour cells [
      • Northcott P.A.
      • et al.
      The miR-17/92 polycistron is up-regulated in sonic hedgehog-driven medulloblastomas and induced by N-myc in sonic hedgehog-treated cerebellar neural precursors.
      ,
      • Uziel T.
      • et al.
      The miR-17∼92 cluster collaborates with the Sonic Hedgehog pathway in medulloblastoma.
      ]. More generally, the upregulation of the miR-17–92 cluster and its paralogs were observed in three types of paediatric brain tumour (MB, ependymoma and pilocytic astrocytoma) and the level of expression correlated with histology and WHO grade [
      • Gruszka R.
      • et al.
      mRNA and miRNA expression analyses of the MYC/E2F/miR-17-92 network in the most common pediatric brain tumors.
      ]. Furthermore, miRNA expression analysis in MB cell lines and human MB tumours showed lower miR-34a expression compared with controls, suggesting that whilst miR-34a could be expendable for normal development, the loss of miR-34a promotes tumorigenesis [
      • Thor T.
      • et al.
      MiR-34a deficiency accelerates medulloblastoma formation in vivo.
      ]. Conversely, Braoudaki et al. found miR-34a to be upregulated in embryonal tumours and in their MB group alone, suggesting potential tissue-specific roles of miR-34a rather than global tumour suppressor functions [
      • Braoudaki M.
      • et al.
      Microrna expression signatures predict patient progression and disease outcome in pediatric embryonal central nervous system neoplasms.
      ], consistent with the observation that the functional role of miRNAs may be tumour- or context-specific [
      • Annese T.
      • et al.
      microRNAs biogenesis, functions and role in tumor angiogenesis.
      ].
      Retinoblastoma (RB) is a malignancy arising from the retina in young children, with approximately 40% of cases being familial. Some patients with RB undergo enucleation due to late diagnosis and/or treatment resistance. Non-invasive markers would, therefore, be of benefit. A circulating miRNA study of RB and control cases revealed an average of 537 and 625 miRNAs detectable in plasma and extracellular vesicles, respectively [
      • Castro-Magdonel B.E.
      • et al.
      Circulating miRNome detection analysis reveals 537 miRNAS in plasma, 625 in extracellular vesicles and a discriminant plasma signature of 19 miRNAs in children with retinoblastoma from which 14 are also detected in corresponding primary tumors.
      ]. A plasma signature of 19 miRNAs was identified that distinguished RB cases from controls [
      • Castro-Magdonel B.E.
      • et al.
      Circulating miRNome detection analysis reveals 537 miRNAS in plasma, 625 in extracellular vesicles and a discriminant plasma signature of 19 miRNAs in children with retinoblastoma from which 14 are also detected in corresponding primary tumors.
      ]. Furthermore, Carvalho et al. identified a single nucleotide polymorphism (SNP) (rs4938723T>C) in the miR-34b/c gene as a potential biomarker for hereditary RB [
      • Carvalho I.N.
      • et al.
      A polymorphism in mir-34b/c as a potential biomarker for early onset of hereditary retinoblastoma.
      ].
      High-throughput screening initially demonstrated that the miR-371-373 cluster was overexpressed in adult gonadal malignant germ cell tumours (GCTs) [
      • Gillis A.J.
      • et al.
      High-throughput microRNAome analysis in human germ cell tumours.
      ]. Importantly, the miR-371-373 and miR-302/367 clusters were then shown to be universally overexpressed in all malignant GCTs, regardless of patient age (paediatric/adult), site (gonadal/extragonadal), or histological subtype [
      • Palmer R.D.
      • et al.
      Malignant germ cell tumors display common microRNA profiles resulting in global changes in expression of messenger RNA targets.
      ]. The miR-302/367 cluster showed further overexpression in the GCT subtype yolk sac tumour (YST) compared with germinoma [
      • Murray M.J.
      • et al.
      The two most common histological subtypes of malignant germ cell tumour are distinguished by global microRNA profiles, associated with differential transcription factor expression.
      ]. This is likely to be functionally significant as downregulated mRNAs in YSTs were enriched for complementary 3′UTR sequences to the common 2-7 nt seed region of miR-302ã miR-302d miRNAs, potentially contributing to their more aggressive nature [
      • Murray M.J.
      • et al.
      The two most common histological subtypes of malignant germ cell tumour are distinguished by global microRNA profiles, associated with differential transcription factor expression.
      ].
      Consequently, a panel of four miRNAs (miR-371a-3p/miR-372-3p/miR-373-3p/miR-367-3p) in serum, and CSF samples showed high sensitivity and specificity in the diagnosis of malignant GCTs and enabled early detection of relapse [
      • Murray M.J.
      • et al.
      A pipeline to quantify serum and cerebrospinal fluid microRNAs for diagnosis and detection of relapse in paediatric malignant germ-cell tumours.
      ]. Of note, a multi-institutional pooled analysis has shown the single miRNA miR-371a-3p to be sufficient for non-invasive malignant testicular GCT diagnosis [
      • Piao J.
      • et al.
      A multi-institutional pooled analysis demonstrates that circulating miR-371a-3p alone is sufficient for testicular malignant germ cell tumor diagnosis.
      ], as reviewed in detail elsewhere [
      • Leao R.
      • et al.
      Circulating MicroRNAs, the next-generation serum biomarkers in testicular germ cell tumours: a systematic review.
      ]. Miyachi et al. showed that circulating levels of the muscle-specific miRNA, miR-206, differentiated between rhabdomyosarcoma (RMS) tumours and non-RMS tumours with high sensitivity and specificity [
      • Miyachi M.
      • et al.
      Circulating muscle-specific microRNA, miR-206, as a potential diagnostic marker for rhabdomyosarcoma.
      ]. Another RMS study demonstrated that miR-486-5p was upregulated in both cells and exosomes and decreased to control value levels post-chemotherapy [
      • Ghamloush F.
      • et al.
      The PAX3-FOXO1 oncogene alters exosome miRNA content and leads to paracrine effects mediated by exosomal miR-486.
      ]. Several studies have also identified multiple miRNA level changes in osteosarcoma (OS) samples [
      • Ma G.
      • et al.
      Construction of microRNA-messenger networks for human osteosarcoma.
      ,
      • Allen-Rhoades W.
      • et al.
      Cross-species identification of a plasma microRNA signature for detection, therapeutic monitoring, and prognosis in osteosarcoma.
      ]. One particular study revealed levels of a four-plasma miRNA signature (miR-195-5p/miR-199a-3p/miR-320a/miR-374a-5p) were increased in OS patients at diagnosis and decreased significantly post-resection, suggesting a further use for monitoring disease response [
      • Lian F.
      • et al.
      Identification of a plasma four-microRNA panel as potential noninvasive biomarker for osteosarcoma.
      ]. The potential for cancer recurrence surveillance using circulating miRNA is increasing and will offer major advantages by allowing non-invasive monitoring, reducing cumulative radiation exposure from serial scans and the associated potential second cancer risk [
      • Tarin T.V.
      • Sonn G.
      • Shinghal R.
      Estimating the risk of cancer associated with imaging related radiation during surveillance for stage I testicular cancer using computerized tomography.
      ]. Scans can be reserved for children with increasing markers in treatment and/or follow-up. Furthermore, for young children, a rational reduction in scans will also reduce the need for associated sedation and/or general anaesthetics required to obtain high-quality images without motion artefacts.

