US20170275703A1

US20170275703A1 - Treatment and detection of melanoma

Info

Publication number: US20170275703A1
Application number: US15/506,545
Authority: US
Inventors: Mitchell Stark
Original assignee: Queensland Institute of Medical Research QIMR
Current assignee: QIMR Berghofer Medical Research Institute
Priority date: 2014-08-25
Filing date: 2015-08-24
Publication date: 2017-09-28
Also published as: EP3186394B1; WO2016029260A1; EP3186394A1; AU2015309684A1; US20220136061A1; AU2022202220A1; EP3186394A4

Abstract

Provided are methods for determining or predicting the diagnosis, prognosis, treatment and therapeutic efficacy of melanoma in a subject, which include evaluating the expression level of one or more miRNA biomarkers including miRNA-4487, miRNA-4706, miRNA-4731 and fragments or variants thereof as an indication of whether the subject may have, or be predisposed to, melanoma.

Description

TECHNICAL FIELD

THIS INVENTION relates to the prognosis, diagnosis and/or treatment of cancers. More particularly, this invention relates to determining expression levels of one or more micro-RNAs correlated with melanoma in a biological sample from a subject.

BACKGROUND

Melanomas are among the most commonly occurring cancers. Incidence rates in Australia²and the USA³reveal more than 11,400 (in 2010) and 76,000 (estimated in 2014) new cases respectively, with more than 1,500 (in 2011) and 9,700 (estimated in 2014) respectively, dying annually of advanced disease. Current staging criteria are based on melanoma progression from an early stage lesion, confined to the epidermis (stage 0) followed by a series of early stages of local invasion (I and II) before moving to the regional lymph nodes (stage III) and finally metastasizing to distal sites (stage IV). The overall 5-year survival for melanoma is 91%. However, if distal metastasis occurs (AJCC 7^thed stage IV), cure rates are <15%¹. Hence, melanoma must be detected in earlier stages to maximize the chances of patient survival. Therefore, the ability to rapidly identify predictors of melanoma progression in patients would therefore be a significant clinical tool.
For many years, melanoma progression markers have been studied intensively with varying levels of success. Serum lactate dehydrogenase (LDH) levels have been reported to be elevated in melanoma patients and have been integrated into current staging regimens¹. Despite the high sensitivity and specificity of serum LDH levels in stage II patients (95% and 83% respectively), performance of this marker has been found to be reduced as disease progresses (stage III, 57% and 87%, respectively)^4-8. In an earlier study of stage IV patients, sensitivity and specificity was increased to 79% and 92% respectively⁹. S-100B, a calcium binding protein, is another serological marker found to be elevated in stage III and IV melanoma patients^{10, 11}. The proportion of patients with elevated S-100B levels has been measured in all stages of disease which vary dependent somewhat upon stage: 0-9% in stage I/II, 5-98% in stage III, and 40-100% in stage IV (reviewed in ref 12). However, serum S 100B is not routinely used in the clinic, highlighting that the current serological methods of progression detection, whilst relatively specific, are inadequate due to somewhat variable sensitivity to detect all stages of disease. To date, there are no biomarkers that are sensitive or specific enough to be beneficial for early diagnosis of melanoma. The use of a blood test (‘circulating’ biomarkers) for early detection of melanoma, prior to distant metastases, could improve treatment and outcomes for melanoma patients.
In order for a circulating biomarker to be effective, it needs to be not only highly sensitive and specific; it also needs to be highly stable and resistant to degradation. In recent years, circulating microRNAs (miRNAs) have been studied intensively for their utility as biomarkers in a wide range of malignancies and disorders^13-15. miRNAs are small (20-22 nt) non-coding RNAs which function primarily is to regulate gene expression in a sequence specific manner and to act as positive or negative regulators of protein levels in the cell. More recently, tumour cells have been shown experimentally to release miRNAs directly into the circulation¹⁶contained primarily in micro-vesicles or exosomes (extracellular vesicles), or bound to AGO2 which is part of the miRNA-mediated silencing complex^13-15. Due to the ‘encapsulation’ of these miRNAs in serum or plasma, they are highly resistant to degradation from RNases (highly concentrated in the blood) thus their potential usefulness as a ‘biomarker’ is relatively high. Not only are they resistant to degradation, they are also inherently stable with many studies highlighting their resistance to extended storage conditions and changes in pH and temperature^13-15.
To date, the identification of circulating melanoma-specific miRNAs has been limited to a small number of studies^{17, 18}. Most recently Friedman et al. (2012) screened 355 miRNAs in sera from 80 melanoma patients using a previously characterised panel of serum-expressed miRNAs¹⁸. The authors found that a five-miRNA signature was able to classify the patients into high and low recurrence risk. Although this study highlighted the potential use of this subset of miRNAs for diagnostic purposes, it only assessed ˜17% of all known miRNAs (there are now >2000 miRNAs in miRBase¹⁹). In addition, the panel of miRNAs selected for validation was not assessed for its specificity to melanoma. Whilst this panel addresses an important question, it is limited to prior diagnosis of melanoma and could not be used to identify the presence of unknown primaries or unidentified metastases. Accordingly, there remains a need for biomarkers, such as miRNA biomarkers, that are useful in the diagnosis, prognosis, therapeutic response monitoring and/or disease monitoring in melanoma patients. Specifically, the use of such biomarkers may address the outstanding clinical needs for more sensitive tools to screen for recurrence following treatment and prognostic tools to guide treatment decisions for patients with advanced disease.

SUMMARY

MicroRNAs represent an important class of biomarkers that provide opportunities for clinical translation. The invention is broadly directed to a method of diagnosis, prognosis, therapeutic response monitoring and/or disease response monitoring of melanoma.
More particularly, the inventors have discovered specific miRNA biomarkers that have proved to be useful in the diagnosis, prognosis and/or disease monitoring of melanoma. Subsequently, methods have been developed to diagnose and/or monitor disease progression in subjects with melanoma as well as to provide an indication of disease prognosis. Furthermore, the inventors have discovered that particular miRNA biomarkers may serve as a prognostic marker with respect to treatment response.
In a first aspect, the invention provides a method of determining whether or not a subject has melanoma, including:
determining an expression level of one or more miRNA biomarkers in a biological sample from a subject, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, or a fragment or variant thereof, and wherein melanoma is detected if said expression level of one or more miRNA biomarkers, is altered or modulated in the biological sample.
In a second aspect, the invention provides a method of determining the prognosis of a subject with melanoma, including:
determining an expression level of one or more miRNA biomarkers in a biological sample obtained from the subject, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, or a fragment or variant thereof, to thereby evaluate the prognosis of melanoma in the subject.
Suitably, if the expression level of said one or more miRNA biomarkers is altered or modulated in the biological sample, the prognosis may be negative or positive.
In one embodiment, the prognosis is used, at least in part, to determine whether the subject would benefit from treatment of melanoma.
In one embodiment, the prognosis is used, at least in part, to develop a treatment strategy for the subject.
In one embodiment, the prognosis is used, at least in part, to determine disease progression in the subject.
In one embodiment, the prognosis is defined as an estimated time of survival.
In one embodiment, the method of this aspect further includes determining suitability of the subject for treatment based, at least in part, on the prognosis.
In one embodiment of the first and second aspects, the method further comprises determining a disease stage and/or grade for melanoma based on the expression level of the one or more miRNA biomarkers.
In one embodiment of the first and second aspects, the expression level of the one or more miRNA biomarkers is determined before, during and/or after treatment.
In a third aspect, the invention provides a method of treating melanoma in a subject including;
determining an expression level of one or more miRNA biomarkers in a biological sample from the subject, before, during and/or after treatment of melanoma, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-473, or a fragment or variant thereof, and based on the determination made, initiating, continuing, modifying or discontinuing a treatment of melanoma.
In a fourth aspect, the invention provides a method of evaluating treatment efficacy of melanoma in a subject including:
determining an expression level of one or more miRNA biomarkers in a biological sample from the subject before, during and/or after treatment, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, or a fragment or variant thereof; and
determining whether or not the treatment is efficacious according to whether said expression level of one or more miRNA biomarkers is altered or modulated in the subject's biological sample.
In one embodiment of the third and fourth aspects, the method further comprises selecting a treatment for melanoma based on the expression level of the miRNA biomarkers.
In particular embodiments, the method of the first, second, third and fourth aspects further comprises measuring an expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p, or a fragment or variant thereof.
In particular embodiments of the method of the first, second, third and fourth aspects, an expression level of a protein or nucleic acid biomarker may also be determined. Suitably, the protein or nucleic acid biomarker is selected from the group consisting of serum lactate dehydrogenase (LDH), S-100B, tyrosinase, melanoma-inhibiting activity (MIA), tumour-associated antigen 90 immune complex (TA90IC), C-reactive protein, galectin-3, melanoma associated antigen, interleukin-8, matrix metalloprotease(MMP)-1, MMP-3 and YKL-40.
In particular embodiments of the method of the first, second, third and fourth aspects, the biological sample comprises tissue, blood, serum, plasma or cerebrospinal fluid. Typically, the miRNAs described herein are obtainable from a non-cellular source. Accordingly, the biological sample is, comprises, or is obtained from a non-cellular source. To this end, the biological sample may be serum, plasma, or cerebrospinal fluid, although without limitation thereto.
Suitably, the subject referred to herein is a mammal. Preferably, the subject is a human.
As used herein, except where the context requires otherwise, the term “comprise” and variations of the term, such as “comprising”, “comprises” and “comprised”, are not intended to exclude further elements, components, integers or steps but may include one or more unstated further elements, components, integers or steps.
It will be appreciated that the indefinite articles “a” and “an” are not to be read as singular indefinite articles or as otherwise excluding more than one or more than a single subject to which the indefinite article refers. For example, “a” cell includes one cell, one or more cells and a plurality of cells.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1. ‘Melanoma-specific’ miRNAs were first identified in a ‘Discovery set’ of 233 miRNAs. The ‘MELmiR-18’ panel was measured firstly in an independent cohort of FFPE melanoma tissues (‘Validation Cohort 1’). These data are summarised if Table 4. The ‘MELmiR-18’ panel was then measured (‘Validation Cohorts 2 and 3’) in two cohorts of serum derived from controls in comparison to AJCC staged melanoma patients (stages I-IV). These data are summarised in Table 2 and FIGS. 6 and 7. The ‘MELmiR-7’ panel was identified and carried forward in subsequent analysis as each miRNA member achieved high significance (P<0.0001) and an AUC score >0.70 in the ‘Validation Cohort 2 and 3’ comparison.

FIG. 2. The diagnostic decision tree summarises the utility of the ‘MELmiR-7’ panel in a diagnostic setting. The ‘MELmiR-7’ panel achieves very high sensitivity and specificity when a diagnostic score of 2-7 is attained (see Table 2 for further details).

FIG. 3. Predictions derived from Binary Logistic Regression (BLR) analysis are summarised. This represents the versatility of the ‘MELmiR-7’ panel for not only diagnosis of melanoma, but also for its predictive power for progression and recurrence of disease.

FIG. 4. Expression levels of miR-211-5p were assigned as high or normal expression based upon the comparison of stage IV melanoma with controls (FIG. 1). Using ROC analysis, high expression was determined if the median-normalised CT value was >85% sensitive. The Kaplan-Meier curve shows clear separation between high and normal miR-211-5p expressers median survival being 1.1 years vs. 3.5 years (HR 3.2, 95% CI 2.02-5.14, p<0.0001)

FIG. 5. Graphical representation of the data presented in Table 4 for stage III and IV FFPE tissues following a Mann-Whitney U test (unpaired). ns=no significance; *=P 0.05-0.01; **=P<0.01; ***=P<0.001; ****=P<0.0001.

FIG. 6. Graphical representation of the data presented in Table 5 for the MELmir-8 panel which were those that showed high significance (P<0.0001) between stage IV vs controls following a Mann-Whitney U test (unpaired). ns=no significance; *=P 0.05-0.01; **=P<0.01; ***=P<0.001; ****=P<0.0001.

FIG. 7. ROC curves derived from the data presented in Table 5 for the MELmir-8 panel which were those that showed high significance (P<0.0001) between stage IV vs controls. All members of the ‘MELmir-7’ panel have AUC scores>0.7. Using this cut-off, miR-204 is excluded.

FIG. 8. Unsupervised hierarchical cluster tree (Euclidean similarity with average linkage) of all miRNA genes and samples (Tables 7 and 8) Distinct separation can be observed for most melanoma (cutaneous) samples (brown) compared with other solid malignancies (green). Controls (melanocytes and melanoblasts) (red), uveal melanoma (UMM) (pink), melanoma patient-derived serum (blue) and nevocyte (grey) are also present in the tree.

FIG. 9. The top 3 upregulated miRNAs were validated using qRT-PCR with all samples present on the microarray along with an extended cohort of melanoma cell lines (Table 10). For comparative purposes the array data was also plotted to highlight the high correlation between the array and qRT-PCR.

FIG. 10. The nucleic acid sequence and observed sequence variation of miRNA-4487.

FIG. 11. The nucleic acid sequence and observed sequence variation of miRNA-4706.

FIG. 12. The nucleic acid sequence and observed sequence variation of miRNA-4731.

FIG. 13. The nucleic acid sequence and observed sequence variation of miRNA-16.

FIG. 14. The nucleic acid sequence and observed sequence variation of miRNA-211.

FIG. 15. The nucleic acid sequence and observed sequence variation of miRNA-509-3p and miRNA-509-5p.

BRIEF DESCRIPTION OF THE SEQUENCES

SEQ ID NO: 1=nucleic acid sequence miRNA-4487 of FIG. 10 (mature sequence of miRNA-4487)
SEQ ID NO: 2=nucleic acid sequence miRNA-4706 of FIG. 11 (mature sequence of miRNA-4706)
SEQ ID NO: 3=nucleic acid sequence miRNA-4731 of FIG. 12 (mature sequence of miRNA-4731-5p)
SEQ ID NO: 4=nucleic acid sequence miRNA-16 of FIG. 13 (mature sequence of miRNA-16-5p)
SEQ ID NO: 5=nucleic acid sequence miRNA-211 of FIG. 14 (mature sequence of miRNA-211-5p)
SEQ ID NO: 6=nucleic acid sequence miRNA-509-3p of FIG. 15 (mature sequence of miRNA-509-3p)
SEQ ID NO: 7=nucleic acid sequence miRNA-509-5p of FIG. 15 (mature sequence of miRNA-509-5p)