      7. MiRNAs in cancer prognostication

      Numerous studies in paediatric patient populations have demonstrated the potential use of miRNAs as prognostic biomarkers, allowing for risk stratification, prediction of clinical outcome and the development of more personalised, tailored therapy. It has been suggested that the use of miRNAs in prognostication has several advantages over mRNAs. For example, miRNAs have a smaller size (18–23 nt) and are protein protected by the RISC complex, and thus are less prone to degradation. Furthermore, miRNAs are more likely to remain intact in formalin-fixed, paraffin-embedded clinical samples [
      • Li J.
      • et al.
      Comparison of miRNA expression patterns using total RNA extracted from matched samples of formalin-fixed paraffin-embedded (FFPE) cells and snap frozen cells.
      ,
      • Lu J.
      • et al.
      MicroRNA expression profiles classify human cancers.
      ].
      Neuroblastoma (NB) accounts for ∼15% of childhood cancer deaths and is clinically heterogeneous, with those with favourable prognosis potentially undergoing spontaneous regression, whereas unfavourable NB can lead to a high mortality, despite intensive multimodality treatment [
      • Schulte J.H.
      • et al.
      Accurate prediction of neuroblastoma outcome based on miRNA expression profiles.
      ]. Unfavourable NB is particularly associated with MYCN amplification and particular chromosomal aberrations [
      • Park J.R.
      • Eggert A.
      • Caron H.
      Neuroblastoma: biology, prognosis, and treatment.
      ]. Studies have demonstrated that MYCN upregulates the miR-17–92 cluster [
      • Schulte J.H.
      • et al.
      MYCN regulates oncogenic MicroRNAs in neuroblastoma.
      ], through direct binding to the promoter [
      • Fontana L.
      • et al.
      Antagomir-17-5p abolishes the growth of therapy-resistant neuroblastoma through p21 and BIM.
      ]. Wei et al. showed that the high expression of miR-17–92 cluster host gene, MIRHG1, correlated with poor prognosis and higher disease stages, as well as the expected MYCN amplification [
      • Wei J.S.
      • et al.
      microRNA profiling identifies cancer-specific and prognostic signatures in pediatric malignancies.
      ]. The miR-17–92 cluster is, therefore, implicated in unfavourable NB, potentially through the promotion of cell growth [
      • Fontana L.
      • et al.
      Antagomir-17-5p abolishes the growth of therapy-resistant neuroblastoma through p21 and BIM.
      ]. A previous study identified a 25-miRNA signature that identified high-risk NB patients with a poorer prognosis in fresh frozen and archived tissue samples [
      • De Preter K.
      • et al.
      miRNA expression profiling enables risk stratification in archived and fresh neuroblastoma tumor samples.
      ]. Interestingly, all but one of the miR-17–92 cluster miRNAs were part of this prognostic signature, and 16 of the 25 miRNAs were MYC/MYCN driven [
      • De Preter K.
      • et al.
      miRNA expression profiling enables risk stratification in archived and fresh neuroblastoma tumor samples.
      ]. Another study showed that a combination of 12 miRNAs, age at diagnosis, and mRNA expression levels of the miRNA biogenesis enzymes Dicer and Drosha could serve as a predictor of clinical outcome and separate NB patients into prognostic groups [
      • Lin R.J.
      • et al.
      microRNA signature and expression of Dicer and Drosha can predict prognosis and delineate risk groups in neuroblastoma.
      ]. Low expression levels of Dicer and Drosha resulted in global downregulation of miRNAs that correlated with poor prognosis [
      • Lin R.J.
      • et al.
      microRNA signature and expression of Dicer and Drosha can predict prognosis and delineate risk groups in neuroblastoma.
      ].
      In contrast to MYCN, TrkA shows higher levels of expression in tumours with a more favourable prognosis. Schulte et al. demonstrated that miR-542-5p expression, the most significant TrkA-correlated miRNA, was positively correlated with event-free survival (EFS) and is repressed in MYCN-amplified tumours [
      • Schulte J.H.
      • et al.
      Accurate prediction of neuroblastoma outcome based on miRNA expression profiles.
      ]. Furthermore, miR-204 expression has also been correlated with EFS and overall survival, potentially via increasing sensitivity to cisplatin and etoposide through targeting the 3′UTR of BCL2 and NTRK2 (TrkB) [
      • Ryan J.
      • et al.
      MicroRNA-204 increases sensitivity of neuroblastoma cells to cisplatin and is associated with a favourable clinical outcome.
      ]. Conversely, the downregulation of several miRNAs has been associated with poor prognosis, including, but not limited to, miR-137 [
      • Althoff K.
      • et al.
      MiR-137 functions as a tumor suppressor in neuroblastoma by downregulating KDM1A.
      ], miR-487b and miR-410 [
      • Gattolliat C.H.
      • et al.
      Expression of miR-487b and miR-410 encoded by 14q32.31 locus is a prognostic marker in neuroblastoma.
      ], miR-497 [
      • Creevey L.
      • et al.
      MicroRNA-497 increases apoptosis in MYCN amplified neuroblastoma cells by targeting the key cell cycle regulator WEE1.
      ], miR-490-5p [
      • Wang J.
      • et al.
      MiR-490-5p functions as tumor suppressor in childhood neuroblastoma by targeting MYEOV.
      ], miR-340 [
      • Das S.
      • et al.
      Modulation of neuroblastoma disease pathogenesis by an extensive network of epigenetically regulated microRNAs.
      ], and 14q32 miRNAs [
      • Roth S.A.
      • et al.
      Next generation sequencing of microRNAs from isogenic neuroblastoma cell lines isolated before and after treatment.
      ].
      Whilst the use of miRNAs as prognostic markers in NB has been particularly well studied, there is growing research in other paediatric solid tumours. OS typically has a poor prognosis, and treatment failure is usually due to metastasis before or after diagnosis [
      • Marko T.A.
      • Diessner B.J.
      • Spector L.G.
      Prevalence of metastasis at diagnosis of osteosarcoma: an international comparison.
      ], and so prognostic biomarkers, particularly of metastatic potential, would have substantial clinical utility. One study demonstrated that miR-221/miR-222 was associated with time to metastasis [
      • Andersen G.B.
      • et al.
      miRNA profiling identifies deregulated miRNAs associated with osteosarcoma development and time to metastasis in two large cohorts.
      ]. Interestingly, these same miRNAs have been shown to be linked with metastasis in other adult tumours, including gastric, colorectal and breast cancer [
      • Andersen G.B.
      • et al.
      miRNA profiling identifies deregulated miRNAs associated with osteosarcoma development and time to metastasis in two large cohorts.
      ]. Furthermore, in OS, the miR-221/PTEN/PI3K/AKT signalling pathway has been implicated for drug resistance [
      • Tang Z.
      • et al.
      Research progress of MicroRNA in chemotherapy resistance of osteosarcoma.
      ]. Berlanga et al. identified a new role for miR-200c as a mediator of lung metastasis in OS and suggested this may be via the upregulation of the reverse process of mesenchymal-to-epithelial transition (MET) [
      • Berlanga P.
      • et al.
      miR-200c and phospho-AKT as prognostic factors and mediators of osteosarcoma progression and lung metastasis.
      ]. Another study showed that miR-340 downregulation correlated with mRNA ROCK1 upregulation, and this combination occurred more often in OS with metastases and poor chemotherapy response [
      • Cai H.
      • et al.
      Combined microRNA-340 and ROCK1 mRNA profiling predicts tumor progression and prognosis in pediatric osteosarcoma.
      ]. Furthermore, miR-27a, miR-181c, miR-135b-5p, and miR-146a-5p have all been identified as potential biomarkers of OS metastasis [
      • Jones K.B.
      • et al.
      miRNA signatures associate with pathogenesis and progression of osteosarcoma.
      ,
      • Lauvrak S.U.
      • et al.
      Functional characterisation of osteosarcoma cell lines and identification of mRNAs and miRNAs associated with aggressive cancer phenotypes.
      ].
      Similar to OS, several potential prognostic miRNA biomarkers have been found in, e.g., RMS, MB, hepatoblastoma (HB), and ependymoma (EP). Low tissue miR-206 levels have been shown to be an independent predictor of shorter survival in PAX3/7-FOXO1 fusion gene negative RMS and are associated with the presence of metastases at diagnosis [
      • Missiaglia E.
      • et al.
      MicroRNA-206 expression levels correlate with clinical behaviour of rhabdomyosarcomas.
      ]. In contrast, high levels of miR-9-5-p were seen in metastatic compared with non-metastatic RMS cases and correlated with poor clinical outcome [
      • Missiaglia E.
      • et al.
      MicroRNA and gene co-expression networks characterize biological and clinical behavior of rhabdomyosarcomas.
      ]. In MB, low tissue levels of miR-495 expression were an independent predictor of overall survival [
      • Wang C.
      • et al.
      MiR-495 is a predictive biomarker that downregulates GFI1 expression in medulloblastoma.
      ]. The same study showed that miR-495 directly interacts with the Gfi1 3′UTR, and previous studies have identified a pro-oncogenic role of Gfi1 in MB groups 3/4 with poor outcome [
      • Wang C.
      • et al.
      MiR-495 is a predictive biomarker that downregulates GFI1 expression in medulloblastoma.
      ,
      • Northcott P.A.
      • et al.
      Enhancer hijacking activates GFI1 family oncogenes in medulloblastoma.
      ]. Increased miR-183–96–182 cluster miRNA levels, which is specific for MB cases expressing photoreceptor transcriptional genes, a MB molecular subgroup, were linked with lower EFS and overall survival [
      • Cho Y.J.
      • et al.
      Integrative genomic analysis of medulloblastoma identifies a molecular subgroup that drives poor clinical outcome.
      ]. For EP, the third most common paediatric brain tumour, high expression of miR-17-5p was associated with reduced EFS and overall survival [
      • Zakrzewska M.
      • et al.
      Altered MicroRNA expression is associated with tumor grade, molecular background and outcome in childhood infratentorial ependymoma.
      ]. The role of miR-17-5p is context-specific and has been shown to act as either a TSG or oncogene depending on the cellular environment [
      • Zakrzewska M.
      • et al.
      Altered MicroRNA expression is associated with tumor grade, molecular background and outcome in childhood infratentorial ependymoma.
      ,
      • Cloonan N.
      • et al.
      The miR-17-5p microRNA is a key regulator of the G1/S phase cell cycle transition.
      ]. Of note, miR-17-5p derives from the miR-17–92 cluster, a prognostic biomarker for NB. For HB, the most common primary liver neoplasm of childhood, the addition of miR-483 to a previously identified four-miRNA HB biomarker signature predicted patients with poorer outcomes [
      • Weiss J.B.W.
      • et al.
      High expression of IGF2-derived intronic miR-483 predicts outcome in hepatoblastoma.
      ]. Dipeptidase 1 (DPEP1) plays an important oncogenic role in multiple tumours, and the miR-193a-5p/DPEP1 axis has been shown to be a prognostic predictor, as well as a potential therapeutic target, in HB patients [
      • Cui X.
      • et al.
      DPEP1 is a direct target of miR-193a-5p and promotes hepatoblastoma progression by PI3K/Akt/mTOR pathway.
      ].

      8. MiRNAs as cancer therapeutic targets

      Whilst the potential clinical utility of diagnostic/prognostic biomarkers can be highlighted without necessarily elucidating any underlying functional relevance, an understanding of the mechanistic relationship between particular miRNA changes and tumorigenesis does offer new therapeutic opportunities. One of the major limitations to the success of conventional chemotherapy, and even targeted agents such as tyrosine kinase inhibitors, is the development of drug resistance. Although investigators have so far identified only a small number of protein-coding targets of miRNAs, one of the potential advantages of using miRNA-mediated therapy is that miRNAs themselves target numerous mRNAs [
      • Lewis B.P.
      • Burge C.B.
      • Bartel D.P.
      Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets.
      ,
      • Lim L.P.
      • et al.
      Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs.
      ], which are often members of multiple, key cellular pathways. As a result, mutations or variations in the 3′UTR sequence of a single mRNA, which may increase the risk of developing cancer per se [
      • Godshalk S.E.
      • et al.
      A Variant in a MicroRNA complementary site in the 3' UTR of the KIT oncogene increases risk of acral melanoma.
      ], are unlikely to be responsible for tumours becoming resistant to miRNA therapy.
      The therapeutic potential of miRNAs has been recognised for a number of years [
      • Blenkiron C.
      • Miska E.A.
      miRNAs in cancer: approaches, aetiology, diagnostics and therapy.
      ,
      • Negrini M.
      • et al.
      MicroRNAs in human cancer: from research to therapy.
      ]. Mechanisms through which this may be realised include the delivery of a single, antisense RNA strand (‘antagomir’) which binds overexpressed miRNAs or a miRNA ‘mimic’ to replace under-expressed miRNAs. The first demonstration of the potential of this technique was published in 2008, when an antagomir to miR-122 (a liver-expressed miRNA important in cholesterol and lipid metabolism) was delivered intravenously to non-human primates, resulting in a long-lasting and reversible reduction in plasma cholesterol without any evidence of toxicity [
      • Elmen J.
      • et al.
      LNA-mediated microRNA silencing in non-human primates.
      ]. Generally, however, there are potential concerns that trying to normalise levels of highly overexpressed miRNAs in cancer cells using antagomirs may lead to deleterious side effects in normal cells, unless (a) those overexpressed miRNAs found in cancer cells are virtually non-expressed in normal tissues/organs, e.g., Ref. [
      • Palmer R.D.
      • et al.
      Malignant germ cell tumors display common microRNA profiles resulting in global changes in expression of messenger RNA targets.
      ], and/or (b) effective target delivery is utilised. This may involve using molecules or antibodies that bind to receptors only expressed on the surface of cancer cells. Perhaps, a more tractable approach to miRNA-mediated therapy will be to identify miRNAs that are downregulated in cancer cells, but expressed at relatively high levels in normal cells, so that the latter are less susceptible to unwanted side effects caused by increases in miRNA concentration through miRNA mimic replenishment methodology. Indeed, this approach was demonstrated in a murine model of hepatocellular carcinoma, where miR-26a is downregulated [
      • Kota J.
      • et al.
      Therapeutic microRNA delivery suppresses tumorigenesis in a murine liver cancer model.
      ]. Systemic administration of miR-26a, using an adeno-associated virus, led to the inhibition of cancer cell proliferation, apoptosis, and protection from disease progression without concomitant toxicity [
      • Kota J.
      • et al.
      Therapeutic microRNA delivery suppresses tumorigenesis in a murine liver cancer model.
      ]. The recent years have seen the translation of miRNA therapeutics in human trials. For example, anti-miR-122 (miravirsen) administration caused dose-dependent reduction in hepatitis C virus (HCV) RNA levels with no evidence of viral resistance in patients with chronic HCV genotype 1 infection [
      • Janssen H.L.
      • et al.
      Treatment of HCV infection by targeting microRNA.