DETAILED DESCRIPTION

The present invention is predicated, at least in part, on the surprising discovery that a novel panel of micro RNAs (miRNAs or miRs) may provide a serum-based prognostic and/or diagnostic biomarker for melanoma. In this regard, the inventors demonstrate herein that the panel may be used to monitor disease progression in melanoma patients, particularly in those patients with metastatic disease. Accordingly, the miRNAs disclosed herein may also have utility in methods of treating and/or evaluating treatment efficacy in melanoma patients.
In an aspect, the invention provides a method of determining whether or not a subject has melanoma, including:
determining an expression level of one or more miRNA biomarkers in a biological sample from a subject, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, and wherein melanoma is detected if said expression level of one or more miRNA biomarkers, is altered or modulated in the biological sample.
As generally used herein, the terms “cancer”, “tumour”, “malignant” and “malignancy” refer to diseases or conditions, or to cells or tissues associated with the diseases or conditions, characterized by aberrant or abnormal cell proliferation, differentiation and/or migration often accompanied by an aberrant or abnormal molecular phenotype that includes one or more genetic mutations or other genetic changes associated with oncogenesis, expression of tumour markers, loss of tumour suppressor expression or activity and/or aberrant or abnormal cell surface marker expression.
“Melanoma”, as used herein, refers to the malignant proliferation of melanocytes. The term “melanoma” is used, unless stated otherwise, in the broadest sense and refers to all stages and all forms of the cancer, such as malignant, metastatic (regional or distant), cutaneous, uveal and recurrent melanoma. As would be understood by the skilled artisan, this includes those subjects with melanoma that may be asymptomatic and/or a diagnosis of melanoma has not been provided or is ambiguous and yet the subject does indeed have melanoma.
The term “determining” includes any form of measurement, and includes determining if an element is present or not. As used herein, the terms “determining”, “measuring”, “evaluating”, “assessing” and “assaying” are used interchangeably and include quantitative and qualitative determinations. Determining may be relative or absolute. “Determining the presence of” includes determining the amount of something present (e.g., an miRNA and/or protein biomarker), and/or determining whether it is present or absent.
As would be understood by the skilled person, the expression level of the one or more miRNA biomarkers is deemed to be “altered” or “modulated” when the amount or expression level of the respective miRNA biomarker is increased or up regulated or decreased or down regulated, as defined herein.
In one embodiment, melanoma is detected if the one or more miRNA biomarkers are at a reduced level, down regulated or absent in the biological sample. In an alternative embodiment, melanoma is detected if the one or more miRNA biomarkers are at an increased level, up regulated or present in the biological sample.
By “enhanced”, “increased” or “up regulated” as used herein to describe the expression level of miRNA, protein and/or nucleic acid biomarkers, refers to the increase in and/or amount or level of one or more miRNA, protein and/or nucleic acid biomarkers, including variants, in a biological sample when compared to a control or reference sample or a further biological sample from a subject. The expression level of a biomarker may be relative or absolute. In some embodiments, the expression level of the one or more miRNA, protein and/or nucleic acid biomarkers is increased if its level of expression is more than about 0.5%, 1%, 2%, 3%, 4%, 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, 150%, 200%, 300%, 400% or at least about 500% higher than the level of expression of the corresponding miRNA, protein and/or nucleic acid biomarkers in a control sample or further biological sample from a subject.
The terms, “reduced” and “down regulated”, as used herein to describe the expression level of miRNA, protein and/or nucleic acid biomarkers, refer to a reduction in and/or amount or level of one or more miRNA, protein and/or nucleic acid biomarkers, including variants, in a biological sample when compared to a control or reference sample or further biological sample from a subject. The expression level of a biomarker may be relative or absolute. In some embodiments, the expression level of one or more miRNA, protein and/or nucleic acid biomarkers is reduced or down regulated if its level of expression is more than about 95%, 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20% or 10%, or even less than about 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.01%, 0.001% or 0.0001% of the level of expression of the corresponding miRNA, protein and/or nucleic acid biomarkers in a control sample or further biological sample from a subject.
The term “control sample” typically refers to a biological sample from a healthy or non-diseased individual not having melanoma. In one embodiment, the control sample may be from a subject known to be free of melanoma. Alternatively, the control sample may be from a subject in remission from melanoma. The control sample may be a pooled, average or an individual sample. An internal control is a marker from the same biological sample being tested.
Suitably, the expression level of any combination of the miRNA biomarkers miRNA-4487, miRNA-4706 and miRNA-4731 in a biological sample may be determined, including for example: miRNA-4487 and miRNA-4706; miRNA-4487 and miRNA-4731; miRNA-4706 and miRNA-4731; or miRNA-4487, miRNA-4706 and miRNA-4731.
In one embodiment the method of determining whether or not a subject has melanoma includes determining the expression level of a further miRNA biomarker in addition to miRNA-4487, miRNA-4706 and miRNA-4731 in a biological sample from the subject, wherein the miRNA biomarkers are selected from the following group: miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p, and wherein melanoma may be detected if at least one of the miRNA biomarkers is at an altered or modulated level or expression in the biological sample when compared to a control or reference sample. By way of example, an increase in the level or up regulation in the biological sample of a further biomarker, such as miRNA-16 and/or miRNA-211, may indicate the presence of melanoma in the subject. Further, a decrease in the level, down regulation or absence in the biological sample of a further biomarker, such as miRNA-509-3p and/or miRNA-509-5p, may indicate the presence of melanoma in the subject.
It will be appreciated that the methods of the invention include methods of determining the expression level of the one or more miRNA biomarkers alone or in combination with one or more protein and/or nucleic acid biomarkers which have been identified as being diagnostic for melanoma. In certain embodiments, the expression level of one or more protein and/or nucleic acid biomarkers may also be determined. In this regard, the one or more protein and/or nucleic acid biomarkers may include, for example, serum lactate dehydrogenase (LDH), S-100B, tyrosinase, VEGF, bFGF, ICAM-1, matrix metalloproteases (MMPs; e.g., MMP-1 and MMP-3), cytokines and/or their receptors (e.g., IL-6, IL-8, IL-10), HLA molecules, TuM2-PK, apoptosis markers (e.g., Fas/CD95) karyopherin-alpha2, MCMs (minichromosome maintenance proteins), melanoma-inhibiting activity (MIA), tumour-associated antigen 90 immune complex (TA90IC), C-reactive protein, galectin-3, melanoma associated antigens (e.g., MAGE proteins), YKL-40, geminin and PCNA, albeit without limitation thereto. Additional diagnostic and prognostic protein and nucleic acid biomarkers for melanoma have been described previously, as reviewed in DIAGNOSTIC AND PROGNOSTIC BIOMARKERS AND THERAPEUTIC TARGETS IN MELANOMA Ed. Murphy, M. J., (Humana Press, 2012), which is incorporated by reference herein.
The expression level of the one or more miRNA, protein and/or nucleic acid biomarkers may be determined by any method known in the art. By way of example, the expression level of miRNA biomarkers may be determined by hybridization-based techniques (e.g., Northern blots, in situ hybridization, RT-PCR, and microarrays), amplification-based techniques (e.g., real-time quantitative PCR; gold nanoparticle-initiated silver enhancement) and cloning-based techniques (e.g., miRAGE).
Suitably, when the expression level of the one or more miRNA biomarkers and/or one or more protein and/or nucleic acid biomarkers are determined, they can be derived from the same or different samples. For example, the expression level of the one or more miRNA biomarkers can be determined in a blood derived sample and the expression level of a protein biomarker can be determined in a tissue sample.
In particular embodiments, the biological sample comprises tissue, blood, serum, plasma or cerebrospinal fluid. Typically, the miRNAs described herein are obtainable from a non-cellular source. Accordingly, the biological sample is, comprises, or is obtained from a non-cellular source. To this end, the biological sample may be serum, plasma, or cerebrospinal fluid, although without limitation thereto.
In some embodiments, the method of determining whether or not a subject has melanoma may be performed in “high throughput” diagnostic tests or procedures such as performed by commercial pathology laboratories or in hospitals. Furthermore or alternatively, the method of the present aspect may be used to confirm a diagnosis of melanoma, including metastatic and recurrent melanoma, such as that initially detected by a different or alternative diagnostic test or procedure.
It would be further appreciated, that such methods of determining miRNA expression in the biological sample from a melanoma patient may have potential utility in characterising disease progression and/or severity of a given patient. Additionally, such methods may be used for selecting patients for particular treatments.
In one embodiment, the method determines whether the subject has metastatic melanoma.
As used herein, “metastasis” or “metastatic”, refers to the migration or transfer of malignant tumour cells, or neoplasms, via the circulatory or lymphatic systems or via natural body cavities, typically from the primary focus of tumour, cancer or a neoplasia to a distant site in the body, and the subsequent development of one or more secondary tumours or colonies thereof in the one or more new locations. “Metastases” refers to the secondary tumours or colonies formed as a result of metastasis and encompasses micro-metastases as well as regional and distant metastases.
In one embodiment, the method determines whether the subject has recurrent melanoma.
In another aspect, the invention provides a method of determining the prognosis of a subject with melanoma, including:
determining an expression level of one or more miRNA biomarkers in a biological sample obtained from the subject, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, to thereby evaluate the prognosis of melanoma in the subject.
Suitably, if the expression level of said one or more miRNA biomarkers, is altered or modulated in the biological sample, the prognosis may be negative or positive.
The terms “prognosis” and “prognostic” are used herein to include making a prognosis, which can provide for predicting a clinical outcome (with or without medical treatment), selecting an appropriate course of treatment (or whether treatment would be effective) and/or monitoring a current treatment and potentially changing the treatment. This may be at least partly based on determining expression levels of one or more miRNA biomarkers by the methods of the invention, which may be in combination with determining the expression levels of additional protein and/or other nucleic acid biomarkers. A prognosis may also include a prediction, forecast or anticipation of any lasting or permanent physical or psychological effects of melanoma suffered by the subject after the melanoma has been successfully treated or otherwise resolved. Furthermore, prognosis may include one or more of determining metastatic potential or occurrence, therapeutic responsiveness, implementing appropriate treatment regimes, determining the probability, likelihood or potential for melanoma recurrence after therapy and prediction of development of resistance to established therapies (e.g., chemotherapy). It would be appreciated that a positive prognosis typically refers to a beneficial clinical outcome or outlook, such as long-term survival without recurrence of the subject's melanoma, whereas a negative prognosis typically refers to a negative clinical outcome or outlook, such as cancer recurrence or progression.
“Resistance” as used herein refers to a diminished or failed response of an organism, disease, tissue or cell to the intended effectiveness of a treatment, such as a chemical or drug. Resistance to a treatment can be already present at diagnosis or the start of treatment (i.e., intrinsic resistance) or it can develop with or after treatment (i.e., acquired resistance).
As described herein, the expression levels of miRNA-4487, miRNA-4706 and miR-4731 are typically decreased or down regulated upon the initial development of melanoma in a subject. The expression levels of these miRNAs, however, may be shown to be altered or modulated (i.e., they may increase and/or decrease) with disease progression, such as progression from Stage I/II to Stage III or Stage IV melanoma and from Stage III to Stage IV melanoma, despite typically remaining decreased or down regulated when compared to a control sample or level of expression.
In this regard, an increase or up regulation of the level in the biological sample of miRNA-4487 in a subject with Stage I/II or Stage III melanoma, may indicate a positive prognosis for the subject, such as disease remission or regression, or alternatively it may indicate a negative prognosis for the subject, such as disease progression to Stage IV melanoma. Further, a decrease or down regulation of the level in the biological sample of miRNA-4487 in a subject with Stage I/II melanoma may indicate a negative prognosis for the subject, such as disease progression to Stage III melanoma.
An increase or up regulation of the level in the biological sample of miRNA-4706 in a subject with Stage I/II or Stage III melanoma, may indicate a positive prognosis for the subject, such as disease regression or remission, or alternatively it may indicate a negative prognosis for the subject, such as disease progression to Stage III or Stage IV melanoma.
An increase or up regulation of the level in the biological sample of miRNA-4731 in a subject with Stage I/II melanoma, may indicate a positive prognosis for the subject, such as disease regression or remission, or alternatively it may indicate a negative prognosis for the subject, such as disease progression to Stage III melanoma. Further, a decrease or down regulation of the level in the biological sample of miRNA-4731 in a subject with Stage III melanoma may indicate a negative prognosis for the subject, such as disease progression to Stage IV melanoma. Additionally, an increase or up regulation of the level in the biological sample of miRNA-4731 in a subject with Stage III or IV melanoma, may indicate a positive prognosis for the subject, such as disease regression or remission.
In one embodiment, the method further comprises measuring an expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p.
In this regard, an increase or up regulation of the level in the biological sample of an additional biomarker, such as miRNA-16 and/or miRNA-211, may indicate a negative prognosis for the subject. Further, a decrease or down regulation of the level in the biological sample of an additional biomarker, such as miRNA-509-3p and/or miRNA-509-5p, may indicate a negative prognosis for the subject. Additionally, a decrease or down regulation of the level in the biological sample of miRNA-16 in a subject with Stage III melanoma, may indicate a positive prognosis for the subject, such as disease regression or remission, or alternatively it may indicate a negative prognosis for the subject, such as disease progression to Stage IV melanoma.
In certain embodiments, the expression level of one or more protein and/or nucleic acid biomarkers, as hereinbefore described, may also be determined.
In particular embodiments, the biological sample comprises tissue, blood, serum, plasma or cerebrospinal fluid. Preferably, the biological sample comprises serum, plasma, or cerebrospinal fluid, although without limitation thereto.
In one embodiment, the method further comprises determining a disease stage and/or grade for the subject's melanoma based on the expression level of the one or more miRNA biomarkers. In this regard, the method may be used to determine whether the subject's melanoma has metastasized regionally, such as to lymph nodes (i.e., Stage III melanoma), and/or distantly to other parts of the subject's body (i.e., Stage IV melanoma).
In one embodiment, the expression level of the one or more miRNA biomarkers is determined before, during and/or after treatment.
In one embodiment, the prognosis is used, at least in part, to determine whether the subject would benefit from treatment of the melanoma.
In one embodiment, the prognosis is used, at least in part, to develop a treatment strategy for the subject.
In one embodiment, the prognosis is used, at least in part, to determine disease progression or recurrence in the subject.
In one embodiment, the prognosis is defined as an estimated time of survival.
In one embodiment, the method of this aspect further includes determining suitability of the subject for treatment based, at least in part, on the prognosis.
In a further aspect, the invention provides a method of treating melanoma in a subject including;
determining an expression level of one or more miRNA biomarkers in a biological sample from a subject, before, during and/or after treatment of melanoma, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706, miRNA-4731 and based on the determination made, initiating, continuing, modifying or discontinuing a treatment of melanoma.
As used herein, “treating”, “treat” or “treatment” refers to a therapeutic intervention, course of action or protocol that at least ameliorates a symptom of melanoma after the cancer and/or its symptoms have at least started to develop. As used herein, “preventing”, “prevent” or “prevention” refers to therapeutic intervention, course of action or protocol initiated prior to the onset of cancer and/or a symptom of cancer so as to prevent, inhibit or delay development or progression of the cancer or the symptom.
In one embodiment, the method further comprises measuring an expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p.
In certain embodiments, the expression level of one or more protein and/or nucleic acid biomarkers, as herein before described, may also be determined.
In particular embodiments, the biological sample comprises tissue, blood, serum, plasma or cerebrospinal fluid. Preferably, the biological sample comprises serum, plasma, or cerebrospinal fluid, although without limitation thereto.
In one embodiment, the method further comprises selecting a treatment for melanoma based on the expression level of the miRNA biomarkers.
It will be appreciated that the method of treating melanoma may include administration of one or more other therapeutic agents that facilitate melanoma treatment or prevention. By way of example only, these include: chemotherapeutic agents such as paclitaxel, doxorubicin, methotrexate, irinotecan, dacarbazine, temozolomide and cisplatin, although without limitation thereto; biotherapeutic agents such as anti-PD-1 antibodies (e.g., Nivolumab) and anti-CTLA4 antibodies (e.g., Ipilimumab), although without limitation thereto; and/or molecularly targeted agents such as MAPK pathway (i.e., RAS-RAF-MEK-ERK signalling) inhibitors and phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathway inhibitors, although without limitation thereto.
In one embodiment, the method further comprises selecting a treatment for melanoma based on the expression level of the miRNA biomarkers.
In yet another aspect, the invention provides a method of evaluating treatment efficacy of melanoma in a subject including;
determining an expression level of one or more miRNA biomarkers in a biological sample from the subject before, during and/or after treatment, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731; and
determining whether or not the treatment is efficacious according to whether said expression level of one or more miRNA biomarkers is altered or modulated in the subject's biological sample.
As noted earlier, the expression levels of miRNA-4487, miRNA-4706 and miRNA-4731, may increase and/or decrease with disease progression, despite typically remaining decreased or down regulated when compared to a control sample.
In this regard, an increase or up regulation of the level in the biological sample of miRNA-4487 in a subject with Stage I/II or Stage III melanoma undergoing treatment, may indicate disease remission or regression in the subject, and as such the treatment is efficacious, or alternatively it may indicate disease progression to Stage IV melanoma and the treatment is inefficacious. Further, a decrease or down regulation of the level in the biological sample of miRNA-4487 in a subject with Stage I/II melanoma undergoing treatment may indicate disease progression to Stage III melanoma and the treatment is inefficacious.
An increase or up regulation of the level in the biological sample of miRNA-4706 in a subject with Stage I/II or Stage III melanoma undergoing treatment, may indicate disease regression or remission and the treatment is efficacious, or alternatively it may indicate disease progression to Stage III or Stage IV melanoma and the treatment is inefficacious.
An increase or up regulation of the level in the biological sample of miRNA-4731 in a subject with Stage I/II melanoma undergoing treatment, may indicate disease regression or remission and the treatment is efficacious, or alternatively it may indicate disease progression to Stage III melanoma and the treatment is inefficacious. Further, a decrease or down regulation of the level in the biological sample of miRNA-4731 in a subject with Stage III melanoma undergoing treatment may indicate disease progression to Stage IV melanoma and the treatment is inefficacious. Additionally, an increase or up regulation of the level in the biological sample of miRNA-4731 in a subject with Stage III or IV melanoma undergoing treatment, may indicate disease regression or remission and the treatment is efficacious.
In one embodiment, the method further comprises measuring the expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p. In this regard, a decrease or down regulation of the level in the biological sample of an additional biomarker, such as miRNA-16 and/or miRNA-211, may indicate the treatment is efficacious. Additionally, for miRNA-16, a decrease or down regulation of its expression level in the biological sample from a subject with Stage I/II or III melanoma, may indicate disease progression to Stage IV melanoma and the treatment is inefficacious. Further, an increase in the level or up regulation in the biological sample of an additional biomarker, such as miRNA-509-3p and/or miRNA-509-5p, may indicate the treatment is efficacious.
In certain embodiments, the expression level of one or more protein and/or nucleic acid biomarkers, as herein before described, may also be determined.
In particular embodiments, the biological sample comprises tissue, blood, serum, plasma or cerebrospinal fluid. Preferably, the biological sample comprises serum, plasma, or cerebrospinal fluid, although without limitation thereto.
It would be appreciated by the skilled artisan that the methods of present invention also include within their scope fragments and variants of the miRNA biomarkers described herein.
In this regard, a miRNA “fragment” includes a nucleic acid sequence that constitutes less than 100%, but at least 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or 95% of said miRNA sequence. In particular embodiments, a miRNA fragment may comprise, for example, at least 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 and 24 contiguous nucleotides of said miRNA biomarkers.
As used herein, “variant” refers to those miRNAs described herein that have one or more nucleic acids deleted, added or substituted by different nucleotides. In particular embodiments, a miRNA variant may comprise, for example, 1, 2, 3, 4, 5, 6, 7, 8 or 9 nucleotide deletions, additions and/or substitutions. In this regard, it is well known in the art that miRNAs can demonstrate some degree of variation from their respective reference miRNA sequence. By way of example and referring to FIGS. 11-15, it will be appreciated that the nucleotide sequences of SEQ ID NOS: 1-7 are “canonical” or preferred sequences and that there may be significant variability with respect to the length of the nucleotide sequence of each miRNA biomarker, in particular.
In some embodiments, a diagnostic, prognostic and/or treatment expression level of the one or more miRNA biomarkers described herein is correlated to melanoma by merely its presence or absence. In other embodiments, a threshold level of a diagnostic, prognostic and or treatment expression level of the one or more miRNA biomarkers can be established, and the level of the one or more miRNA biomarkers in a subject's biological sample can simply be compared to the threshold level.
In some embodiments, multiple time points prior to, during and/or after treatment of a subject with melanoma may be selected to determine the expression level of the one or more miRNA biomarkers, with or without other biomarkers, to determine a diagnosis, prognosis or treatment efficacy. For example, the expression level of the one or more miRNA biomarkers with or without additional specific miRNA biomarkers, protein biomarkers and/or nucleic acid biomarkers can be determined at an initial time point and then again at one, two, three or more time points.
Suitably, the time points may be selected throughout a treatment cycle or over a desired time period. Over a desired time period, for example, the time points may be prior to treatment, mid way through treatment and/or after treatment has been completed. Suitably, an altered or modulated expression level, such as a decrease or reduction in the expression level, of the one or more miRNA biomarkers, such as miRNA-4487, miRNA-4706 and/or miRNA-4731, may be utilised by the methods of the invention from the first to second and/or third time points may provide a poor prognosis for a subject with melanoma. Alternatively, an altered or modulated expression level, such as an increase or upregulation in the expression level, of the one or more miRNA biomarkers, such as miRNA-4487, miRNA-4706 and/or miRNA-4731, utilised by the methods of the invention from the first to second and/or third time points may provide a positive prognosis for a subject with melanoma.
As would be appreciated by the skilled artisan, the alteration or modulation in expression level of the one or more miRNA biomarkers may also be related to the severity, stage, recurrence or progression of melanoma and/or the efficacy of the treatment. By way of example, differences in the expression levels of the one or more miRNA biomarkers may be used to delineate stage III (i.e., regional metastatic) from stage IV (i.e., distant metastatic) melanoma patients. In addition, differences in the expression levels of the one or more miRNA biomarkers may be used to delineate stage I/II (i.e., non-metastatic) from stage III/IV (i.e., metastatic) melanoma patients.
In one embodiment, biological samples may be sourced and/or collected from a subject at diagnosis and then prior to each cycle of treatment. Suitably, there may be any number of treatment cycles, depending on the subject and the nature and/or stage of melanoma, including but not limited to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15 and/or 20 cycles. The treatment cycles may be close together, spread out over a period of time and/or intense cycles at defined time points over a period of time, or any combination of the above.
In further embodiments, samples may be taken both during treatment and/or after treatment has been completed. Suitably, samples may be sourced from a subject at any time point after treatment has been completed, examples of which include 1, 2, 3, 4, 5, 10, 15, 20, 25 and/or 30 days post treatment, 1, 2 and/or 3 weeks post treatment and/or 1, 3, 6 and/or 9 months post treatment and/or 1, 2, 3, 4, 5, 10, 15, 20 and/or 30 years post treatment. The treatment may be completed once the subject is in remission or after at least one or more treatment cycles, depending on the subject and the melanoma.
In some embodiments, subjects are sampled every three to six months and/or every year post treatment. It will be understood by a person of skill in the art that a subject may be in remission or may have non-responsive or relapsed melanoma. If subjects have non-responsive or relapsed disease, then monitoring may be more frequent.
In one embodiment, a decrease or no change in the expression level of the one or more miRNA biomarkers, such as miRNA-4487, miRNA-4706 and/or miRNA-4731, in a second, third, fourth and/or fifth etc., biological sample from a subject collected either after treatment or during treatment as compared to the expression level in a first earlier sample, may indicate progression of melanoma or failure of the treatment, such as the presence and/or development of resistance.
In one embodiment, an increase in the expression level of the one or more miRNA biomarkers, such as miRNA-4487, miRNA-4706 and/or miRNA-4731, in a second, third, fourth and/or fifth etc., biological sample from a subject collected either after treatment or during treatment as compared to the expression level in a first earlier sample, may indicate that the treatment is efficacious.
The methods described herein may be suitable for any biological sample from a subject. In particular embodiments, the biological sample is or comprises tissue, blood, serum, plasma or cerebrospinal fluid. Typically, the miRNAs are obtainable from a non-cellular source. Accordingly, the biological sample is, comprises, or is obtained from a non-cellular source. The biological sample may be serum, plasma, or cerebrospinal fluid, although without limitation thereto.
Additionally, the one or more biomarkers described herein may be used in in vitro assays of response by subject-derived test cells or tissue to predict an in vivo response to the treatment. As such, the one or more biomarkers described herein, may be used, for example, to predict and monitor how particular subjects with melanoma respond to therapeutic intervention with the treatment. A biomarker expression pattern correlating with sensitivity or resistance of cells following exposure of the cells to the treatment may provide a useful tool for screening one or more tumour samples from the subject before treatment. The screening allows a prediction of cells of a biological sample of the melanoma exposed to the treatment, based on the expression results of the one or more biomarkers, as to whether or not the melanoma, and hence a patient harbouring the melanoma, will or will not respond to the treatment.
It would be understood that the methods described herein may be applicable to any mammal. In particular embodiments, the term “mammal” includes but is not limited to humans, performance animals (such as horses, camels, greyhounds), livestock (such as cows, sheep, horses) and companion animals (such as cats and dogs). Preferably, the subject is a human.
So that the present invention may be more readily understood and put into practical effect, the skilled person is referred to the following non-limiting examples.

EXAMPLE 1

Materials and Methods

Patient Specimen Details and Inclusion Criteria
VALIDATION Cohort 1. Stage III melanoma tissues (‘PAH-tissue’) were obtained from a prospective database of stage IIIA-C cutaneous melanoma patients, presenting to the Princess Alexandra Hospital (PAH) Melanoma Unit and affiliated private hospitals, which has been maintained since 1997. Permission to collect and use information was approved by the hospital ethics committee (HREX Reference number: HREX/11/QPAH/650; SSA reference number: SSA/11/QPAH/694). Formalin-fixed paraffin embedded (FFPE) tumour samples from 72 patients who met these criteria were reviewed according to the AJCC staging system for cutaneous melanoma seventh edition¹by an experienced melanoma pathologist. Pathologic and treatment-related variables were prospectively recorded.
Stage III and IV melanoma tissues ‘MIA-tissue’ were obtained from a prospective database of melanoma patients, presenting to the Melanoma Institute Australia (formerly Sydney Melanoma Unit) and affiliated private hospitals, which has been maintained since 1967. Cases and corresponding available FFPE tissue specimens were selected that fell into one of the following three categories, (1) diagnosed with distant metastatic melanoma and survived less than 1 year, (2) diagnosed with distant metastatic melanoma and survived greater than 5 years, and (3) diagnosed with regional lymph node metastasis. Informed written consent was obtained for each patient under approved protocols (Protocol No X10-0305 &HREC/10/RPAH/539 and Protocol No X10-0300 HREC/10/RPAH/530) governed by the Human Research Ethics Committee of the Royal Prince Alfred Hospital (Sydney NSW, Australia)
VALIDATION Cohort 2 and 3. Sera from ‘Healthy Controls’ were ascertained from a large cohort of participants collected as part of the Australian Cancer Study (ACS) (QIMR Berghofer HREC approved project no. P399). As part of the ACS, potential controls were randomly selected from the Australian Electoral Roll (enrollment is compulsory). Controls were prospectively sampled from within strata of age (in 5 year age-groups) and state of residence. Of 3,258 potentially eligible control participants, 41 could not be contacted and 175 were excluded because they were deceased (16), too ill (61), or unable to read or write in English (98). Of 3,042 controls meeting the inclusion criteria, 1680 (55%) gave their consent to take part. Completed questionnaires were returned by 1580 controls (48% of all potentially eligible controls selected from the roll). See Table 1 for participant descriptive statistics.
Sera from ‘High nevus count’ and ‘History of melanoma’ participants were prospectively collected from cohorts who were enrolled in a large study titled: ‘Pigmentation genotypes and phenotypic correlations with dermoscopic naevus types and distribution’. These samples were included as ‘controls’ to determine the level of expression measurable in sera derived from patients with a high melanocyte burden. All study participants were enrolled in the following human ethics approved projects: QIMR HREC/P1237, The Metro South Health District HREC/09/QPAH/162, and UQ HREC approval number is 2009001590. Participants with a history of melanoma were recruited through the Melanoma Unit and Dermatology Department of the Princess Alexandra Hospital, Brisbane, Queensland, Australia, between May 2012 and November 2012. Control participants, with no personal history of melanoma, were recruited from the Brisbane Twin Naevus Study between August 2012 and November 2012. All participants had 16-panel full-body images and dermoscopic images of significant naevi recorded²¹. Significant naevi were defined as naevi greater than or equal to 5 mm on all body sites except the scalp, buttocks, mucosal surfaces and genitals, and greater than or equal to 2 mm on the back of both males and females and on the legs of females. All significant naevi were classified by the predominant dermoscopic pattern (reticular, globular, or non-specific), colour, and profile (flat, raised, domed or papillomatous). See Table 1 for participant descriptive statistics.
Sera from Stage I/II and IV melanoma patients (staged according to the current AJCC staging manual¹) had blood drawn and serum stored as part of a large prospectively collected cohort from the university department of dermatology in Tubingen, Germany (‘Tubingen’ cohort). Bio-banking and the usage of corresponding patient data for this study was approved by the Ethics Committee, University of Tubingen (approvals 657/2012BO2). An additional cohort of patients was obtained from the Melanoma Institute of Australia, Sydney (‘MIA’ cohort). Patients were staged I-IV according to the current AJCC staging guidelines¹. Informed written consent was obtained for each patient under approved protocols (Protocol No X10-0305 &HREC/10/RPAH/539 and Protocol No X10-0300 HREC/10/RPAH/530) governed by the Human Research Ethics Committee of the Royal Prince Alfred Hospital (Sydney NSW, Australia).
Total RNA extraction from ‘Validation cohort 1’: FFPE tissue. The extraction of total RNA from FFPE tissue was performed using miRNeasy FFPE Kits (QIAGEN). A sterile disposable biopsy punch (BPP-20F; Kai Medical, Japan) was used to retrieve tumour content from blocks that had been scored and marked for content via H&E histological staining. This process allowed for minimal stromal contamination. This kit is specifically designed to reverse the formalin crosslinking of RNA, efficiently releasing RNA from tissue sections, while avoiding significant RNA degradation.
Total RNA extraction from ‘Validation cohorts 2 and 3’: serum. Total RNA was extracted from serum following the manufacturer's protocol using the miRNeasy Serum/Plasma Kits (QIAGEN) which efficiently purify RNA from up to 200 μl serum or plasma. As there is currently no consensus as to which miRNA should be used as an endogenous control¹⁵, a synthetic miRNA mimic (miRNeasy Serum/Plasma Spike-In Control: C. elegans miR-39 miRNA mimic) was spiked into each sample (5.6×10⁸copies/200 μl) to allow for normalization of expression data.
Selection criteria for ‘melanoma-specific’ miRNAs. In our previously published miRNA microarray data²⁰we found a total of 233/1898 miRNAs (‘Discovery’ set) (FIG. 1) that were differentially expressed (DE; P≦0.05 and ≧2 fold) between a group of cutaneous melanoma cell lines (n=55) as compared to a group of other (non-melanoma) solid malignancies (n=34). We then applied filtering criteria to the 233 DE miRNAs to identify which miRNAs would be suitable to measure in patient derived serum. The following strict criteria were used to filter the ‘Discovery’ set: ≧15 fold higher expression in cutaneous melanoma vs ‘other’ solid malignancies (n=14/14), or ≧2 fold higher expression in cutaneous melanoma vs ‘other’ solid malignancies with no detectable expression in melanocytes or melanoblasts (n=3/6). In addition, miR-16, which is known to be highly expressed in blood, was also included as a control. This identified an 18 miRNA panel (‘MELmiR-18’) that consisted of: miR-211-5p, miR-514a-3p, miR-509-3p, miR-204-5p, miR-509-5p, miR-513b, miR-145-5p, miR-146a-5p, miR-508-3p, miR-506-3p, miR-513c-5p, miR-4731-5p, miR-508-5p, miR-363-3p, miR-4487, miR-4469, miR-4706, and hsa-miR-16. This panel was carried forward for validation in independent cohorts of patient derived sera and subsequent correlation with panel analysis in FFPE of melanoma tumours.
Reverse transcription and PreAmplification using custom primer and assay pools. The standard TaqMan® MicroRNA Assay protocol calls for an individual gene specific reverse transcription (RT) reaction for each assay, however in order to maximize the full capability of the Biomark HD (Fluidigm) system, we developed a modified protocol (User bulletin publication part number 4465407, revision date January 2013 (Rev. C)). Briefly, a custom RT primer pool consisting of equal amounts of miRNA-specific RT primer plus an additional pool of the corresponding TaqMan® MicroRNA Assay (PreAmp Primer Pool) were used to pre-amplify the RT reaction. The TaqMan® PreAmp Master Mix was used to amplify small amounts of cDNA reactions (this was specifically designed to avoid amplification bias). The RT protocol was as per manufacturers guidelines and the PreAmplification protocol was modified in the following way: total reaction volume was scaled down to 10 μL (2.5 μL of RT-product, 5 μL of TaqMan PreAmp Master Mix (2×), and 1.5 μL of PreAmp Primer Pool) with a total of 14 cycles used in the PreAmplification.
qRT-PCR using the Fluidigm® 96.96 Dynamic Array™ and BioMark™ HD. The 96.96 Dynamic™ Array on the Fluidigm system allows simultaneous measurement of 96 assays with 96 samples totaling 9216 reactions with minimal cDNA and (2.7 ul) assay (2.5 ul) input. The Dynamic Array has an on-chip network of microfluidic channels, chambers and valves that automatically assemble individual PCR reactions. The BioMark system has the ability to detect a single copy in a 6.75 nl reaction volume, a feat not possible with conventional real-time PCR instruments (5 ul minimum volume). In this way, it is possible to combine up to 96 primer assays to create a high throughput, specific and highly sensitive custom miRNA expression panel, which maximizes the information that can be obtained from each RNA sample. This ultra-sensitivity allows for confidence in its ability to detect subtle changes in gene expression—a requirement for serum miRNA detection.
qRT-PCR analysis to determine the median normalized Ct expression value. The expression of the ‘MELmiR-18’ panel (FIG. 1) was assayed in each sample with at least 4 replicate Taqman assays to determine their expression. Firstly, real-time expression data was analyzed and extracted using Fluidigm's Real-Time PCR Analysis software using default parameters (Quality Threshold=0.65, Baseline Correction Method=Linear, and Ct Threshold Method=Auto (Global)). The combined data from all 96.96 Dynamic Array runs were then combined in Microsoft Excel so as to perform global normalization (median normalized Ct value method)²². Firstly, the average Ct value was determined for each sample-assay combination (including the synthetic spike-in control cel-miR-39). The normalization factor was then derived by subtracting the mean cel-miR-39 Ct value of each sample from median cel-miR-39 Ct value of all samples. The median normalized Ct value was then calculated by adding the normalization factor to the averaged Ct value of each sample-assay combination.
For the FFPE samples, again, the average Ct value was determined for each sample-assay combination (including the endogenous control RNU-6). The normalization factor was then derived by subtracting the mean RNU-6 Ct value of a positive control sample with known quantity (15 ng) of total RNA (added to the initial cDNA reaction) from each RNU-6 assay-sample combination. The median normalized Ct value was then calculated by adding the normalization factor to the averaged Ct value of each sample-assay combination.
Statistical Analysis. All data analysis (Mann-Whitney U test, ROC analysis, and Binary Logistic Regression) was performed in Graph Pad Prism 6 or SPSS for Windows, version 21.0 (Statistical Package for the Social Sciences, SPSS, Inc., Chicago, Ill.) with median-normalized Ct expression values using default parameters.
Diagnostic inclusion criteria. To maximize the chances of having a positive signal expressed in the patients serum, only the combined stage IV cohorts were used (n=119; ‘TUBINGEN’ and ‘MIA’) to compare against disease-free ‘controls’ (n=130; no history of melanoma or nevi, prior history of melanoma, high nevus count with no melanoma). Initially, all members of the ‘MELmiR-18’ panel underwent a simple Mann-Whitney U test to identify the highly significant (p.<0.0001) miRNAs to be included in the next step (FIG. 1). Those miRNAs that met these criteria then underwent ROC analysis to determine their Area under the Curve (AUC) (FIGS. 1 and 2). The miRNAs that had an AUC≧0.70 (‘MELmiR-7’ panel) were interrogated further to classify the median-normalized Ct values as ‘high’ or ‘low’ expression for diagnostic test purposes along with survival analysis. AUC scores≧0.7 were deemed to be diagnostically useful²³. Interpretation of Ct expression values used to determine ROC curves were evaluated with an arbitrary cut-off of ‘85% sensitivity’. Those that were greater than this cut-off were classed as ‘high’ and those that were below were ‘low’. Binary Logistic Regression (SPSS for Windows, version 21.0 (Statistical Package for the Social Sciences, SPSS, Inc., Chicago, Ill.) analysis was also performed to provide further justification of the inclusion or exclusion of particular miRNAs in the MELmiR-7 panel.
Diagnostic Score assignment. For those miRNAs that met the criteria for inclusion in the diagnostic panel (FIGS. 1 and 2), the patient was given a diagnostic score (ranging from 0-7) determined by the number of miRNAs that were present as ‘high’ and ‘low’ (graphing of Ct values as part of the Mann-Whitney U test allowed a visual determinant of the stage IV cohort being higher or lower than the ‘control’ cohort for each miRNA) or ‘normal’ (most like the ‘control’ cohort). To be deemed positive for melanoma, the patients sample must have had a score≧2 (max 7). A negative test was a score of 0 or 1.
Diagnostic test evaluation. The following formulas were used to determine diagnostic test ability:
Positive Predictive Value (PPV) or Precision=True Positive (TP)/(TP+False Positive (FP))
Negative Predictive Value (NPV)=True Negative (TN)/(False Negative (FN)+TN)
Sensitivity=TP/(TP+FN)
Specificity=TN/(FP+TN)
False Positive Rate=1−Specificity
False Negative Rate=1−Sensitivity
Power=Sensitivity
Likelihood Ratio Positive=Sensitivity/1−Specificity
Likelihood Ratio Negative=1−Sensitivity/Specificity
Diagnostic Odds Ratio (DOR)=(TP/FN)/(FP/TN)
Survival Curve analysis. All Kaplan Meier curves were drawn in Graph Pad Prism 6 using date of last follow-up to compare to date of diagnosis (stage IV) to determine survival years and Alive/Dead status. ‘High’ and ‘Low’ categories were derived as discussed. P values were determined using the Mantel-Cox (Log-rank) test and the Hazard Ratio (HR) and Confidence Intervals (CI) were determined by the Mantel-Haenszel test. Serum LDH and S100B data were available from routine testing carried out in the ‘Tubingen’ cohort.