      ]. In mesothelioma, a miR-16-based mimic entered a human phase 1 clinical trial with an acceptable safety profile [
      • van Zandwijk N.
      • et al.
      Safety and activity of microRNA-loaded minicells in patients with recurrent malignant pleural mesothelioma: a first-in-man, phase 1, open-label, dose-escalation study.
      ]. The translation of our understanding of miRNA expression in childhood cancers to in vitro studies, and subsequently to appropriate in vivo models (Table 1), will be essential if we are to fully realise the role that these small ncRNAs play in paediatric malignancy.
      OS is highly malignant and overcoming drug resistance is important for advances to be made in clinical outcomes [
      • Tang Z.
      • et al.
      Research progress of MicroRNA in chemotherapy resistance of osteosarcoma.
      ]. Recently, the therapeutic potential of many miRNAs identified in the pathogenesis of OS through in vitro and in vivo experiments has been highlighted. The overexpression of multiple miRNAs has been shown to reduce OS cell proliferation and/or migration, via targeting of mRNA targets, including miR-29b-1 [
      • Di Fiore R.
      • et al.
      MicroRNA-29b-1 impairs in vitro cell proliferation, selfrenewal and chemoresistance of human osteosarcoma 3AB-OS cancer stem cells.
      ], miR-29b via CDK6 [
      • Zhu K.
      • et al.
      MiR-29b suppresses the proliferation and migration of osteosarcoma cells by targeting CDK6.
      ], miR-486 via PKC-δ signalling pathways [
      • He M.
      • et al.
      miR-486 suppresses the development of osteosarcoma by regulating PKC-delta pathway.
      ], miR-143-3p via FOS-like antigen 2 (FOSL2) [
      • Sun X.
      • et al.
      miR-143-3p inhibits the proliferation, migration and invasion in osteosarcoma by targeting FOSL2.
      ], miR-590-3p via SOX9 [
      • Wang W.T.
      • et al.
      miR-590-3p is a novel microRNA which suppresses osteosarcoma progression by targeting SOX9.
      ], miR-34c-5p via FLOT2 [
      • Wang Y.
      • et al.
      The study of mechanism of miR-34c-5p targeting FLOT2 to regulate proliferation, migration and invasion of osteosarcoma cells.
      ], miR-5093p via ARHGAP1 [
      • Patil S.L.
      • et al.
      MicroRNA-509-3p inhibits cellular migration, invasion, and proliferation, and sensitizes osteosarcoma to cisplatin.
      ], miR-598 via PDGFB and MET [
      • Liu K.
      • et al.
      MiR-598: a tumor suppressor with biomarker significance in osteosarcoma.
      ], miR-216a via CDK14 [
      • Ji Q.
      • et al.
      miR-216a inhibits osteosarcoma cell proliferation, invasion and metastasis by targeting CDK14.
      ], miR-377 via HAT1-mediated Wnt signalling [
      • Xia P.
      • et al.
      MicroRNA-377 exerts a potent suppressive role in osteosarcoma through the involvement of the histone acetyltransferase 1-mediated Wnt axis.
      ] and miR-16-1-3p and miR-16-2-3p via FGFR2 [
      • Maximov V.V.
      • et al.
      MiR-16-1-3p and miR-16-2-3p possess strong tumor suppressive and antimetastatic properties in osteosarcoma.
      ]. Similarly, the inhibition of some miRNAs has been shown to reduce proliferation in OS cells, including miR-27a-3p via TET1 [
      • Liu J.
      • et al.
      miR-27a-3p promotes the malignant phenotypes of osteosarcoma by targeting ten-eleven translocation 1.
      ], miR-107 via TPM1 (tropomyosin 1) [
      • Jiang R.
      • et al.
      MicroRNA-107 promotes proliferation, migration, and invasion of osteosarcoma cells by targeting tropomyosin 1.
      ], and miR-199a-5p via PIAS3 and p27 [
      • Wang C.
      • et al.
      MicroRNA-199a-5p promotes tumour growth by dual-targeting PIAS3 and p27 in human osteosarcoma.
      ]. Aberrant expression of differentiation antagonising non-protein-coding RNA (DANCR) has been reported in OS, with DANCR silencing suppressing OS cell progression through targeting of miR-216a-5p and downregulation of SOX [
      • Pan Z.
      • et al.
      LncRNA DANCR silence inhibits SOX5-medicated progression and autophagy in osteosarcoma via regulating miR-216a-5p.
      ]. Some studies on OS have shown that miRNAs can increase sensitisation to chemotherapy agents, thus offering the potential of combined therapy that may combat drug resistance and reduce the toxicity of conventional therapy. For example, Gao et al. demonstrated that miR-199a-3p transfection increased the sensitivity of OS cells to doxorubicin, potentially via the downregulation of CD44 [
      • Gao Y.
      • et al.
      CD44 is a direct target of miR-199a-3p and contributes to aggressive progression in osteosarcoma.
      ]. MiR-22 has been shown to inhibit the proliferation of OS cells and increase the anti-proliferative action of cisplatin, both in vitro and in vivo [
      • Meng C.Y.
      • et al.
      MicroRNA22 mediates the cisplatin resistance of osteosarcoma cells by inhibiting autophagy via the PI3K/Akt/mTOR pathway.
      ]. Bioengineered miR-34a prodrug, with and without doxorubicin, has been shown to suppress OS xenograft tumour growth, representing a putative novel therapeutic agent [
      • Zhao Y.
      • et al.
      Combination therapy with bioengineered miR-34a prodrug and doxorubicin synergistically suppresses osteosarcoma growth.
      ,
      • Zhao Y.
      • et al.
      Genetically engineered pre-microRNA-34a prodrug suppresses orthotopic osteosarcoma xenograft tumor growth via the induction of apoptosis and cell cycle arrest.
      ].
      In NB, exogenous miR-34a administration decreased cell proliferation, supporting the role for miR-34a as a TSG [
      • Welch C.
      • Chen Y.
      • Stallings R.L.
      MicroRNA-34a functions as a potential tumor suppressor by inducing apoptosis in neuroblastoma cells.
      ,
      • Cole K.A.
      • et al.
      A functional screen identifies miR-34a as a candidate neuroblastoma tumor suppressor gene.
      ,
      • Li Z.
      • Chen H.
      miR-34a inhibits proliferation, migration and invasion of paediatric neuroblastoma cells via targeting HNF4alpha.
      ]. Recently, delivery of both miR-34a and let-7b via NB-targeted nanoparticles, resulted in synergistic reduction in proliferation, neo-angiogenesis, and tumour growth/burden in orthotopic xenografts and increased survival [
      • Di Paolo D.
      • et al.
      Combined replenishment of miR-34a and let-7b by targeted nanoparticles inhibits tumor growth in neuroblastoma preclinical models.
      ]. Another study showed that among miR-34 family members, miR-34b downregulated DLL1 expression and arrested NB cell proliferation, thus suggesting that the Notch ligand DLL1 could be an attractive therapeutic target in NB through miRNA interaction [
      • Bettinsoli P.
      • et al.
      Notch ligand Delta-like 1 as a novel molecular target in childhood neuroblastoma.
      ]. However, it should be noted that a phase 1 clinical trial of a liposomal miR-34a mimic, MRX34, in adult patients with advanced solid tumours, was stopped early due to serious immune-related adverse events, so the clinical application of these in vitro and in vivo studies remains a key barrier to overcome [
      • Hong D.S.
      • et al.
      Phase 1 study of MRX34, a liposomal miR-34a mimic, in patients with advanced solid tumours.