Results

Validation of ‘melanoma-specific’ miRNA expression in an independent cohort of stage III and stage IV FFPE melanoma tumors shows significant differences between stages. The ‘MELmiR-18’ panel (see Materials and Methods for selection criteria) was assessed (along with the endogenous control RNU-6) in a retrospective collection (‘Validation Cohort 1’) of FFPE melanoma tissues derived from stage III melanoma patients (n=82) and stage IV melanoma patients (n=10) (FIG. 1). Each assay had a serial dilution of a positive control sample (known expression for all miRNAs in panel) that had a total input of 1 ng, 3 ng, 15 ng, and 45 ng in the original cDNA reaction. The BioMark HD Real-time PCR system (Fluidigm) was used to maximize the chances for miRNA detection. Detectable expression was observed in all dilutions of the positive control (except miR-4469 which had assay failure) as well as being expressed to varying degrees in the FFPE tissue-derived RNA. Grouping of the samples into stages (III vs IV) allowed for statistically significant separation (p.0.031-p.<0.0001) in 13/18 miRNAs with 4/18 (miR-145-5p, miR-363-3p, miR-4706, and miR-514b) being non-significant (ns). The most strongly significant miRNA was miR-4731 (p.<0.0001) followed by miR-4487 (p.0.001) followed by most of the members of the miR-506-514-cluster (Table 4 and FIG. 5). Subsequent ROC analysis (stage III vs IV) confirmed the Mann-Whitney U tests where all previously identified significant miRNAs also had ‘good’ to ‘strong’ AUC scores ranging from 0.709-0.854. (Table 4). The highest AUC scores were 0.854 (miR-4731) and 0.821 (miR-4487). Members of the miR-506-514 cluster had AUC scores ranging from (0.709-0.793) with the highest scores being shared by miR-508-3p and miR-509-5p.
Circulating miRNA expression is readily detectable in serum using a high-throughput ultra-sensitive method of detection. The same ‘MELmiR-18’ panel was then assessed (along with a spiked-in C. elegans (cel-miR-39) miRNA for normalization purposes) in a large series of retrospective and prospective cohorts of patient sera (‘Validation Cohort 2’) with varied stages of disease: no melanoma (n=102), no melanoma but high mole count (n=12), prior history of melanoma (n=16), and stage IV melanoma patients (n=119) (FIG. 1). An ultra-sensitive, high-throughput method of detection (Biomark HD, Fluidigm in combination with TaqMan® MicroRNA Assays, Life Technologies) was used to maximize the chance of detecting low abundance miRNAs which may have been present in the specimens. In miRNA derived from serum, detectable expression was measured in 13/18 miRNAs (Table 2) with 5/18 miRNAs showing no detectable expression. Those miRNAs that had no expression included: miR-4469, miR-508-3p, miR-508-5p, miR-513b, and miR-513c (data not shown). In the 13/18 miRNAs having detectable expression, eight (miR-16, miR-204-5p, miR-211-5p, miR-4487, miR-4706, miR-4731, miR-509-3p, and miR-509-5p) of these showed highly significant differences (Mann-Whitney U-test: p<0.0001) between ‘controls’ (no melanoma, no melanoma but high mole count, and a prior history of melanoma) and patients with stage IV disease (Tables 2 and 5 shows the p values and AUC scores when compared to healthy controls alone). Importantly, the eight highly significant (p<0.0001) miRNAs represented in FIG. 5 show that miR-16 and miR-211-5p have increased expression in melanoma cases compared to controls and the remainder have lower expression as compared to controls.
Table 2 also highlights the significance of the 13 detectable miRNAs in relation to stage I/II (n=86) vs. ‘controls’ along with stage III (n=50) vs. ‘controls’ (‘Validation Cohort 3’). It is important to note that the same highly significant miRNAs (except miR-211-5p) that discriminated stage IV vs. ‘controls’ are also able to discriminate the lesser stages.
Circulating ‘melanoma-specific’ miRNA expression shows significant differences between stages of melanoma. Highly significant expression (Man-Whitney U-test p<0.0001) differences were also observed between stage III vs. stage IV melanoma patients along with stages I/II vs. stage IV melanoma patients (Table 2 and FIG. 6).
A panel of 7 miRNAs has the ability to diagnose melanoma with high sensitivity and specificity. Receiver operating characteristic (ROC) curve analysis revealed which of the highly significant miRNAs (Table 2) have the potential to be used for diagnostic/prognostic purposes (FIG. 2). Particular attention is paid to those miRNAs that were able to detect all stages of disease (stages I-IV) (Table 2). The following miRNAs fitted these criteria: miR-16, miR-211-5p, miR-4487, miR-4706, miR-4731, miR-509-3p, and miR-509-5p (‘MELmiR-7’). FIG. 7 shows all of the ROC curves associated with stage IV vs. ‘controls’.
Analysis of the ROC curve data (see Materials and Methods) was used to categorise expression levels into ‘high’ or ‘low’ vs. ‘controls’, or ‘normal’ (control-like) (FIG. 2). A diagnostic score was then applied to each sample (see Materials and Methods) which ranged from 0 or 1 (low likelihood of melanoma) and 2-7 (high likelihood of melanoma). Upon applying the derived diagnostic score, the ‘MELmiR-7’ panel was evaluated as a group. We found that it had the ability to diagnose melanoma (independent of stage), when ≧2 miRNAs were expressed (96% sensitivity and ≧80% specificity) (Table 3). The sensitivity of the ‘MELmiR-7’ panel increases to 98% when stage IV is compared to ‘controls’ (FIG. 2). Table 3 provides a summary of the effectiveness of the ‘MELmiR-7’ panel in relation to other stages (see Materials and Methods), particularly for the Diagnostic Odds Ratio (DOR). This ratio was used to determine the lowest diagnostic score possible for the ‘MELmiR-7’ panel yet still maintaining very high sensitivity and specificity.
The ‘MELmiR-7’ panel is more sensitive then serum LDH and S100B in diagnosing stage IV melanoma patients. Stage IV melanoma patients (‘TUBINGEN’ cohort) had serum LDH and S100B levels determined as part of their standard of care regimen. Elevated levels of serum LDH and S100B were found in 39% (27/69) and 63% (43/68) of these patients (data not shown). In the same patients, the ‘MELmiR-7’ achieved 96% (66/69) sensitivity and ≧80% specificity. Specificity could not be determined for serum LDH and S100B as ‘controls’ were not assayed.
Binary logistic regression models predicts the accuracy of the MELmiR-7 panel. In regression analysis, using just the median normalised Ct values, high predictive accuracy was achieved when using a binary logistic regression model (stepwise backwards conditional) which was able to classify stage IV melanoma and ‘controls’ 95% and 94% respectively (FIG. 3). The same accuracy was achieved when in a stepwise fashion; miR-16 and miR-4487 were left out of the model (Table 6). When stage III were compared to ‘controls’, the highest level of accuracy (90% and 99% respectively) was achieved when miR-4731 and miR-509-3p were left out of the model (Table 5). Likewise, when stage I/II were compared to ‘controls’, the same level of predictive accuracy was maintained following 3 stepwise iterations. The highest percentages were 94% and 97% prediction when miR-211-5p and miR-4731 were left out of the model.
Binary logistic regression models predict recurrence in stage III melanoma patients using the MELmiR-7 panel. Interestingly, when stage III patients were compared to stage IV patients using the same regression model, the complete MELmiR-7 panel achieves an accuracy of 80% (40/50) prediction of stage III vs. 92% prediction of stage IV (FIG. 3). Closer inspection of the stage III patients (at time of blood collection) that were misclassified (according to the regression model) as stage IV (10/50) revealed that 60% (6/10) subsequently developed stage IV recurrence. The combinations of stage III vs. stage IV and stage I/II vs. stage IV can be seen in Table 6.
In stage III melanoma patients with known dates of disease recurrence (stage IV diagnosis) following blood-draw (Median follow-up=20 months, Progression free survival=15.7 months), binary logistic regression was able to predict recurrence in 8/13 (62%) patients and no sign of recurrence (NSR) in 33/37 (89%) patients (FIG. 3). The prediction of NSR was increased to 95% (35/37) when only miR-16, miR-4731, and miR-509-5p were measured (see Table 6 for all possible iterations) however prediction for recurrence was reduced to 54% (7/13).
Binary logistic regression models predict short and long survival in stage IV patients using the MELmiR-7 panel. Stage IV melanoma patients can be classified as short (<2years) and long (>2 years) survivors using the ‘MELmiR-7’ panel. Using binary logistic regression models, short and long survivors have a prediction of 85% (57/67) and 64% (33/52) respectively (FIG. 3). The highest level of prediction for short survivors increases to 88% (59/67) when only miR-211-5p, miR-4706, miR-4731, and miR-miR-509-5p are measured (see Table 6 for all possible iterations).
Binary logistic regression models predict M1a-c stage in stage IV melanoma patients using the MELmiR-7 panel. Stage IV patients are classed as M1a, M1b, and M1c, depending upon the site of distal metastasis. In stage IV patients (MIA cohort; n=40) previously classed as M1a-c at last follow-up, using binary logistic regression, the MELmiR-7 panel was able to predict M1a in 14/14 (100%) cases and M1b in 4/6 (67%) cases respectively (FIG. 3). Prediction of M1b was increased to 83% (5/6) and was decreased to 93% (13/14) for M1a, when only miR-16, miR-4487, miR-4706, miR-4731, and miR-509-3p were expressed. When M1b was compared to M1c, the complete MELmiR-7 panel was able to predict 4/6 (67%) and 15/16 (94%) cases respectively. However, when M1a cases were compared to M1c, 100% prediction was achieved respectively (FIG. 3) (see Table 6 for all possible iterations).
miRNA expression is strongly associated with survival in stage IV patients. The individual miRNAs from the ‘MELmiR-7’ panel identified following ROC analysis were then assessed for their prognostic ability in terms of survival using date of last follow-up data along with alive/dead from melanoma status (up to 5 years post stage IV diagnosis). One miRNA, miR-211-5p, showed statistically significant association with survival of stage IV patients. Kaplan-Meier survival curve analysis identified high miR-211 expression as the most strongly significant predictor of survival, with median survival being 1.1 years vs. 3.5 years (HR 3.2, 95% CI 2.02-5.14, p<0.0001) (FIG. 4) for high and normal miR-211-5p expression.
Additionally, for the stage IV patients (Tubingen cohort) that had S100B (n=67) and serum LDH (n=69) status available, median survival times observed for elevated S100B, serum LDH, and miR-211-5p, was 1.21, 1.22, and 1.26 years respectively (data not shown).

Discussion

The linear progression model for melanoma has been in place for many years and by and large, most of the observable tumors fit into this framework. Staging of melanoma adheres to strict guidelines devised by the American Joint Committee on Cancer (AJCC) 1 and is based upon evidence-based gross observation and currently available serological markers (eg serum LDH). This is a highly effective system and dictates the level of treatments which are offered to melanoma patients. However, this model which does not take into account the presence of metastatic disease unobservable to the treating physician by conventional methods (e.g. palpation, CT scans). Currently, there are no serological tests available that are sensitive enough to detect the presence of melanoma in a patient before it manifests into ‘observable’ disease.
We have successfully confirmed the expression of a ‘melanoma-specific’ panel of miRNAs that are not only diagnostically accurate in FFPE tissue, distinguishing stage III from stage IV tumours, but also, via a minimally-invasive blood test, a proportion of these miRNAs (‘MELmiR-7’) allow for highly accurate diagnosis and prognosis of melanoma patients (independent of AJCC stage).
miRNAs are commonly found in the circulation and have been associated with cancer progression^{13, 14}. miRNAs are also inherently resistant to degradation (if contained within exosomes or bound to AGO2)13-15 and are readily detectable in serum and plasma, using precise methods of isolation and detection. Using a comprehensive approach to identify a panel of miRNAs (MELmiR-18)²⁰that could be deemed ‘melanoma-specific’, combined with using an ultra-sensitive method of detection, we have successfully identified a highly sensitive (96%) and specific (≧80%) diagnostic panel of seven miRNAs (‘MELmiR-7’ including: miR-16, miR-211-5p, miR-509-3p, miR-509-5p, mir-4487, miR-4731-5p, and miR-4706) which have the ability to correctly diagnose melanoma (n=242) independent of AJCC stages I-IV. Most importantly, the sensitivity increases to 98% in stage IV patients (n=119) when 2 or more miRNAs are expressed as compared to ‘controls’. In fact, the panel also achieved high sensitivity and specificity in stage I/II (98% and ≧80% respectively) as well as stage III (91% and ≧79% respectively) thus paving the way for early diagnosis to be achieved.
There is currently an unmet need for a highly specific and predictive serum biomarker of melanoma burden, as for many years the consensus whether or not to use currently available serological markers (S100B and serum LDH) is often disputed due to the ranges of sensitivity and specificities that some studies have reported^{4-8, 10-12}. As such, dependent upon the local treatment regimen, these are not commonly used. Despite the lack of consensus for sensitivity and specificity, a recent study did find that elevated levels of S100B were able to predict survival times in unresectable melanoma patients²⁴.
We have subsequently found that elevated expression levels of miR-211-5p (a member of our MELmiR-7 panel) have the ability to predict prognosis in stage IV melanoma patients. Interestingly in the same group of patients, these survival data mirror those found for S100B, thus making the MELmir-7 panel not only highly sensitive as a diagnostic panel, but also highly informative in terms of prognosis.
Stage IV patients are further categorized into M-stage which is given according to where the tumor has metastasized (M1a: skin, subcutaneous, or distant lymph nodes; M1b: lung; and M1c: other visceral metastases or any distant metastasis with elevated serum LDH)1. We have found that the MELmir-7 panel allows for 100% discrimination between M1a and M1c. These data are highly suggestive of the level of MELmiR-7 panel expression being relative to the presence of tumour in the circulation. This panel would therefore be ideally suited to monitor tumour progression in patients diagnosed with late stage disease (IV) following treatment to enable a change in therapy following a subsequent recurrence. In addition, in a cohort of stage III melanomas (MIA) with known dates of recurrence, the MELmiR-7 panel was able to predict recurrence in 62% of patients and confirm no sign of recurrence (NSR) in 89% of patients.
According to the AJCC Staging committee, stage III melanoma patients have a 50% chance of survival beyond 5 years 1; these patients also remain the most difficult to provide effective treatments/surveillance regimens and accurate survival estimates. Melanoma patients diagnosed with stage III disease have tumours that have spread to regional lymph nodes or have developed in-transit metastases or satellites but have no evidence of distant metastasis. The treatment for stage III melanoma relies heavily on the accuracy of the clinical staging of the disease and (which can vary across melanoma centres) as the most effective treatment is surgical removal of the affected lymph nodes/satellites25. Following treatment, stage III patients are subjected to a series of standard physical examinations including computer tomography (CT) scans and serum LDH/S100B (depending on treatment centre) testing every 3 months for the first year, every 4 months in the second year, every 6 months in the third-fifth year, and then yearly for every subsequent year after 5 years. The frequency of these tests is deemed necessary for early detection of distant metastases; however this causes a significant burden to both the patient and the healthcare system. To alleviate this burden, the MELmiR-7 panel could be offered to patients to complement physical examination. If the diagnostic score for melanoma positivity has changed from earlier measurements, then this may indicate the presence of disease recurrence and as such, these patients may qualify earlier for adjuvant, systemic, or targeted therapies that would other-wise be only offered to stage IV patients. The use of this miRNA panel in this manner has the potential to greatly increase the chances of survival by earlier and more precise detection of the presence of metastases.
In terms of prognosis, miR-211-5p measurement may also allow better triaging of patients diagnosed with stage IV disease, into good and poor prognosis which would be highly informative for not only the treating clinician but also for the quality of life of the patients.

TABLE 3

Summaries of diagnostic test statistics generated when
AJCC staged melanoma is compared with controls.

MELmiR-panel	MELANOMA	Stage I/II	Stage III	Stage IV

Diagnostic	≧2	≧3	≧2	≧3	≧2	≧3	≧2	≧3
Score
Sensitivity	96%	91%	98%	92%	91%	86%	98%	93%
Specificity
	80%	90%	80%	90%	80%	90%	80%	90%
False Positive	20%	10%	20%	10%	20%	10%	20%	10%
Rate
False Negative	4%	9%	2%	8%	9%	14%	2%	7%
Rate
Positive	89%	94%	72%	85%	60%	77%	78%	89%
Predictive
Value (PPV)
Negative	93%	84%	98%	95%	96%	94%	98%	94%
Predictive
Value (PPV)
Likelihood	4.75	8.84	4.83	8.94	4.49	8.33	4.82	8.99
Ratio Positive
Likelihood	0.05	0.10	0.03	0.08	0.11	0.15	0.03	0.08
Ratio Negative
Diagnostic	100.45	92.08	169.39	106.48	39.39	54.17	156.26	114.59
Odds Ratio
(DOR)

TABLE 6

Tables were compiled following Binary Logistic Regression analysis performed in SPSS
for Windows. version 21.0. To represent the overall analysis performed for each cohort.
only the “classification table” and the “variables in the equation” are shown.

Controls vs Stage I/II
Classification Table^a

Predicted

Cohorts

Percentage

Observed	Controls	Stage I/II	Correct

Step

1	Cohorts	Controls	126	4	96.9
		Stage I/II	5	81	94.2

Overall Percentage

95.8

Step 2	Cohorts	Controls	126	4	96.9
		Stage I/II	5	81	94.2

Overall Percentage

95.8

Step 3	Cohorts	Controls	126	4	96.9
		Stage I/II	5	81	94.2

	Overall Percentage			95.8

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig.	Exp(B)

Step 1^a	miR16	−1.270	.444	8.169	1	.004	.281
	miR211	025	.110	.052	1	.820	1.025
	miR4487	1.164	.305	14.579	1	.000	3.204
	miR4706	137	.114	1.453	1	.228	1.147
	miR4731	.171	.246	.487	1	.485	1.187
	miR509.3p	−.629	.358	3.096	1	.079	533
	miR509.5p	297	.088	11.266	1	.001	1.346
	Constant	−19.668	7.505	6.867	1	.009	.000
Step 2^a	miR16	−1.244	.427	8.506	1	.004	.288
	miR4487	1.160	.304	14.543	1	.000	3.188
	miR4706	.129	.108	1.436	1	.231	1.138
	miR4731	185	.239	.602	1	.438	1.203
	miR509.3p	−.611	344	3.159	1	.076	.543
	miR509.5p	.299	.088	11.585	1	.001	1.348
	Constant	−19.677	7.474	6.931	1	.008	.000
Step 3^a	miR16	−1.248	.416	8.996	1	.003	.287
	miR4487	1.218	.304	16.009	1	.000	3.380
	miR4706	.165	.101	2.670	1	.102	1.179
	mIR509.3p	−.576	.336	2.936	1	.087	.562
	miR509.5p	.278	.082	11.623	1	.001	1.321
	Constant	−17.449	6.719	6.745	1	.009	.000

^aVariable(s) entered on step 1: miR16, miR211, miR4487, miR4706, miR4731, miR509.3p, miR509.5p.

Controls vs Stage III

Classification Table^a

Predicted

Cohorts

Percentage

Observed	Controls	Stage III	Correct

Step

1	Cohorts	Controls		127	3	97.7
		Stage III	5	45	90.0

Overall Percentage

95.6

Step 2	Cohorts	Controls		127	3	97.7
		Stage III	5	45	90.0

Overall Percentage

95.6

Step 3	Cohorts	Controls	128	2	98.5
		Stage III	5	45	90.0

Overall Percentage

96.1

Step 4	Cohorts	Controls	129	1	99.2
		Stage III	6	44	88.0

Overall Percentage

96.1

Step 5	Cohorts	Controls	129	1	99.2
		Stage III	6	44	88.0

Overall Percentage

96.1

Step 6	Cohorts	Controls	129	2	98.5
		Stage III	6	44	88.0

	Overall Percentage			95.6

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig.	Exp(B)

Step 1^a	miR16	−1.531	.479	10.210	1	.001	.216
	miR211	.139	.096	2.101	1	.147	1.149
	miR4487	938	.329	8.146	1	.004	2.555
	miR4706	158	.129	1.494	1	222	1.171
	miR4731	−.118	.289	167	1	683	889
	miR509.3p	−.118	.362	107	1	.744	889
	miR509.5p	090	.091	.984	1	.321	1.094
	Constant	−12.453	6.711	3.443	1	.064	.000
Step 2^a	miR16	−1.620	.404	16.060	1	000	.198
	miR211	.135	.094	2.029	1	.154	1.144
	miR4487	.935	.328	8.110	1	.004	2.548
	miR4706	161	.128	1.590	1	.207	1.175
	miR4731	−.109	.287	143	1	705	897
	miR509.5p	.084	0.68	.905	1	.341	1.087
	Constant	−13.915	5.181	7.213	1	.007	000
Step 3^a	miR16	−1.595	.394	15.340	1	.000	.203
	miR211	.125	091	1.916	1	166	1.134
	miR4487	.860	250	11.804	1	.001	2.362
	miR4706	.144	118	1.490	1	.222	1.155
	miR509.5p	099	078	1.593	1	.207	1.104
	Constant	−14.800	4.740	9.748	1	.002	.000
Step 4^a	miR16	−1.537	379	15.455	1	000	.215
	miR211	.086	086	1.011	1	315	1.090
	miR4487	1.029	239	18.610	1	000	2.799
	miR509.5p	103	077	1.815	1	178	1.109
	Constant	−13.936	4.632	9.052	1	.003	.000
Step 5^a	miR16	−1.431	343	17.446	1	000	.239
	miR4487	1.056	241	19.245	1	.000	2.876
	miR509.5p	.097	076	1.024	1	203	1.102
	Constant	−13.081	4.559	8.234	1	.004	000
Step 6^a	miR16	−1.497	.332	20.273	1	.000	224
	miR4487	1.150	226	27.243	1	000	3.255
	Constant	−12.700	4.451	8.142	1	004	000

Controls vs Stage IV

Classification Table^a

Predicted

Cohorts

Percentage

Observed	Controls	Stage IV	Correct

Step

1	Cohorts	Controls	122	8	93.8
		Stage IV	6	113	95.0

Overall Percentage

94.4

Step 2	Cohorts	Controls	122	8	93.8
		Stage IV	7	112	94.1

Overall Percentage

94.0

Step 3	Cohorts	Controls	122	8	93.8
		Stage IV	6	113	95.0

	Overall Percentage			94.4

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig	Exp(B)

Step 1^a	miR509.5p	386	090	19.622	1	000	1.474
	mir16	209	319	973	1	350	1.347
	miR211	−491	101	23.546	1	000	612
	miR4487	183	279	434	1	510	1.201
	miR4706	233	099	5.488	1	016	1.262
	miR4731	938	245	14.669	1	000	2.555
	miR509.3p	321	103	9.737	1	002	1.379
	Constant	−47.261	10.728	19.409	1	.000	000
Step 2^a	miR509.5p	402	088	20.744	1	.000	1.49
	miR16	.280	.306	.935	1	.361	1.323
	miR211	−494	100	24.343	1	.000	610
	miR4706	238	099	5.728	1	017	1.268
	miR4731	1.018	217	22.041	1	000	2.768
	miR509.3p	308	094	10.761	1	.001	1.361
	Constant	−44.729	9.547	21.951	1	.000	.000
Step 3^a	miR509.5p	399	088	20.659	1	000	1.499
	miR211	−484	.098	24.197	1	.000	.616
	miR4706	237	.097	5.930	1	.014	1.207
	miR4731	1.002	218	21.113	1	.000	2.724
	miR509.3p	297	093	10.262	1	001	1.345
	Constant	−40.322	7.744	27.109	1	000	000

^aVariable(s) entered on step 1: miR509.5p, miR16, miR211, miR4487, miR4706, miR509.3p.

Stage I/II vs Stage III

Classification Table^a

Predicted

Cohorts

Percentage

Observed	Stage I/II	Stage III	Correct

	Cohorts	Stage I/II	73	13	84.9
Step 1		Stage III	32	18	36.0

Overall Percentage

66.9

	Cohorts	Stage I/II	73	13	84.9
Step 2		Stage III	32	18	36.0

Overall Percentage

66.9

	Cohorts	Stage I/II	76	10	88.4
Step 3		Stage III	35	15	30.0

Overall Percentage

66.9

	Cohorts	Stage I/II	75	11	87.2
Step 4		Stage III	33	17	34.0

Overall Percentage

67.6

	Cohorts	Stage I/II	76	10	68.4
Step 5		Stage III	32	18	36.0

	Overall Percentage			69.1

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig.	Exp(B)

Step 1^a	miR16	−.157	.159	.982	1	.322	.854
	miR211	−.004	048	.005	1	.937	996
	miR4487	.305	.097	10.001	1	.002	1.357
	miR4706	−.003	.065	.931	1	.335	.939
	miR4731	−185	.077	5.886	1	.015	829
	miR509.3p	.176	.200	.770	1	.380	1.192
	miR509.5p	−.083	.040	3.193	1	.074	.920
	Constant	−.532	4.284	.015	1	.991	.587
Step 2^a	miR16	−.160	158	1.057	1	304	652
	miR4487	.306	.098	10.266	1	.001	1.359
	miR4706	−0.62	.005	.926	1	336	939
	miR4731	−188	.077	5.934	1	.015	828
	miR509.3p	.173	.197	.772	1	.380	1.189
	miR509.9p	−.083	.046	3.255	1	.071	.920
	Constant	−.572	4.253	.018	1	893	564
Step 3^a	miR16	−.124	.150	.083	1	408	883
	miR4487	.318	.095	11.225	1	.001	1.375
	miR4706	−054	.064	.709	1	.400	948
	miR4731	−195	.077	6.422	1	.011	823
	miR509.5p	−.080	.046	3.066	1	.080	923
	Constant	2.191	2.892	.574	1	.449	8.944
Step 4^a	miR4487	.295	.058	11.120	1	.001	1.342
	miR4706	−.043	.052	.479	1	.489	955
	miR4731	−.193	.077	6.354	1	.012	.825
	miR509.5p	−.080	.048	3.055	1	.060	.923
	Constant	953	2.467	.149	1	.099	2.594
Step 5^a	miR4487	.284	.086	10.781	1	.001	1.328
	miR4731	−.212	.072	8.549	1	.003	809
	miR509.5p	−.087	.044	3.816	1	.051	.917
	Constant	.430	2.370	.033	1	.856	1.537

Stage I/II vs Stage IV

Classification Table^a

Predicted

Cohorts

Percentage

Observed	Stage I/II	Stage IV	Correct

Step

1	Cohorts	Stage I/II	81	5	94.2
		Stage IV	10	109	91.6

Overall Percentage

92.7

	Cohorts	Stage I/II	82	4	95.3
Step 2		Stage IV	10	109	91.6

Overall Percentage

93.2

	Cohorts	Stage I/II	79	7	91.9
Step 3		Stage IV	11	108	90.8

Overall Percentage

91.2

	Cohorts	Stage I/II	79	9	90.7
Step 4		Stage IV	12	107	89.9

	Overall Percentage			90.2

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig	Exp(B)

Step 1^a	miR16	.979	.217	20.407	1	.000	2.661
	miR211	−.350	.076	21.463	1	.000	.705
	miR4487	−479	.156	9.156	1	.002	823
	miR4706	−.065	.058	.570	1	.450	.937
	miR4731	170	.101	2.851	1	.091	1.185
	miR509.3p	.905	217	17.470	1	.000	2.473
	miR509.5p	−035	065	291	1	.589	955
	Constant	−11.035	5.856	3.552	1	.059	.000
Step 2^a	miR16	.964	213	20.579	1	.000	2.623
	miR211	−.361	073	24.355	1	.000	697
	miR4487	−480	157	9.378	1	.002	619
	miR4706	−083	080	1.095	1	.295	920
	miR4731	163	099	2.718	1	.099	1.177
	miR509.3p	.900	.216	17.293	1	.000	2.450
	Constant	−10.566	5.773	3.350	1	.007	000
Step 3^a	miR16	954	210	20.857	1	.000	2.596
	miR211	−.373	073	25.848	1	.000	.089
	miR4487	−506	155	10.657	1	001	003
	miR4731	.147	096	2.347	1	.126	1.159
	miR509.3p	.888	214	17.196	1	.000	2.430
	Constant	−11.773	5.668	4.314	1	038	000
Step 4^a	miR16	882	.202	19.127	1	.000	2.417
	miR211	−.373	.071	27.606	1	.000	.689
	miR4487	−.425	145	8.765	1	003	652
	miR509.3p	.914	210	18.979	1	.000	2.496
	Constant	−9.277	5.345	3.012	1	.083	000

Stage III vs Stage IV

Classification Table^a

Predicted

Cohorts

Percentage

Observed	MIA III	Stage IV	Correct

Step

1	Cohorts	Stage III		40	10	90.0
		Stage IV	9	110	92.4

Overall Percentage

98.8

Step 2	Cohorts	Stage III		40	10	80.0
		Stage IV	8	111	93.3

Overall Percentage

89.3

Step 3	Cohorts	Stage III		40	10	90.0
		Stage IV	8	111	93.3

	Overall Percentage			89.3

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig	Exp(B)

Step 1^a	miR16	1.007	269	14.032	1	.000	2.737
	miR211	−300	076	15.503	1	000	740
	miR4487	−732	194	14.289	1	000	491
	miR4706	020	112	032	1	.859	1.020
	miR4731	387	149	6.753	1	.000	1.473
	miR509.3p	.768	247	9.094	1	.002	2.155
	miR509.5p	021	077	071	1	790	1.021
	Constant	−13.209	6.285	4.417	1	.086	000
Step 2^a	miR16	1.005	268	14.089	1	.000	2.733
	miR211	−299	075	15.840	1	000	742
	miR4487	−728	192	14.359	1	000	.493
	miR4731	391	148	6.057	1	008	1.478
	miR509.3p	764	245	9.751	1	.002	2.147
	miR509.5p	029	066	.175	1	675	1.023
	Constant	−12.880	5.997	4.613	1	032	000
Step 3^a	miR16	1.013	259	14.170	1	000	2.753
	miR211	−.294	073	15.983	1	.000	.746
	miR4487	−710	185	14.656	1	000	.492
	miR4731	386	147	6.905	1	.009	1.472
	miR509.3p	781	243	10.331	1	.001	2.184
	Constant	−12.954	6.017	4.634	1	031	000

Stage III (NSR) vs Stage III (REC)

Classification Table^a

Predicted

Cohorts

Percentage

Observeo	NSR	REC	Correct

Step

1	Cohorts	NSR	33	4	89.2
		REC	5	8	81.5

Overall Percentage

82.0

Step 2	Cohorts	NSR	32	5	86.5
		REC	5	8	81.5

Overall Percentage

80.0

Step 3	Cohorts	NSR	34	3	91.9
		REC	5	8	61.5

Overall Percentage

84.0

Step 4	Cohorts	NSR	33	4	89.2
		REC	5	7	53.8

Overall Percentage

80.0

Step 5	Cohorts	NSR	35	2	94.8
		REC	6	7	53.8

	Overall Percentage			84.0

Variables in the Equation

		B	S.E.	Wald	df	Sig	Exp(B)

Step 1^a	miR16	941	480	3.842	1	.050	2.563
	miR211	−161	.099	2.635	1	.105	852
	miR4487	−188	103	1.327	1	.249	.829
	miR4706	.027	155	.031	1	.861	1.027
	miR4731	−410	.238	2.976	1	.084	.664
	miR509.3p	.266	.436	.371	1	.542	1.304
	miR509.5p	.293	121	5.872	1	.015	1.340
	Constant	−7.428	9.701	.579	1	.447	.001
Step 2^a	miR16	.903	.423	4.547	1	.033	2.467
	miR211	.161	.099	2.650	1	.104	851
	miR4487	−.156	.163	1.294	1	.255	.830
	miR4731	−389	.209	3.654	1	.055	678
	miR509.3p	283	.428	438	1	.508	1.327
	miR509.5p	.209	.118	6.370	1	.012	1.247
	Constant	−7.190	9.731	.546	1	.460	.001
Step 3^a	miR16	.898	.418	4.661	1	.031	2.454
	miR211	−.152	.097	2.448	1	.118	.859
	miR4487	−.109	.159	1.107	1	.293	.845
	miR4731	−.354	.190	3.479	1	.062	.702
	miR509.5p	.297	.115	6.524	1	.011	1.346
	Constant	−2.508	6.746	.156	1	.693	.069
Step 4^a	miR16	.880	.417	4.457	1	.035	2.410
	miR211	−.131	.094	1.933	1	.164	.877
	miR4731	−.415	.186	4.955	1	.026	.001
	miR509.5p	250	.105	5.533	1	.019	1.285
	Constant	−1.510	6.620	.485	1	.486	.010
Step 5^a	miR16	.795	.390	4.143	1	.042	2.214
	miR4731	−.409	.182	5.067	1	.024	004
	miR509.5p	.272	.104	6.797	1	.009	1.213
	Constant	−8.295	5.915	1.982	1	.161	.000