      ]. Other miRNAs identified as potential TSGs in NB pathogenesis in functional experiments include miR-542-3p and miR-542-5p [
      • Bray I.
      • et al.
      MicroRNA-542-5p as a novel tumor suppressor in neuroblastoma.
      ,
      • Althoff K.
      • et al.
      miR-542-3p exerts tumor suppressive functions in neuroblastoma by downregulating Survivin.
      ], miR-204 [
      • Ooi C.Y.
      • et al.
      Network modeling of microRNA-mRNA interactions in neuroblastoma tumorigenesis identifies miR-204 as a direct inhibitor of MYCN.
      ], miR-137 [
      • Althoff K.
      • et al.
      MiR-137 functions as a tumor suppressor in neuroblastoma by downregulating KDM1A.
      ], miR-497 [
      • Creevey L.
      • et al.
      MicroRNA-497 increases apoptosis in MYCN amplified neuroblastoma cells by targeting the key cell cycle regulator WEE1.
      ], miR-323a-5p and miR-342-5p [
      • Soriano A.
      • et al.
      Functional high-throughput screening reveals miR-323a-5p and miR-342-5p as new tumor-suppressive microRNA for neuroblastoma.
      ], miR-15a-5p, miR-15b-5p and miR-16-5p [
      • Chava S.
      • et al.
      miR-15a-5p, miR-15b-5p, and miR-16-5p inhibit tumor progression by directly targeting MYCN in neuroblastoma.
      ], miR-26a-5p and miR-26b-5p [
      • Beckers A.
      • et al.
      MYCN-driven regulatory mechanisms controlling LIN28B in neuroblastoma.
      ], and miR-184 [
      • Chen Y.
      • Stallings R.L.
      Differential patterns of microRNA expression in neuroblastoma are correlated with prognosis, differentiation, and apoptosis.
      ,
      • Foley N.H.
      • et al.
      MicroRNA-184 inhibits neuroblastoma cell survival through targeting the serine/threonine kinase AKT2.
      ]. Conversely, miR-558 has been shown to have an oncogenic role as overexpression increased growth, invasion, metastasis and angiogenesis of NB cells via enhancing heparanase, an endogenous endoglycosidase that degrades heparan sulphate proteoglycans [
      • Qu H.
      • et al.
      miRNA-558 promotes tumorigenesis and aggressiveness of neuroblastoma cells through activating the transcription of heparanase.
      ].
      In MB studies, the transfection of miR-4521 reduced cell proliferation in vitro [
      • Senfter D.
      • et al.
      High impact of miRNA-4521 on FOXM1 expression in medulloblastoma.
      ], and overexpression of miR-124 reduced proliferation both in vitro and in vivo [
      • Silber J.
      • et al.
      Expression of miR-124 inhibits growth of medulloblastoma cells.
      ]. Conversely, the inhibition of miR-17–92 cluster miRNAs reduced cell proliferation in vitro and tumour growth in vivo [
      • Murphy B.L.
      • et al.
      Silencing of the miR-17∼92 cluster family inhibits medulloblastoma progression.
      ]. Many of the miRNA therapeutic studies described above have focused on increasing tumour sensitisation to chemotherapy regimens. Interestingly, Bharambe et al. observed that miR-193a expression increased the radiation sensitivity of MB cells [
      • Bharambe H.S.
      • et al.
      Restoration of miR-193a expression is tumor-suppressive in MYC amplified Group 3 medulloblastoma.
      ]. MiR-22 inhibits the proliferation of MB [
      • Xu Q.F.
      • et al.
      MiR-22 is frequently downregulated in medulloblastomas and inhibits cell proliferation via the novel target PAPST1.
      ], OS [
      • Meng C.Y.
      • et al.
      MicroRNA22 mediates the cisplatin resistance of osteosarcoma cells by inhibiting autophagy via the PI3K/Akt/mTOR pathway.
      ] and RMS [
      • Bersani F.
      • et al.
      Deep sequencing reveals a novel miR-22 regulatory network with therapeutic potential in rhabdomyosarcoma.
      ] cells. RMS is classified molecularly as PAX3/7-FOXO1 fusion positive or negative [
      • Hanna J.A.
      • et al.
      PAX7 is a required target for microRNA-206-induced differentiation of fusion-negative rhabdomyosarcoma.
      ]. Of note, differentiation therapy represents a possible therapeutic avenue for tumours, such as RMS, and miR-206 acts as a TSG promoting cell cycle exit and myogenic differentiation in fusion-negative RMS through various targets, including CCND2, NOTCH3, PAX3 and PAX7 [
      • Hanna J.A.
      • et al.
      PAX7 is a required target for microRNA-206-induced differentiation of fusion-negative rhabdomyosarcoma.
      ]. In fusion-positive RMS, miR-486-5p is activated by PAX3-FOXO1 and promotes proliferation, invasion and growth. Thus, miR-486-5p inhibition in vivo reduced tumour growth [
      • Hanna J.A.
      • et al.
      PAX3-FOXO1 drives miR-486-5p and represses miR-221 contributing to pathogenesis of alveolar rhabdomyosarcoma.
      ]. In contrast, the re-expression of miR-27a promoted chemotherapy sensitisation both in vitro and for the more aggressive alveolar RMS subtype [
      • Bharathy N.
      • et al.
      The HDAC3-SMARCA4-miR-27a axis promotes expression of the PAX3:FOXO1 fusion oncogene in rhabdomyosarcoma.
      ]. Interestingly, the overexpression of miR-7 has been shown to impair metastatic lung colonisation in RMS models, thus offering a novel potential therapeutic approach [
      • Molist C.
      • et al.
      miRNA-7 and miRNA-324-5p regulate alpha9-Integrin expression and exert anti-oncogenic effects in rhabdomyosarcoma.
      ].
      The EWS-FLI-1 fusion gene, which occurs as a result of the t(11;22)(q24;q12) translocation and is present in approximately 85–90% of all Ewing sarcomas (ES), initiates reprogramming towards cancer stem cells in vitro, through the modulation of miR-145 and the pluripotency factor SOX2 [
      • Riggi N.
      • et al.
      EWS-FLI-1 modulates miRNA145 and SOX2 expression to initiate mesenchymal stem cell reprogramming toward Ewing sarcoma cancer stem cells.
      ]. Vito et al. showed that the systemic delivery of synthetic miR-143 or miR-145 -reduced ES cell clonogenicity and tumour growth in vivo [
      • De Vito C.
      • et al.
      A TARBP2-dependent miRNA expression profile underlies cancer stem cell properties and provides candidate therapeutic reagents in Ewing sarcoma.
      ]. Unfortunately, ES has a tendency for metastasis and, as a result, has a poor survival rate that has not improved significantly over the last 30 years [
      • Gaspar N.
      • et al.
      Ewing sarcoma: current management and future approaches through collaboration.
      ]. Nevertheless, miRNAs have a potential clinical utility in reducing metastatic potential in ES and improving chemotherapy sensitisation. For example, enforced miR-34a expression in ES cell lines made the cells less proliferative and sensitised them to vincristine and doxorubicin [
      • Nakatani F.
      • et al.
      miR-34a predicts survival of Ewing's sarcoma patients and directly influences cell chemo-sensitivity and malignancy.
      ]. The re-introduction of miR-708 and/or use of transcriptional cofactor, EYA3, inhibitors could re-sensitise ES cells to chemotherapy [
      • Robin T.P.
      • et al.
      EWS/FLI1 regulates EYA3 in Ewing sarcoma via modulation of miRNA-708, resulting in increased cell survival and chemoresistance.