Surival of Stage IV (<2 yr) vs Stage IV (>2 yr)

Classification Table^a

Predicted

Cohorts

	Short	Long	Percentage
Observed	Survivor	Survivor	Correct

Step

1	Cohorts	Short Survivor	57	10	85.1
		Long Survivor	19	33	63.5

Overall Percentage

75.6

Step 2	Cohorts	Short Survivor	56	11	83.6
		Long Survivor	19	33	53.5

Overall Percentage

74.8

Step 3	Cohorts	Short Survivor	55	12	82.1
		Long Survivor	19	33	63.5

Overall Percentage

73.9

Step 4	Cohorts	Short Survivor	59	8	88.1
		Long Survivor	19	33	63.5

	Overall Percentage			77.3

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig.	Exp(B)

Step 1^a	miR16	−101	.191	.279	1	.597	.904
	miR211	.230	.056	16.789	1	.000	1.259
	miR4487	071	134	.282	1	.595	1.074
	miR4706	.255	.088	8.439	1	004	1.290
	miR4731	.097	.083	1.356	1	244	1.102
	miR509.3p	−008	039	048	1	826	.992
	miR509.5p	−113	051	4.978	1	026	.893
	Constant	−15.131	7.002	4.590	1	032	.000
Step 2^a	miR16	−.089	183	.239	1	625	915
	miR211	228	.053	18.199	1	000	1.254
	miR4487	.071	134	.285	1	593	1.074
	miR4706	258	088	8.541	1	003	1.292
	miR4731	.094	.082	1.303	1	.254	1.099
	miR509.5p	−115	.050	5.163	1	023	892
	Constant	−15.338	6.997	4.904	1	028	.000
Step 3^a	miR211	.228	053	18.065	1	000	1.254
	miR4487	098	123	634	1	426	1.103
	miR4706	257	088	5.601	1	003	1.293
	miR4731	115	.071	2.610	1	108	1.122
	miR509.5p	−.112	.050	5.015	1	025	.894
	Constant	−17.804	4.762	14.122	1	000	.000
Step 4^a	miR211	219	052	17.917	1	000	1.244
	miR4706	.257	087	8.830	1	003	1.293
	miR4731	142	.063	5.026	1	.025	1.152
	miR509.5p	−109	.050	4.859	1	.028	.897
	Constant	−15.991	4.029	15.756	1	000	000

M Stage Prediction

Classification Table^a

Predicted

Cohorts

Percentage

Observed	M1a	M1c	Correct

Step

1	Cohorts	M1a		14	0	100.0
		M1c	0	16	100.0

	Overall Percentage			100.0

^aThe cut value is .500

Variables in the Equation

		B	S.E	wald	df	Sig	Exp(B)

Step 1^a	miR16	−88.890	3906.283	.001	1	.980	000
	miR211	11.432	435.925	001	1	979	92223.759
	miR4487	195.085	6137.855	001	1	975	5.299E+84
	miR4706	−157.880	5104.067	.001	1	975	000
	miR4731	−97.471	3095.981	001	1	975	000
	miR509.3p	−37.669	1191.835	001	1	975	.000
	miR509.5p	−17.797	614.043	.001	1	977	000
	Constant	6098.464	1.971E+05	.001	1	975

Classification Table^a

Predicted

Cohorts

Percentage

Observed	M1a	M1b	Correct

Step

1	Cohorts	M1a		14	0	100.0
		M1b	2	4	66.7

Overall Percentage

90.0

Step 2	Cohorts	M1a		14	0	100.0
		M1b	2	4	66.7

Overall Percentage

90.0

Step 3	Cohorts	M1a		13	1	92.9
		M1b	1	5	53.3

	Overall Percentage			90.0

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig	Exp(B)

Step 1^a	miR16	−.721	1.078	.448	1	.503	486
	miR211	145	220	435	1	510	1.156
	miR4487	1.157	.854	1.835	1	176	3.182
	miR4706	−.790	.527	2.249	1	.134	.454
	miR4731	−.490	.354	1.918	1	.166	.613
	miR509.3p	−270	.215	1.576	1	.209	.764
	miR509.5p	−.010	.213	.002	1	.962	.690
	Constant	25.245	20.936	.711	1	.399
Step 2^a	miR16	−.693	.888	.609	1	.435	.500
	miR211	.145	.219	.435	1	.508	1.150
	miR4487	1.165	.847	1.892	1	.169	3.205
	miR4706	−.801	.481	2.776	1	.096	.449
	miR4731	−.487	.350	1.941	1	.164	.614
	miR509.3p	−.267	.205	1.697	1	.193	.766
	Constant	24.617	25.819	.843	1	.359
Step 3^a	miR16	−.935	.878	1.133	1	.287	.393
	miR4487	.910	.584	1.769	1	.183	2.485
	miR4706	−.728	.425	2.909	1	.058	.484
	miR4731	−.507	.344	2.174	1	.140	.602
	miR509.3p	−221	.155	2.048	1	.152	.801
	Constant	34.560	25.078	1.899	1	.108

Classification Table^a

Predicted

Cohorts

Percentage

Observed	M1b	M1c	Correct

Step

1	Cohorts	M1b	4	2	66.7
		M1c	1	15	93.8

Overall Percentage

86.4

Step 2	Cohorts	M1b	4	2	66.7
		M1c	1	15	93.8

Overall Percentage

86.4

Step 3	Cohorts	M1b	4	2	66.7
		M1c	1	15	93.8

Overall Percentage

86.4

Step 4	Cohorts	M1b	2	4	33.3
		M1c	1	15	93.8

Overall Percentage

77.3

Step 5	Cohorts	M1b	2	4	33.3
		M1c	2	14	87.5

Overall Percentage

72.7

Step 6	Cohorts	M1b	2	4	33.3
		M1c	1	15	93.8

Overall Percentage

77.3

Step 7	Cohorts	M1b	1	5	16.7
		M1c	0	16	100.0

Overall Percentage

77.3

Step 8	Cohorts	M1b	0	6	0.0
		M1c	0	16	100.0

	Overall Percentage			72.7

^aThe cut value is .500

Variables in the Equation

		B	S.E.	Wald	df	Sig	Exp(B)

Step 1^a	miR16	1.034	1.230	697	1	.404	2.813
	miR211	−.191	.133	2.055	1	.152	.826
	miR4487	.385	.419	.854	1	.355	1.473
	miR4706	.039	.220	.031	1	.851	1.030
	miR4731	−.377	.291	1.075	1	.196	.686
	miR509.3p	−103	.122	.707	1	.401	.903
	miR509.5p	.008	.140	.003	1	.053	1.008
	Constant	−2.212	16.670	.015	1	.894	.109
Step 2^a	miR16	1.014	1.189	.727	1	.394	2.757
	miR211	−188	.121	2.393	1	.122	.829
	miR4487	.390	.419	.865	1	.352	1.477
	miR4706	.043	.208	.044	1	.834	1.044
	miR4731	−376	.292	1.059	1	.198	.086
	miR509.3p	−102	.122	.702	1	.402	.905
	Constant	−2.079	10.453	.015	1	.899	.125
Step 3^a	miR16	.911	1.062	.736	1	.391	2.487
	miR211	−.181	.115	2.458	1	.117	.835
	miR4487	.415	.402	1.064	1	.302	1.514
	miR4731	−.392	.281	1.948	1	.163	.076
	miR509.3p	−.098	.120	.869	1	.413	.907
	Constant	.037	13.133	.000	1	.998	1.038
Step 4^a	miR16	.002	609	.553	1	.457	1.825
	miR211	−.202	115	3.055	1	.080	.817
	miR4487	.349	.360	.943	1	.332	1.416
	miR4731	−.318	.232	1.883	1	.170	.727
	Constant	.193	12.401	.000	1	988	1.213
Step 5^a	miR211	−160	.094	2.905	1	.066	.852
	miR4487	.255	.332	.591	1	.442	1.290
	miR4731	−.314	.215	2.127	1	145	.731
	Constant	8.004	8.765	.826	1	.361	2891.756
Step 6^a	miR211	−.149	.091	2.707	1	.100	.861
	miR4731	−.233	.186	1.532	1	.216	.782
	Constant	12.449	7.167	3.140	1	.082
Step 7^a	miR211	−128	.089	2.087	1	148	.880
	Constant	4.287	2.418	3.140	1	.076	72.747
Step 8^a	Constant	.981	.479	4.198	1	.040	2.667

^aVariable(s) entered on step 1: mir16, miR211, miR4487, miR4706, miR4731, miR509.3p, miR509.5p

EXAMPLE 2

Materials and Methods

Cell Culture and Total RNA extraction. All melanoma (cutaneous and uveal) cell lines (Table 8) were established and have been previously described63, 64. All other solid cancer cell lines (Table 8) were kind donations from investigators at QIMR Berghofer. Most of the solid cancer cell lines are available from the cell line repositories ATCC or CellBank Australia.
All cell lines were cultured in RPMI (#31800-089, Life Technologies, Foster City, Calif., USA) supplemented with 10% FBS (Life Technologies, Foster City, Calif., USA) with HEPES, 100 U/ml penicillin and 100 μg/ml streptomycin (Life Technologies, Foster City, Calif., USA) at 37° C. (5% CO2) and were periodically authenticated via short tandem repeat profiling according to the manufacturer's instructions (AmpFISTR Profiler Plus ID kit; Life Technologies. Foster City, Calif., USA). Primary human melanoblasts (QF1160MB) and melanocytes (MELA) were established from human neonatal foreskin and cultured as described^{65, 66}. Cells were harvested from the plate and column-purified using the miRNeasy Kit (QIAGEN, Hilden, Germany) according to the manufacturer's instructions.
Serum/Plasma collection and Total RNA extraction. Ethical approval was granted by the QIMR Berghofer's Human Research Ethics Committee (HREC) approval number P1237. Serum and Plasma were processed using standard methodologies and total RNA was extracted using the Plasma/Serum Circulating RNA Purification Kit (#30000; Norgen Biotek, Ontario, Canada) according to manufacturer's instructions.
microRNA microarray profiling and data analysis. 5 μg of total RNA was shipped to LC Sciences (Houston, USA) for miRNA profiling using a custom array platform (μParaflo® technology) containing 1898 miRNAs (miRBase V18)(67). All QC, labelling (Cy-3), hybridization, scanning, signal background subtraction and global normalisation (LOWES) were performed by LC Sciences as per technical note: (www.lcsciences.com/documents/application-notes/Tech-Bull-MicroRNA-Microarmy-Data-Analysis.pdf).
Advanced data analyses were performed in Genespring GX12.5 (Agilent Technologies, Santa Clara, USA) using the LOWES normalised signal intensity values. All values <30 were considered ‘background expression’ (personal communication with LC Sciences) and changed to 0.01 prior to log₂transformation. So as to identify ‘melanoma-specific’ miRNAs that were potentially more relevant to melanoma, samples were classified as either ‘melanoma’ or ‘other ca’ (melanocytes, melanoblasts. nevocyte, and serum derived samples were excluded from these categories). To identify differentially expressed miRNAs, a Mann-Whitney U-test (unpaired) was applied to a ‘volcano-plot’ analysis with thresholds set at p<0.05 and ≧2 fold. The gene list derived from these analyses was then applied to all samples in an unsupervised manner using hierarchical clustering (Euclidean similarity with average linkage).
miScript quantitative RT-PCR validation. A selection of miRNAs was validated using quantitative real-time-PCR to check the integrity of the microarray data. Briefly, all samples included on the microarray along with an extended cohort of melanoma cell lines (as described in ref 54) were reverse transcribed using the miScript II RT Kit (QIAGEN, Hilden, Germany) according to the manufacturer's instructions. Real-time PCR was subsequently performed with a miScript SYBR Green PCR Kit (QIAGEN, Hilden, Germany) using the 7900HT Fast Real Time PCR System (Life Technologies, Foster City, Calif., USA). Data were analysed in Microsoft Excel using the ΔCT method compared to RNU6 which was assessed in every sample.

Results

‘Melanoma-specific’ miRNA discovery profiling in melanoma. This comprehensive analysis of a large panel of melanoma cell lines identified distinct ‘tissue-specific’ expression as compared to other solid malignancies. We used miRNA microarrays (LC Sciences, Houston, USA. miRBaseV18; 1898 miRNAs) to comprehensively study the miRNA profile of cutaneous melanoma (n=55), uveal melanoma (n=7), and RNA derived from melanoma patient scrum/plasma (n=3) in relation to ‘other’ solid malignancies (n=34). Normal pigment (melanocytes and melanoblasts) and pre-malignant (nevocyte) cells were also included as controls (‘Discovery cohort’). In this ‘Discovery’ set, a total of 233/1898 differentially expressed (≧2 fold, p<0.05; See Materials and Methods) miRNAs were identified which may fit this criteria when ‘melanomas’ were compared to ‘other’ cancers (FIG. 8 and Table 7). We identified a significant number of miRNAs (Table 8) with known and novel relationship to melanoma. Table 9 summarizes the top up and down regulated miRNAs by at least 10 fold, most of which have a known association either with melanoma or tumourigenesis.
‘Lineage-specific’ miRNA validation. Notably the top two miRNAs up-regulated in melanoma were hsa-miR-211-5p (miR-211) and hsa-miR-514a-3p (miR-514a) with profound average fold changes of 276 and 204 respectively compared to other solid malignancies (Table 8). To test the robustness of the array data, quantitative real-time PCR validation was performed with these miRNAs along with a selection of other miRNAs from the array (FIG. 9 and Table 10).

Discussion

MicroRNAs have been proven to be powerful regulators of protein coding gene expression in almost all facets of cell biology. In cancer, their dysregulation can be far reaching, spanning a multitude of different signalling pathways due to the number of genes being directly and indirectly regulated and as such, they have been intensively studied for over a decade. In melanoma, the comprehensive analysis of miRNA has been limited, with most studies performed using only a few hundred miRNAs²⁸. Newly characterised miRNAs have seen the expansion of miRBASE (curated miRNA database), largely due to the advent of deep sequencing technologies. As most of the newly identified miRNAs have no known association with melanoma, here we sort to survey a large panel of melanoma cell lines in relation to other solid malignancies to strengthen prior studies into melanoma-specific miRNAs and to identify those that were previously unknown to be involved in melanomagenesis. Identification of a panel in a comprehensive manner such as this has the ability to open new avenues for research in this burgeoning field, which may provide a greater understanding of this complex disease. In our comprehensive analysis of 1898 miRNAs (miRBaseV18), we have successfully identified a large number of miRNAs (n=233) that were found to be statistically significant (P<0.05, ≧2 fold) when melanoma (n=55) was compared with other solid cancer (n=34). Indeed, many of the identified miRNAs have a known association with melanoma however most have no known association. The main premise of the study was to identify a panel of melanoma-specific miRNAs (or more predominantly expressed) and these are more readily observed in the up-regulated collection of miRNAs. However it is worth noting that many of the down-regulated miRNAs have been previously associated with melanoma. Our dataset confirms and strengthens the associated loss of the miR-200 family (including miR-141)^{47, 48}along with the frequently silenced miRNA, miR-205^49-53.

TABLE 7

Differentially expressed miRNAs in melanoma

GeneName	p (Corr)	p	FC (abs)	Regulation

hsa-miR-200c-3p	1.29E−06	2.24E−08	250.7467	down
hsa-miR-204-5p	5.50E−06	1.39E−07	93.67072	up
hsa-miR-145-5p	1.31E−04	6.27E−06	54.31673	up
hsa-miR-211-5p	3.43E−07	3.97E−09	276.8409	up
hsa-miR-363-3p	0.001081	8.32E−05	17.00228	up
hsa-miR-510	0.001869	1.69E−04	11.33646	up
hsa-miR-514a-5p	0.004085	4.44E−04	8.814804	up
hsa-miR-509-3-5p	3.46E−06	7.48E−08	55.9232	up
hsa-miR-509-3p	5.15E−07	6.24E−09	139.3508	up
hsa-miR-514a-3p	2.96E−07	3.12E−09	204.0894	up
hsa-miR-506-3p	1.47E−04	7.38E−06	27.69136	up
hsa-miR-509-5p	2.29E−06	4.59E−08	65.2407	up
hsa-miR-513c-5p	1.56E−04	7.97E−06	26.93033	up
hsa-miR-508-3p	7.57E−05	3.03E−06	31.90864	up
hsa-miR-513b	1.08E−05	3.13E−07	61.25974	up
hsa-miR-205-5p	0.001084	8.52E−05	19.47374	down
hsa-miR-200a-5p	4.47E−06	1.06E−07	31.4116	down
hsa-miR-200c-5p	2.14E−04	1.17E−05	13.61658	down
hsa-miR-141-3p	1.08E−08	5.70E−11	206.6276	down
hsa-miR-203	2.10E−08	1.44E−10	273.6629	down
hsa-miR-200b-3p	3.83E−10	1.01E−12	1717.021	down
hsa-miR-200a-3p	6.01E−10	1.90E−12	680.575	down
hsa-miR-429	3.83E−10	9.84E−13	515.7665	down
hsa-miR-183-5p	3.29E−11	1.73E−14	220.7499	down
hsa-miR-96-5p	1.73E−08	1.10E−10	120.4834	down
hsa-miR-299-3p	0.001145	9.29E−05	10.29763	down
hsa-miR-183-3p	4.11E−09	1.73E−11	101.0339	down
hsa-miR-375	1.84E−05	6.29E−07	34.56165	down
hsa-miR-335-3p	9.59E−07	1.36E−08	40.52689	down
hsa-miR-335-5p	2.00E−07	1.89E−09	138.4271	down
hsa-miR-27b-5p	0.002927	2.98E−04	9.261446	down
hsa-miR-4520a-3p	0.022563	0.003721	6.89	down
hsa-miR-224-5p	1.05E−06	1.66E−08	93.71726	down
hsa-miR-452-5p	3.02E−06	6.36E−08	28.70594	down
hsa-miR-2115-3p	1.64E−04	8.45E−06	13.62161	down
hsa-miR-610	0.006736	8.06E−04	6.325314	down
hsa-miR-4474-5p	0.004085	4.46E−04	7.169808	down
hsa-miR-4768-3p	3.71E−04	2.23E−05	11.11902	down
hsa-miR-3120-3p	9.80E−07	1.50E−08	25.58355	down
hsa-miR-3131	0.002805	2.81E−04	8.680573	down
hsa-miR-3064-5p	0.03839	0.006966	4.001939	down
hsa-miR-3146	0.001636	1.43E−04	9.022174	down
hsa-miR-29b-2-5p	4.74E−06	1.17E−07	24.81251	down
hsa-miR-23b-5p	0.010427	0.001494	5.084345	down
hsa-miR-29c-5p	0.018817	0.002954	4.759145	down
hsa-miR-181c-3p	0.003249	3.41E−04	6.808747	down
hsa-miR-934	0.004177	4.58E−04	5.713632	down
hsa-miR-224-3p	8.00E−04	5.69E−05	6.372761	down
hsa-miR-3664-3p	0.046873	0.008718	2.658377	down
hsa-miR-182-3p	0.004453	4.97E−04	3.530612	down
hsa-miR-26b-3p	0.007891	9.94E−04	3.759939	down
hsa-miR-1224-3p	0.02713	0.004603	3.188798	down
hsa-miR-4639-3p	0.046873	0.008718	2.577534	down
hsa-miR-136-3p	0.009735	0.001349	3.622702	down
hsa-miR-889	0.022175	0.003622	2.626288	down
hsa-miR-146a-3p	0.006231	7.35E−04	8.144401	up
hsa-miR-551b-3p	0.002793	2.75E−04	13.10246	up
hsa-miR-4436b-5p	0.029436	0.005087	5.61163	up
hsa-miR-4655-3p	0.007966	0.001032	9.83113	up
hsa-miR-3909	0.001779	1.60E−04	9.591683	up
hsa-miR-4469	0.026951	0.004558	4.714384	up
hsa-miR-199b-5p	0.010553	0.001518	6.022791	up
hsa-miR-20b-3p	0.030173	0.005262	4.286351	up
hsa-miR-3189-5p	0.010122	0.001419	6.377651	up
hsa-miR-3944-3p	0.010203	0.001441	6.720376	up
hsa-miR-3972	0.002615	2.54E−04	7.432085	up
hsa-miR-4654	0.047201	0.008845	2.915873	up
hsa-miR-3677-5p	0.046873	0.008704	3.970655	up
hsa-miR-4785	0.033194	0.005929	3.763893	up
hsa-miR-4665-5p	0.047031	0.008772	4.908286	up
hsa-miR-646	0.028359	0.004856	3.96194	up
hsa-miR-4722-3p	0.006789	8.18E−04	7.179308	up
hsa-miR-4731-5p	1.47E−04	7.29E−06	24.1787	up
hsa-miR-31-3p	0.001084	8.62E−05	22.97718	down
hsa-miR-31-5p	0.003738	4.00E−04	7.086216	down
hsa-miR-5701	0.008838	0.001197	10.01555	up
hsa-miR-4781-5p	0.014003	0.002132	5.305426	up
hsa-miR-548q	0.004951	5.61E−04	5.547026	up
hsa-miR-4708-3p	0.001086	8.76E−05	12.24264	up
hsa-miR-654-5p	0.009938	0.001386	7.699157	up
hsa-miR-658	0.010971	0.001607	6.930552	up
hsa-miR-4719	0.007075	8.69E−04	11.2437	down
hsa-miR-4539	0.021122	0.003405	6.615453	up
hsa-miR-4321	0.008674	0.00117	8.399978	up
hsa-miR-92a-1-5p	0.032566	0.005782	5.652854	down
hsa-let-7a-3p	0.031593	0.005576	4.507036	down
hsa-miR-16-1-3p	0.010709	0.001557	7.805436	down
hsa-miR-3127-5p	0.021434	0.003479	4.025753	down
hsa-miR-149-5p	1.47E−04	7.24E−06	7.747861	down
hsa-miR-103b	0.013613	0.002044	5.532397	down
hsa-miR-1306-3p	4.41E−04	2.74E−05	5.499783	down
hsa-miR-4277	0.019562	0.003082	4.243334	down
hsa-miR-7-1-3p	0.013779	0.002091	7.00258	down
hsa-miR-192-5p	0.008451	0.001122	5.301906	down
hsa-miR-194-5p	0.010414	0.001487	6.899473	down
hsa-miR-625-3p	6.00E−04	3.98E−05	5.362473	down
hsa-miR-1244	1.30E−05	4.04E−07	13.87607	down
hsa-miR-3609	0.00615	7.19E−04	5.480611	down
hsa-miR-374a-5p	0.005488	6.33E−04	5.316285	down
hsa-miR-374b-5p	7.74E−04	5.45E−05	4.971394	down
hsa-miR-374c-5p	0.002805	2.80E−04	6.082301	down
hsa-miR-4644	7.74E−04	5.47E−05	5.508903	down
hsa-miR-3148	2.36E−04	1.33E−05	3.753175	down
hsa-miR-670	1.31E−04	6.20E−06	2.954435	down
hsa-miR-3125	1.56E−05	5.17E−07	10.45982	down
hsa-miR-3682-3p	8.17E−05	3.31E−06	13.37263	down
hsa-miR-617	1.49E−05	4.87E−07	10.12681	down
hsa-miR-196b-3p	0.035312	0.006363	2.420995	down
hsa-miR-631	0.047798	0.008991	2.766758	down
hsa-miR-4712-3p	0.008318	0.001099	4.874474	down
hsa-miR-1183	4.14E−06	9.60E−08	13.50259	down
hsa-miR-1305	1.48E−05	4.76E−07	8.817349	down
hsa-miR-3616-3p	0.01974	0.00312	3.940568	down
hsa-miR-4750	0.007443	9.29E−04	4.045383	down
hsa-miR-622	5.67E−06	1.46E−07	6.138279	down
hsa-miR-1914-5p	0.0043	4.75E−04	4.031873	up
hsa-miR-557	0.002559	2.45E−04	3.978653	up
hsa-miR-3937	7.27E−05	2.87E−06	8.556762	up
hsa-miR-612	0.018014	0.002819	2.968695	up
hsa-miR-498	0.01123	0.001651	5.406655	up
hsa-miR-596	3.88E−04	2.39E−05	4.02502	up
hsa-miR-637	5.32E−04	3.42E−05	3.097873	up
hsa-miR-1914-3p	5.84E−04	3.84E−05	3.166935	up
hsa-miR-4783-5p	6.75E−04	4.59E−05	2.468174	up
hsa-miR-593-5p	4.55E−05	1.70E−06	7.19713	up
hsa-miR-138-5p	0.001269	1.06E−04	7.184394	up
hsa-miR-146a-5p	3.50E−10	5.54E−13	36.32244	up
hsa-miR-4468	0.007443	9.29E−04	3.498524	down
hsa-miR-4435	0.001689	1.49E−04	3.713703	down
hsa-miR-4675	0.011764	0.001742	3.270106	down
hsa-miR-4303	7.50E−04	5.18E−05	4.217161	down
hsa-miR-200b-5p	9.31E−04	6.99E−05	4.466938	down
hsa-miR-505-3p	0.030783	0.005417	2.509697	down
hsa-miR-4689	0.002805	2.80E−04	2.843713	down
hsa-miR-182-5p	1.59E−10	1.67E−13	16.29153	down
hsa-miR-3126-3p	0.005012	5.70E−04	2.052046	down
hsa-miR-134	0.010971	0.001605	2.169331	down
hsa-miR-135b-3p	0.021434	0.00349	2.782404	down
hsa-miR-15b-3p	3.97E−05	1.40E−06	4.796393	down
hsa-miR-16-2-3p	0.005456	6.27E−04	3.170327	down
hsa-miR-7-5p	5.58E−08	4.12E−10	8.046594	down
hsa-miR-98	9.47E−04	7.18E−05	2.361425	down
hsa-miR-23c	4.38E−05	1.61E−06	3.371118	down
hsa-miR-454-3p	0.002884	2.92E−04	2.117273	down
hsa-miR-331-3p	9.31E−04	6.97E−05	2.204815	down
hsa-miR-5011-5p	2.97E−04	1.72E−05	2.12799	down
hsa-miR-26b-5p	2.04E−06	3.86E−08	2.913235	down
hsa-miR-27b-3p	0.001741	1.55E−04	2.012356	down
hsa-miR-4753-5p	0.004079	4.41E−04	2.579822	down
hsa-miR-3667-5p	2.33E−04	1.30E−05	3.667121	down
hsa-miR-601	1.18E−05	3.53E−07	3.796099	down
hsa-miR-595	0.001086	8.75E−05	2.123006	down
hsa-miR-3149	1.18E−04	5.33E−06	2.016392	down
hsa-miR-568	6.07E−04	4.06E−05	2.588553	down
hsa-miR-4700-5p	1.30E−04	5.95E−06	3.067645	down
hsa-miR-4701-3p	9.92E−05	4.18E−06	2.177509	down
hsa-miR-1273f	0.0043	4.76E−04	2.661917	down
hsa-miR-3156-5p	0.029612	0.005149	2.818424	down
hsa-miR-4489	0.008561	0.001149	2.226271	down
hsa-miR-1538	0.022286	0.003652	2.002481	up
hsa-miR-602	4.35E−05	1.56E−06	3.929173	up
hsa-miR-152	0.001604	1.39E−04	2.903808	up
hsa-miR-361-3p	0.004367	4.85E−04	2.0819	up
hsa-miR-660-5p	4.69E−04	2.94E−05	3.816789	up
hsa-miR-532-3p	8.07E−04	5.87E−05	2.41832	up
hsa-miR-532-5p	0.001258	1.04E−04	2.049957	up
hsa-miR-500a-5p	1.92E−04	1.01E−05	2.592344	up
hsa-miR-502-3p	2.08E−06	4.05E−08	2.867455	up
hsa-miR-500a-3p	2.81E−06	5.78E−08	3.288926	up
hsa-miR-500b	4.14E−06	9.15E−08	3.306766	up
hsa-miR-146b-5p	9.66E−07	1.42E−08	6.577508	up
hsa-miR-4758-3p	0.007443	9.27E−04	2.140459	up
hsa-miR-572	0.007836	9.83E−04	2.122475	up
hsa-miR-4707-3p	2.78E−04	1.58E−05	2.890006	up
hsa-miR-4467	1.45E−05	4.59E−07	3.269783	up
hsa-miR-211-3p	0.01656	0.002572	2.137208	up
hsa-miR-514b-5p	4.14E−06	9.46E−08	2.451853	up
hsa-miR-584-5p	2.00E−06	3.68E−08	4.172919	up
hsa-miR-140-3p	2.94E−04	1.69E−05	2.380397	up
hsa-miR-185-5p	1.18E−05	3.55E−07	2.566472	up
hsa-miR-4306	9.11E−06	2.59E−07	2.30042	up
hsa-miR-33b-3p	0.001779	1.60E−04	2.029654	up
hsa-miR-718	0.001673	1.47E−04	2.019544	up
hsa-miR-513a-5p	1.57E−05	5.28E−07	2.5497	up
hsa-miR-3180	1.03E−04	4.44E−06	2.207144	up
hsa-miR-4634	7.24E−08	5.72E−10	2.783406	up
hsa-miR-187-5p	4.38E−05	1.62E−06	2.040338	up
hsa-miR-3195	8.82E−10	3.25E−12	3.056716	up
hsa-miR-4532	1.08E−08	5.66E−11	2.402943	up
hsa-miR-4497	3.24E−07	3.58E−09	2.42442	up
hsa-miR-3196	1.22E−08	7.09E−11	2.634303	up
hsa-miR-4488	2.77E−07	2.77E−09	2.560814	up
hsa-miR-3940-5p	7.03E−06	1.89E−07	2.071707	up
hsa-miR-4466	4.49E−06	1.09E−07	2.079208	up
hsa-miR-4707-5p	7.75E−06	2.16E−07	2.062567	up
hsa-miR-3178	1.96E−06	3.51E−08	3.343691	up
hsa-miR-1469	1.21E−05	3.71E−07	2.23717	up
hsa-miR-663a	1.75E−07	1.49E−09	2.543887	up
hsa-miR-4440	0.013465	0.002015	2.267684	up
hsa-miR-4487	0.007933	0.001024	2.725489	up
hsa-miR-3907	7.04E−06	1.93E−07	6.43762	up
hsa-miR-501-3p	6.88E−07	9.19E−09	6.462504	up
hsa-miR-5587-5p	3.80E−04	2.30E−05	5.271441	up
hsa-miR-4706	0.029612	0.005149	2.311177	up
hsa-miR-5008-5p	0.011546	0.001703	3.408319	up
hsa-miR-628-3p	0.02025	0.003233	4.679494	up
hsa-miR-3184-5p	0.001423	1.23E−04	8.294511	up
hsa-miR-3194-3p	0.002812	2.83E−04	4.127772	up
hsa-miR-3619-3p	0.008931	0.001214	4.414144	up
hsa-miR-508-5p	6.88E−07	9.42E−09	21.59629	up
hsa-miR-218-2-3p	7.74E−04	5.43E−05	7.18456	up
hsa-miR-4323	9.31E−04	7.02E−05	4.921013	up
hsa-miR-636	0.032688	0.005821	2.65164	up
hsa-miR-3944-5p	0.001908	1.74E−04	4.912326	up
hsa-miR-4665-3p	5.11E−04	3.26E−05	4.010174	up
hsa-miR-181a-3p	0.022309	0.003667	3.537355	up
hsa-miR-30b-3p	0.047798	0.009016	2.869683	up
hsa-miR-362-3p	2.02E−04	1.08E−05	6.514883	up
hsa-miR-501-5p	0.002222	2.07E−04	5.113965	up
hsa-miR-140-5p	0.002789	2.73E−04	7.660998	up
hsa-miR-584-3p	9.67E−05	4.02E−06	10.36175	up
hsa-miR-10a-3p	0.001258	1.04E−04	15.86066	down
hsa-miR-196b-5p	1.03E−04	4.44E−06	23.30624	down
hsa-miR-654-3p	0.040924	0.007482	5.409095	down
hsa-miR-3713	0.006907	8.44E−04	9.587804	down
hsa-miR-3170	5.75E−04	3.75E−05	8.304263	down
hsa-miR-4672	1.31E−04	6.27E−06	22.65654	down
hsa-miR-215	3.54E−04	2.11E−05	30.10806	down
hsa-miR-206	3.85E−04	2.35E−05	18.04436	down
hsa-miR-539-5p	1.20E−06	2.03E−08	43.96934	down
hsa-miR-4311	0.010702	0.001551	5.535323	down
hsa-miR-2278	0.001325	1.12E−04	8.703324	down
hsa-miR-297	0.010271	0.001461	9.661793	down