      ]. Finally, miR-130b has been shown to increase ES metastatic potential in vivo, and thus miR-130b and its mRNA targets represent novel approaches for preventing metastasis [
      • Satterfield L.
      • et al.
      miR-130b directly targets ARHGAP1 to drive activation of a metastatic CDC42-PAK1-AP1 positive feedback loop in Ewing sarcoma.
      ].
      Wilms tumour (WT), also known as nephroblastoma, is the most common paediatric renal malignancy [
      • Liu Z.
      • et al.
      miR-140-5p could suppress tumor proliferation and progression by targeting TGFBRI/SMAD2/3 and IGF-1R/AKT signaling pathways in Wilms' tumor.
      ], in which the lin28/let-7 pathway has been implicated in tumorigenesis [
      • Urbach A.
      • et al.
      Lin28 sustains early renal progenitors and induces Wilms tumor.
      ]. Furthermore, the overexpression of miR-195 inhibited WT survival in vitro, the effect of which was restored by LINC00473, a long ncRNA (lncRNA), implicating LINC00473 as an oncogene [
      • Zhu S.
      • et al.
      LINC00473 antagonizes the tumour suppressor miR-195 to mediate the pathogenesis of Wilms tumour via IKKalpha.
      ]. Both miR-140-5p and miR-185 have been identified as having TSG functions in WT [
      • Liu Z.
      • et al.
      miR-140-5p could suppress tumor proliferation and progression by targeting TGFBRI/SMAD2/3 and IGF-1R/AKT signaling pathways in Wilms' tumor.
      ,
      • Imam J.S.
      • et al.
      MicroRNA-185 suppresses tumor growth and progression by targeting the Six1 oncogene in human cancers.
      ]. Conversely, miR-483-3p and miR-483-5p, intragenic miRNAs located within intron 2 of the IGF2 gene, are overexpressed in WT and appear to have anti-apoptotic roles [
      • Veronese A.
      • et al.
      Oncogenic role of miR-483-3p at the IGF2/483 locus.
      ,
      • Liu M.
      • et al.
      The IGF2 intronic miR-483 selectively enhances transcription from IGF2 fetal promoters and enhances tumorigenesis.
      ]. Interestingly, the IGF2/miR-483 locus is located at 11p15.5, a common region of genetic and epigenetic abnormality in WT [
      • Veronese A.
      • et al.
      Oncogenic role of miR-483-3p at the IGF2/483 locus.
      ]. At present, there have been few studies systematically examining miRNA expression in WT and linking such findings with clinicopathological data and outcome. However, such work will facilitate the identification of miRNA/mRNA networks that drive tumorigenesis.
      Malignant GCTs are clinically complex and heterogeneous yet share functionally important molecular abnormalities. In malignant GCTs, regardless of histological subtype, site, or patient age, LIN28 is abundantly expressed with let-7 family downregulation [
      • Murray M.J.
      • et al.
      LIN28 Expression in malignant germ cell tumors downregulates let-7 and increases oncogene levels.
      ]. Thus, the LIN28/let-7 axis is a promising therapeutic target, offering strategies including protective small molecule targeting of pre-let-7 stem-loop binding motifs, induction of stem-loop binding protein KSRP, promoting maturation of miRNA subsets including let-7, use of let-7 mimics and inhibition of the terminal-uridyl-transferase (TUTase) ZCCHC11, which ultimately leads to pre-let-7 degradation [
      • Murray M.J.
      • et al.
      LIN28 Expression in malignant germ cell tumors downregulates let-7 and increases oncogene levels.
      ]. LIN28 and the C19MC oncogenic miRNA cluster have also been identified as potential therapeutic targets for certain CNS embryonal tumours, such as those with multilayered rosettes (ETMRs) [
      • Sin-Chan P.
      • et al.
      A C19MC-LIN28A-MYCN oncogenic circuit driven by hijacked super-enhancers is a distinct therapeutic vulnerability in ETMRs: a lethal brain tumor.
      ,
      • Neumann J.E.
      • et al.
      A mouse model for embryonal tumors with multilayered rosettes uncovers the therapeutic potential of Sonic-hedgehog inhibitors.
      ,
      • Lambo S.
      • et al.
      The molecular landscape of ETMR at diagnosis and relapse.
      ,
      • Spence T.
      • et al.
      CNS-PNETs with C19MC amplification and/or LIN28 expression comprise a distinct histogenetic diagnostic and therapeutic entity.
      ]. CNS GCTs can be histologically subdivided into germinomas and nongerminomatous malignant GCTs (NGMGCT) [
      • Hsieh T.H.
      • et al.
      Global DNA methylation analysis reveals miR-214-3p contributes to cisplatin resistance in pediatric intracranial nongerminomatous malignant germ cell tumors.
      ]. MiR-214-3p overexpression has been shown in vitro to reduce the expression of the pro-apoptotic protein BCL2-like 11 and induce cisplatin resistance, offering new mechanistic insight into underlying treatment resistance [
      • Hsieh T.H.
      • et al.
      Global DNA methylation analysis reveals miR-214-3p contributes to cisplatin resistance in pediatric intracranial nongerminomatous malignant germ cell tumors.
      ]. Paediatric low-grade gliomas (pLGGs) have the tendency to recur, and supratentorial lesions are difficult to resect [
      • Catanzaro G.
      • et al.
      The miR-139-5p regulates proliferation of supratentorial paediatric low-grade gliomas by targeting the PI3K/AKT/mTORC1 signalling.
      ]. Catanzaro et al. showed that miR-139-5p was significantly downregulated in supratentorial pLGGs, and this drives cell proliferation by derepressing PI3K/AKT signalling pathways [
      • Catanzaro G.
      • et al.
      The miR-139-5p regulates proliferation of supratentorial paediatric low-grade gliomas by targeting the PI3K/AKT/mTORC1 signalling.
      ]. MiR-125 is also downregulated in pLGG, and its overexpression leads to reduced cell growth and apoptosis induction [
      • Yuan M.
      • et al.
      MicroRNA (miR) 125b regulates cell growth and invasion in pediatric low grade glioma.
      ]. In glioblastoma (GBM), miR-1300 ectopic expression reduced GBM growth in an orthotopic model, and it was identified as a regulator of endomitosis [
      • Boissinot M.
      • et al.
      Profiling cytotoxic microRNAs in pediatric and adult glioblastoma cells by high-content screening, identification, and validation of miR-1300.
      ].
      Thus, many in vitro and in vivo studies across a broad range of paediatric solid tumours have elucidated critical roles for miRNAs in tumorigenesis and highlighted many as possible therapeutic targets via the use of inhibitors or mimics. Nevertheless, despite many preclinical experiments, few miRNA candidates have reached clinical development and trials, and further research is needed. One important exception is for malignant GCTs, where circulating miRNA biomarkers are embedded in clinical trials which include paediatric enrolment, e.g., AGCT1531 (ClinicalTrials.gov identifier: NCT03067181) and P3BEP (NCT02582697). There are many challenges that need addressing for the delivery of miRNA-based therapeutics, which include but are not limited to miRNA degradation by nucleases, poor target tissue delivery, immune reactions, unwanted off-target effects, and poor binding affinity for complementary sequences [
      • Annese T.
      • et al.
      microRNAs biogenesis, functions and role in tumor angiogenesis.
      ,
      • Shah V.
      • Shah J.
      Recent trends in targeting miRNAs for cancer therapy.
      ].

      9. Conclusion

      This review highlights key published observations regarding miRNA expression/profiles across a range of paediatric cancers observed in clinical practice. It will be important to extend and further the observations described here in order to translate these findings for clinical benefit. The physical properties of miRNAs, which make them resistant to degradation in body tissues and fluids, make them ideal candidates to explore as biomarkers of paediatric malignancies. MiRNA biomarkers are of particular relevance for paediatric malignancies, which typically have a low mutational burden compared with their adult counterparts; thus, the identification of DNA mutations and ctDNA as biomarkers are of more limited use. This review has demonstrated the opportunities that miRNAs hold for diagnosis, monitoring of treatment response, detection of early asymptomatic disease recurrence, prognostication, and as potential therapeutic targets of paediatric solid tumours. Further research is required to improve tissue-specific delivery of miRNA therapeutics and minimise off-target effects. It is not unreasonable to envisage the future use of miRNA mimics or antagonists, based on miRNA signatures, to deliver personalised, targeted therapy. A systematic investigation of the roles of dysregulated miRNAs across all paediatric malignancies, involving international collaboration where necessary, is warranted to afford substantial improvements in the management and outcomes of children affected by cancer.

      CRediT roles

      Nicholas Coleman, Matthew J. Murray - Conceptualisation; Karan R. Chadda, Ellen E. Blakey, Matthew J. Murray - Data curation; Karan R. Chadda, Ellen E. Blakey, Nicholas Coleman, Matthew J. Murray - Formal analysis; Karan R. Chadda, Ellen E. Blakey, Nicholas Coleman, Matthew J. Murray - Investigation and methodology; Nicholas Coleman, Matthew J. Murray - Supervision; Karan R. Chadda, Matthew J. Murray - Roles/Writing – original draft; Karan R. Chadda, Ellen E. Blakey, Nicholas Coleman, Matthew J. Murray - Writing – review & editing.

      Conflict of interest statement

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Acknowledgements/Funding

      We thank Dr Bob Philips, Senior Clinical Academic and Honorary Consultant Paediatric Oncologist, Centre for Reviews and Dissemination, University of York, York, UK, for invaluable advice regarding the review format. We apologise to those whose work we could not include due to space and reference constraints. The authors acknowledge grant funding the St. Baldrick’s Foundation (reference 358099) and the NIHR Cambridge Biomedical Research Centre (BRC-1215-20014∗). We are grateful for support from the Max Williamson Fund and from Christiane and Alan Hodson, in memory of their daughter Olivia.

      References

        • Kapranov P.
        • et al.
        RNA maps reveal new RNA classes and a possible function for pervasive transcription.
        Science. 2007; 316: 1484-1488
        • He L.
        • Hannon G.J.
        MicroRNAs: small RNAs with a big role in gene regulation.
        Nat Rev Genet. 2004; 5: 522-531
        • Lee R.C.
        • Feinbaum R.L.
        • Ambros V.
        The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14.
        Cell. 1993; 75: 843-854
        • Kloosterman W.P.
        • et al.
        In situ detection of miRNAs in animal embryos using LNA-modified oligonucleotide probes.
        Nat Methods. 2006; 3: 27-29
        • Condrat C.E.
        • et al.
        miRNAs as biomarkers in disease: latest findings regarding their role in diagnosis and prognosis.
        Cells. 2020; 9
        • Alexandrov L.B.
        • et al.
        Signatures of mutational processes in human cancer.
        Nature. 2013; 500: 415-421
        • O'Brien J.
        • et al.
        Overview of MicroRNA biogenesis, mechanisms of actions, and circulation.
        Front Endocrinol (Lausanne). 2018; 9: 402
        • Denli A.M.
        • et al.
        Processing of primary microRNAs by the Microprocessor complex.
        Nature. 2004; 432: 231-235
        • Ha M.
        • Kim V.N.
        Regulation of microRNA biogenesis.
        Nat Rev Mol Cell Biol. 2014; 15: 509-524
        • Yoda M.
        • et al.
        ATP-dependent human RISC assembly pathways.
        Nat Struct Mol Biol. 2010; 17: 17-23
        • Krol J.
        • Loedige I.
        • Filipowicz W.
        The widespread regulation of microRNA biogenesis, function and decay.
        Nat Rev Genet. 2010; 11: 597-610
        • Lewis B.P.
        • Burge C.B.
        • Bartel D.P.
        Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets.
        Cell. 2005; 120: 15-20
        • Nilsen T.W.
        Mechanisms of microRNA-mediated gene regulation in animal cells.
        Trends Genet. 2007; 23: 243-249
        • Baek D.
        • et al.
        The impact of microRNAs on protein output.
        Nature. 2008; 455: 64-71
        • Guo H.
        • et al.
        Mammalian microRNAs predominantly act to decrease target mRNA levels.
        Nature. 2010; 466: 835-840
        • Lim L.P.
        • et al.
        Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs.
        Nature. 2005; 433: 769-773
        • Farh K.K.
        • et al.
        The widespread impact of mammalian MicroRNAs on mRNA repression and evolution.
        Science. 2005; 310: 1817-1821
        • Calin G.A.
        • Croce C.M.
        MicroRNA signatures in human cancers.
        Nat Rev Cancer. 2006; 6: 857-866
        • Cho W.C.
        OncomiRs: the discovery and progress of microRNAs in cancers.
        Mol Cancer. 2007; 6: 60
        • Esquela-Kerscher A.
        • Slack F.J.
        Oncomirs - microRNAs with a role in cancer.
        Nat Rev Cancer. 2006; 6: 259-269
        • Farazi T.A.
        • et al.
        miRNAs in human cancer.
        J Pathol. 2011; 223: 102-115
        • Kent O.A.
        • Mendell J.T.
        A small piece in the cancer puzzle: microRNAs as tumor suppressors and oncogenes.
        Oncogene. 2006; 25: 6188-6196
        • Zhang B.
        • et al.
        microRNAs as oncogenes and tumor suppressors.
        Dev Biol. 2007; 302: 1-12
        • Calin G.A.
        • et al.
        Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia.
        Proc Natl Acad Sci U S A. 2002; 99: 15524-15529
        • Cimmino A.
        • et al.
        miR-15 and miR-16 induce apoptosis by targeting BCL2.
        Proc Natl Acad Sci U S A. 2005; 102: 13944-13949
        • Tagawa H.
        • Seto M.
        A microRNA cluster as a target of genomic amplification in malignant lymphoma.
        Leukemia. 2005; 19: 2013-2016
        • Mishra S.
        • Yadav T.
        • Rani V.
        Exploring miRNA based approaches in cancer diagnostics and therapeutics.
        Crit Rev Oncol Hematol. 2016; 98: 12-23
        • Calin G.A.
        • Croce C.M.
        MicroRNAs and chromosomal abnormalities in cancer cells.
        Oncogene. 2006; 25: 6202-6210
        • Calin G.A.
        • et al.
        Human microRNA genes are frequently located at fragile sites and genomic regions involved in cancers.
        Proc Natl Acad Sci U S A. 2004; 101: 2999-3004
        • Fabbri M.
        • Calin G.A.
        Epigenetics and miRNAs in human cancer.
        Adv Genet. 2010; 70: 87-99
        • Stumpel D.J.
        • et al.
        Hypermethylation of specific microRNA genes in MLL-rearranged infant acute lymphoblastic leukemia: major matters at a micro scale.
        Leukemia. 2010;
        • Li X.
        • et al.
        Gene silencing of MIR22 in acute lymphoblastic leukaemia involves histone modifications independent of promoter DNA methylation.
        Br J Haematol. 2010; 148: 69-79
        • Saito Y.
        • et al.
        Specific activation of microRNA-127 with downregulation of the proto-oncogene BCL6 by chromatin-modifying drugs in human cancer cells.
        Cancer Cell. 2006; 9: 435-443
        • Peng Y.
        • Croce C.M.
        The role of MicroRNAs in human cancer.
        Signal Transduct Targeted Ther. 2016; 115004
        • Arrate M.P.
        • et al.
        MicroRNA biogenesis is required for Myc-induced B-cell lymphoma development and survival.
        Cancer Res. 2010; 70: 6083-6092
        • Kumar M.S.
        • et al.
        Impaired microRNA processing enhances cellular transformation and tumorigenesis.
        Nat Genet. 2007; 39: 673-677
        • Thomson J.M.
        • et al.
        Extensive post-transcriptional regulation of microRNAs and its implications for cancer.
        Genes Dev. 2006; 20: 2202-2207
        • Kawahara Y.
        • et al.
        Redirection of silencing targets by adenosine-to-inosine editing of miRNAs.
        Science. 2007; 315: 1137-1140
        • Muralidhar B.
        • et al.
        Global microRNA profiles in cervical squamous cell carcinoma depend on Drosha expression levels.
        J Pathol. 2007; 212: 368-377
        • Muralidhar B.
        • et al.
        Functional evidence that Drosha overexpression in cervical squamous cell carcinoma affects cell phenotype and microRNA profiles.
        J Pathol. 2011; 224: 496-507
        • Trabucchi M.
        • et al.
        The RNA-binding protein KSRP promotes the biogenesis of a subset of microRNAs.
        Nature. 2009; 459: 1010-1014
        • Viswanathan S.R.
        • Daley G.Q.
        • Gregory R.I.
        Selective blockade of microRNA processing by Lin28.
        Science. 2008; 320: 97-100
        • Johnson C.D.
        • et al.
        The let-7 microRNA represses cell proliferation pathways in human cells.
        Cancer Res. 2007; 67: 7713-7722
        • Viswanathan S.R.
        • Daley G.Q.
        • Lin28
        A microRNA regulator with a macro role.
        Cell. 2010; 140: 445-449
        • Viswanathan S.R.
        • et al.
        Lin28 promotes transformation and is associated with advanced human malignancies.
        Nat Genet. 2009; 41: 843-848
        • Sandberg R.
        • et al.
        Proliferating cells express mRNAs with shortened 3' untranslated regions and fewer microRNA target sites.
        Science. 2008; 320: 1643-1647
        • Mayr C.
        • Bartel D.P.
        Widespread shortening of 3'UTRs by alternative cleavage and polyadenylation activates oncogenes in cancer cells.
        Cell. 2009; 138: 673-684
        • Godshalk S.E.
        • et al.
        A Variant in a MicroRNA complementary site in the 3' UTR of the KIT oncogene increases risk of acral melanoma.
        Oncogene. 2011; 30: 1542-1550
        • Zhang W.
        • et al.
        A let-7 microRNA-binding site polymorphism in 3'-untranslated region of KRAS gene predicts response in wild-type KRAS patients with metastatic colorectal cancer treated with cetuximab monotherapy.
        Ann Oncol. 2010; (In Press)
        • Kedde M.
        • et al.
        RNA-binding protein Dnd1 inhibits microRNA access to target mRNA.
        Cell. 2007; 131: 1273-1286
        • Murray M.J.
        • et al.
        "Future-proofing" blood processing for measurement of circulating miRNAs in samples from biobanks and prospective clinical trials.
        Cancer Epidemiol Biomarkers Prev. 2018; 27: 208-218
        • Caby M.P.
        • et al.
        Exosomal-like vesicles are present in human blood plasma.
        Int Immunol. 2005; 17: 879-887
        • Valadi H.
        • et al.
        Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells.
        Nat Cell Biol. 2007; 9: 654-659
        • Chen X.
        • et al.
        Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and other diseases.
        Cell Res. 2008; 18: 997-1006
        • Mitchell P.S.
        • et al.
        Circulating microRNAs as stable blood-based markers for cancer detection.
        Proc Natl Acad Sci U S A. 2008; 105: 10513-10518
        • Rupaimoole R.
        • et al.
        miRNA deregulation in cancer cells and the tumor microenvironment.
        Cancer Discov. 2016; 6: 235-246
        • Seijo L.M.
        • et al.
        Biomarkers in lung cancer screening: achievements, promises, and challenges.
        J Thorac Oncol. 2019; 14: 343-357
        • Weber J.A.
        • et al.
        The microRNA spectrum in 12 body fluids.
        Clin Chem. 2010; 56: 1733-1741
        • Hanke M.
        • et al.
        A robust methodology to study urine microRNA as tumor marker: microRNA-126 and microRNA-182 are related to urinary bladder cancer.
        Urol Oncol. 2010; 28: 655-661
        • Kurahashi R.
        • et al.
        MicroRNA-204-5p: a novel candidate urinary biomarker of Xp11.2 translocation renal cell carcinoma.
        Cancer Sci. 2019; 110: 1897-1908
        • Baraniskin A.
        • et al.
        Identification of microRNAs in the cerebrospinal fluid as biomarker for the diagnosis of glioma.
        Neuro Oncol. 2012; 14: 29-33
        • Murray M.J.
        • et al.
        A pipeline to quantify serum and cerebrospinal fluid microRNAs for diagnosis and detection of relapse in paediatric malignant germ-cell tumours.
        Br J Cancer. 2016; 114: 151-162
        • Murray M.J.
        • et al.
        Solid tumors of childhood display specific serum microRNA profiles.
        Cancer Epidemiol Biomarkers Prev. 2015; 24: 350-360
        • Brase J.C.
        • et al.
        Serum microRNAs as non-invasive biomarkers for cancer.
        Mol Cancer. 2010; 9: 306
        • Northcott P.A.
        • et al.
        The miR-17/92 polycistron is up-regulated in sonic hedgehog-driven medulloblastomas and induced by N-myc in sonic hedgehog-treated cerebellar neural precursors.
        Cancer Res. 2009; 69: 3249-3255
        • Uziel T.
        • et al.
        The miR-17∼92 cluster collaborates with the Sonic Hedgehog pathway in medulloblastoma.
        Proc Natl Acad Sci U S A. 2009; 106: 2812-2817
        • Gruszka R.
        • et al.
        mRNA and miRNA expression analyses of the MYC/E2F/miR-17-92 network in the most common pediatric brain tumors.
        Int J Mol Sci. 2021; 22
        • Thor T.
        • et al.
        MiR-34a deficiency accelerates medulloblastoma formation in vivo.
        Int J Cancer. 2015; 136: 2293-2303
        • Braoudaki M.
        • et al.
        Microrna expression signatures predict patient progression and disease outcome in pediatric embryonal central nervous system neoplasms.
        J Hematol Oncol. 2014; 7: 96
        • Annese T.
        • et al.
        microRNAs biogenesis, functions and role in tumor angiogenesis.
        Front Oncol. 2020; 10581007
        • Castro-Magdonel B.E.
        • et al.
        Circulating miRNome detection analysis reveals 537 miRNAS in plasma, 625 in extracellular vesicles and a discriminant plasma signature of 19 miRNAs in children with retinoblastoma from which 14 are also detected in corresponding primary tumors.
        PLoS One. 2020; 15: e0231394
        • Carvalho I.N.
        • et al.
        A polymorphism in mir-34b/c as a potential biomarker for early onset of hereditary retinoblastoma.