TABLE 9

The top up and down regulated miRNAs by at least 10 fold in melanoma

				Reg-
GeneName	p (Corr)	p	FC (abs)	ulation

hsa-miR-211-5p	3.43E−07	3.97E−09	276.84088	up
hsa-miR-514a-	2.96E−07	3.12E−09	204.08942	up
3p
hsa-miR-509-3p	5.15E−07	6.24E−09	139.35081	up
hsa-miR-204-5p	5.50E−06	1.39E−07	93.670715	up
hsa-miR-509-5p	2.29E−06	4.59E−08	65.2407	up
hsa-miR-513b	1.08E−05	3.13E−07	61.259735	up
hsa-miR-509-3-	3.46E−06	7.48E−08	55.923203	up
5p
hsa-miR-145-5p	1.31E−04	6.27E−06	54.31673	up
hsa-miR-146a-5p	3.50E−10	5.54E−13	36.32244	up
hsa-miR-508-3p	7.57E−05	3.03E−06	31.908644	up
hsa-miR-506-3p	1.47E−04	7.38E−06	27.691364	up
hsa-miR-513c-	1.56E−04	7.97E−06	26.930325	up
5p
hsa-miR-4731-	1.47E−04	7.29E−06	24.178703	up
5p
hsa-miR-508-5p	6.88E−07	9.42E−09	21.59629	up
hsa-miR-363-3p	0.001080965	8.32E−05	17.002277	up
hsa-miR-551b-	0.002793076	2.75E−04	13.102462	up
3p
hsa-miR-4708-	0.001086084	8.76E−05	12.242638	up
3p
hsa-miR-510	0.001869193	1.69E−04	11.336461	up
hsa-miR-584-3p	9.67E−05	4.02E−06	10.361748	up
hsa-miR-5701	0.008838477	0.00119678	10.0155525	up
hsa-miR-617	1.49E−05	4.87E−07	10.126807	down
hsa-miR-299-3p	0.001144688	9.29E−05	10.297629	down
hsa-miR-3125	1.56E−05	5.17E−07	10.459817	down
hsa-miR-4768-	3.71E−04	2.23E−05	11.119021	down
3p
hsa-miR-4719	0.00707513	8.69E−04	11.243696	down
hsa-miR-3682-	8.17E−05	3.31E−06	13.372626	down
3p
hsa-miR-1183	4.14E−06	9.60E−08	13.502589	down
hsa-miR-200c-5p	2.14E−04	1.17E−05	13.616578	down
hsa-miR-2115-	1.64E−04	8.45E−06	13.621608	down
3p
hsa-miR-1244	1.30E−05	4.04E−07	13.876072	down
hsa-miR-10a-3p	0.001257804	1.04E−04	15.860663	down
hsa-miR-182-5p	1.59E−10	1.67E−13	16.291529	down
hsa-miR-206	3.85E−04	2.35E−05	18.044355	down
hsa-miR-205-5p	0.001083732	8.52E−05	19.473736	down
hsa-miR-4672	1.31E−04	6.27E−06	22.656536	down
hsa-miR-31-3p	0.001083732	8.62E−05	22.977182	down
hsa-miR-196b-	1.03E−04	4.44E−06	23.306242	down
5p
hsa-miR-29b-2-	4.74E−06	1.17E−07	24.812508	down
5p
hsa-miR-3120-	9.80E−07	1.50E−08	25.58355	down
3p
hsa-miR-452-5p	3.02E−06	6.36E−08	28.705942	down
hsa-miR-215	3.54E−04	2.11E−05	30.10806	down
hsa-miR-200a-5p	4.47E−06	1.06E−07	31.411602	down
hsa-miR-375	1.84E−05	6.29E−07	34.56165	down
hsa-miR-335-3p	9.59E−07	1.36E−08	40.526886	down
hsa-miR-539-5p	1.20E−06	2.03E−08	43.96934	down
hsa-miR-224-5p	1.05E−06	1.66E−08	93.71726	down
hsa-miR-183-3p	4.11E−09	1.73E−11	101.033875	down
hsa-miR-96-5p	1.73E−08	1.10E−10	120.48337	down
hsa-miR-335-5p	2.00E−07	1.89E−09	138.42706	down
hsa-miR-141-3p	1.08E−08	5.70E−11	206.62756	down
hsa-miR-183-5p	3.29E−11	1.73E−14	220.74986	down
hsa-miR-200c-3p	1.29E−06	2.24E−08	250.74672	down
hsa-miR-203	2.10E−08	1.44E−10	273.66293	down
hsa-miR-429	3.83E−10	9.84E−13	515.76654	down
hsa-miR-200a-3p	6.01E−10	1.90E−12	680.57495	down
hsa-miR-200b-	3.83E−10	1.01E−12	1717.0214	down
3p

TABLE 10

Cell lines in which quantitative real-time PCR validation was performed

			qRT PCR
Sample ID	Type	Cohort	validation

QF1160MB	Melanoblast	Discovery	Yes
MELA	Melanocyte	Discovery	Yes
MM653	Nevus	Discovery	Yes
22rV-1	Other	Discovery	Yes
786-O	Other	Discovery	Yes
ALVA-1	Other	Discovery	Yes
BT474	Other	Discovery	Yes
BxPC-3	Other	Discovery	Yes
CAKI-1	Other	Discovery	Yes
CAKI-2	Other	Discovery	Yes
CaOV-3	Other	Discovery	Yes
CAPAN-1	Other	Discovery	Yes
CAPAN-2	Other	Discovery	Yes
Co115	Other	Discovery	Yes
Du145	Other	Discovery	Yes
HT29	Other	Discovery	Yes
LIM1899	Other	Discovery	Yes
LIM2405	Other	Discovery	Yes
LN-18	Other	Discovery	Yes
LNCAP	Other	Discovery	Yes
MCF-7	Other	Discovery	Yes
MDAMB231	Other	Discovery	Yes
OAW28	Other	Discovery	Yes
OVCAR-3	Other	Discovery	Yes
PANC-1	Other	Discovery	Yes
PC-3	Other	Discovery	Yes
PE04	Other	Discovery	Yes
PL45	Other	Discovery	Yes
SKBR3	Other	Discovery	Yes
SK-OV-3	Other	Discovery	Yes
SN12K-1	Other	Discovery	Yes
SW48	Other	Discovery	Yes
SW839	Other	Discovery	Yes
T46	Other	Discovery	Yes
T47D	Other	Discovery	Yes
T50	Other	Discovery	Yes
U118	Other	Discovery	Yes
MM200	STAGE I/II	Validation	Yes
MM229	STAGE I/II	Validation	Yes
MM329	STAGE I/II	Validation	Yes
MM540	STAGE I/II	Validation	Yes
C001	STAGE_III_MM	Discovery	Yes
C002	STAGE_III_MM	Discovery	Yes
C003	STAGE_III_MM	Discovery	Yes
C004	STAGE_III_MM	Discovery	Yes
C006	STAGE_III_MM	Discovery	Yes
C008	STAGE_III_MM	Discovery	Yes
C011	STAGE_III_MM	Discovery	Yes
C012	STAGE_III_MM	Discovery	Yes
C013	STAGE_III_MM	Discovery	Yes
C016	STAGE_III_MM	Discovery	Yes
C017	STAGE_III_MM	Discovery	Yes
C021	STAGE_III_MM	Discovery	Yes
C022	STAGE_III_MM	Discovery	Yes
C025	STAGE_III_MM	Discovery	Yes
C027	STAGE_III_MM	Discovery	Yes
C028	STAGE_III_MM	Discovery	Yes
C037	STAGE_III_MM	Discovery	Yes
C038	STAGE_III_MM	Discovery	Yes
C042	STAGE_III_MM	Discovery	Yes
C043	STAGE_III_MM	Discovery	Yes
C044	STAGE_III_MM	Discovery	Yes
C045	STAGE_III_MM	Discovery	Yes
C050	STAGE_III_MM	Discovery	Yes
C052	STAGE_III_MM	Discovery	Yes
C054	STAGE_III_MM	Discovery	Yes
C057	STAGE_III_MM	Discovery	Yes
C058	STAGE_III_MM	Discovery	Yes
C060	STAGE_III_MM	Discovery	Yes
C062	STAGE_III_MM	Discovery	Yes
C065	STAGE_III_MM	Discovery	Yes
C067	STAGE_III_MM	Discovery	Yes
C071	STAGE_III_MM	Discovery	Yes
C074	STAGE_III_MM	Discovery	Yes
C076	STAGE_III_MM	Discovery	Yes
C077	STAGE_III_MM	Discovery	Yes
C078	STAGE_III_MM	Discovery	Yes
C079	STAGE_III_MM	Discovery	Yes
C080	STAGE_III_MM	Discovery	Yes
C081	STAGE_III_MM	Discovery	Yes
C083	STAGE_III_MM	Discovery	Yes
C084	STAGE_III_MM	Discovery	Yes
C086	STAGE_III_MM	Discovery	Yes
C088	STAGE_III_MM	Discovery	Yes
C089	STAGE_III_MM	Discovery	Yes
C091	STAGE_III_MM	Discovery	Yes
C092	STAGE_III_MM	Discovery	Yes
C094	STAGE_III_MM	Discovery	Yes
C096	STAGE_III_MM	Discovery	Yes
C097	STAGE_III_MM	Discovery	Yes
C100	STAGE_III_MM	Discovery	Yes
C106	STAGE_III_MM	Discovery	Yes
C108	STAGE_III_MM	Discovery	Yes
A02	STAGE_IV_MM	Discovery	Yes
A04	STAGE_IV_MM	Discovery	Yes
A06	STAGE_IV_MM	Validation	Yes
A13	STAGE_IV_MM	Validation	Yes
A15	STAGE_IV_MM	Validation	Yes
AF6	STAGE_IV_MM	Validation	Yes
C-32	STAGE_IV_MM	Validation	Yes
CJM	STAGE_IV_MM	Validation	Yes
D01	STAGE_IV_MM	Validation	Yes
D04	STAGE_IV_MM	Validation	Yes
D05	STAGE_IV_MM	Validation	Yes
D05	STAGE_IV_MM	Validation	Yes
D10	STAGE_IV_MM	Validation	Yes
D11	STAGE_IV_MM	Validation	Yes
D14	STAGE_IV_MM	Validation	Yes
D17	STAGE_IV_MM	Validation	Yes
D20	STAGE_IV_MM	Validation	Yes
D22	STAGE_IV_MM	Validation	Yes
D24	STAGE_IV_MM	Validation	Yes
D25	STAGE_IV_MM	Validation	Yes
D28	STAGE_IV_MM	Validation	Yes
D29	STAGE_IV_MM	Validation	Yes
D32	STAGE_IV_MM	Validation	Yes
D35	STAGE_IV_MM	Validation	Yes
D36	STAGE_IV_MM	Validation	Yes
D40	STAGE_IV_MM	Validation	Yes
D41	STAGE_IV_MM	Validation	Yes
D59	STAGE_IV_MM	Discovery	Yes
HT144	STAGE_IV_MM	Validation	Yes
MM127	STAGE_IV_MM	Validation	Yes
MM253	STAGE_IV_MM	Validation	Yes
MM370	STAGE_IV_MM	Validation	Yes
MM386	STAGE_IV_MM	Validation	Yes
MM415	STAGE_IV_MM	Validation	Yes
MM426	STAGE_IV_MM	Validation	Yes
MM466	STAGE_IV_MM	Validation	Yes
MM473	STAGE_IV_MM	Validation	Yes
MM485	STAGE_IV_MM	Validation	Yes
MM537	STAGE_IV_MM	Validation	Yes
MM548	STAGE_IV_MM	Validation	Yes
MM649	STAGE_IV_MM	Validation	Yes
MM96L	STAGE_IV_MM	Validation	Yes
SKMEL28	STAGE_IV_MM	Validation	Yes
SKMEL5	STAGE_IV_MM	Validation	Yes
92_1	Uveal MM	Discovery	Yes
MEL202	Uveal MM	Discovery	Yes
MEL270	Uveal MM	Discovery	Yes
MEL285	Uveal MM	Discovery	Yes
MEL290	Uveal MM	Discovery	Yes
OCM8	Uveal MM	Discovery	Yes
OMM1	Uveal MM	Discovery	Yes

Throughout the specification the aim has been to describe the preferred embodiments of the invention without limiting the invention to any one embodiment or specific collection of features. It will therefore be appreciated by those of skill in the art that, in light of the instant disclosure, various modifications and changes can be made in the particular embodiments exemplified without departing from the scope of the present invention.
All computer programs, algorithms, patent and scientific literature referred to herein is incorporated herein by reference.

REFERENCES

1. Balch C M, Gershenwald J E, Soong S J, et al. Final version of 2009 AJCC melanoma staging and classification. J Clin Oncol 2009; 27(36): 6199-206.
2. AIHW. Australian Cancer Incidence and Mortality (ACIM) books: Melanoma of the skin: Canberra: AIHW; 2014.
3. SEER. SEER Cancer Statistics Review, 1975-2011, National Cancer Institute. Bethesda, Md. National Cancer Institute., 2014.
4. Brochez L, Naeyaert J M. Serological markers for melanoma. Br J Dermatol 2000; 143(2): 256-68.
5. Finck S J, Giuliano A E, Morton D L. LDH and melanoma. Cancer 1983; 51(5): 840-3.
6. Karakousis C P, Balch C M, Urist M M, Ross M M, Smith T J, Bartolucci A A. Local recurrence in malignant melanoma: long-term results of the multiinstitutional randomized surgical trial. Ann Surg Oncol 1996; 3(5): 446-52.
7. Sirott M N, Bajorin D F, Wong G Y, et al. Prognostic factors in patients with metastatic malignant melanoma. A multivariate analysis. Cancer 1993; 72(10): 3091-8.
8. Weide B, Elsasser M, Buttner P, et al. Serum markers lactate dehydrogenase and S100B predict independently disease outcome in melanoma patients with distant metastasis. Br J Cancer 2012; 107(3): 422-8.
9. Deichmann M, Benner A, Bock M, et al. S100-Beta, melanoma-inhibiting activity, and lactate dehydrogenase discriminate progressive from nonprogressive American Joint Committee on Cancer stage IV melanoma. J Clin Oncol 1999; 17(6): 1891-6.
10. Guo H B, Stoffel-Wagner B, Bierwirth T, Mezger J, Klingmuller D. Clinical significance of serum S100 in metastatic malignant melanoma. Eur J Cancer 1995; 31A(6): 924-8.
11. Smit L H, Korse C M, Hart A A, et al. Normal values of serum S-100B predict prolonged survival for stage IV melanoma patients. Eur J Cancer 2005; 41(3): 386-92.
12. Kruijff S, Bastiaannet E, Kobold A C, van Ginkel R J, Suurmeijer A J, Hoekstra H J. S-100B concentrations predict disease-free survival in stage III melanoma patients. Ann Surg Oncol 2009; 16(12): 3455-62.
13. Allegra A, Alonci A, Campo S, et al. Circulating microRNAs: new biomarkers in diagnosis, prognosis and treatment of cancer (review). Int J Oncol 2012; 41(6): 1897-912.
14. De Guire V, Robitaille R, Tetreault N, et al. Circulating miRNAs as sensitive and specific biomarkers for the diagnosis and monitoring of human diseases: promises and challenges. Clinical biochemistry 2013; 46(10-11): 846-60.
15. Grasedieck S, Sorrentino A, Langer C, et al. Circulating microRNAs in hematological diseases: principles, challenges, and perspectives. Blood 2013; 121(25): 4977-84.
16. Mitchell P S, Parkin R K, Kroh E M, et al. Circulating microRNAs as stable blood-based markers for cancer detection. Proc Nati Acad Sci USA 2008; 105(30): 10513-8.
17. Kanemaru H, Fukushima S, Yamashita J, et al. The circulating microRNA-221 level in patients with malignant melanoma as a new tumor marker. J Dermatol Sci 2011; 61(3): 187-93.
18. Friedman E B, Shang S, de Miera E V, et al. Serum microRNAs as biomarkers for recurrence in melanoma. J Transl Med 2012; 10: 155.
19. Griffiths-Jones S. The microRNA Registry. Nucleic Acids Res 2004; 32(Database issue): D109-11.
20. Stark M S, Boyle G M, Bonazzi V F, et al. miR-514a regulation of the tumour-suppressor gene NF1 confers resistance to PLX4032 targeted therapy of melanoma. In preparation 2014.
21. Sturm R A, Fox C, McClenahan P, et al. Phenotypic characterization of nevus and tumor patterns in MITF E318K mutation carrier melanoma patients. J Invest Dermatol 2014; 134(1): 141-9.
22. Kroh E M, Parkin R K. Mitchell P S, Tewari M. Analysis of circulating microRNA biomarkers in plasma and serum using quantitative reverse transcription-PCR (qRT-PCR). Methods 2010; 50(4): 298-301.
23. Wians F H J. Clinical Laboratory Tests: Which, Why, and What Do The Results Mean? Lab Medicine 2009; 40(2): 105-13.
24. Weide B, Richter S, Buttner P, et al. Serum S100B, lactate dehydrogenase and brain metastasis are prognostic factors in patients with distant melanoma metastasis and systemic therapy. PLoS One 2013; 8(11): e81624.
25. Morton D L, Thompson J F, Cochran A J, et al. Final trial report of sentinel-node biopsy versus nodal observation in melanoma. N Engl J Med 2014; 370(7): 599-609.
26. Hanahan D, Weinberg R A. Hallmarks of cancer: the next generation. Cell. 2011; 144: 646-74.
27. Alexandrov L B. Nik-Zainal S, Wedge D C, Aparicio S A, Behjati S, Biankin A V, et al. Signatures of mutational processes in human cancer. Nature. 2013;500:415-21.
28. Bonazzi V F, Stark M S, Hayward N K. MicroRNA regulation of melanoma progression. Melanoma Res. 2012;22:101-13.
29. Gaur A, Jewell D A, Liang Y, Ridzon D, Moore J H, Chen C, et al. Characterization of microRNA expression levels and their biological correlates in human cancer cell lines. Cancer Res. 2007;67:2456-68.
30. Ferracin M, Pedriali M, Veronese A, Zagatti B, Gafa R, Magri E, et al. MicroRNA profiling for the identification of cancers with unknown primary tissue-of-origin. J Pathol. 2011.
31. Poliseno L, Haimovic A, Segura M F, Hanniford D, Christos P J, Darvishian F, et al. Histology-specific microRNA alterations in melanoma. J Invest Dermatol. 2012;132:1860-8.
32. Stark M S, Tyagi S, Nancarrow D J, Boyle G M, Cook A L, Whiteman D C, et al. Characterization of the Melanoma miRNAome by Deep Sequencing. PLoS One. 2010;5:e9685.
33. Kozubek J, Ma Z, Fleming E, Duggan T, Wu R, Shin D G, et al. In-depth characterization of microRNA transcriptorne in melanoma. PLoS One. 2013;8:e72699.
34. Streicher K L, Zhu W, Lehmann K P, Georgantas R W, Morehouse C A, Brohawn P, et al. A novel oncogenic role for the miRNA-506-514 cluster in initiating melanocyte transformation and promoting melanoma growth. Oncogene. 2012;31:1558-70.
35. Wani S, Cloonan N. Profiling direct mRNA-microRNA interactions using synthetic biotinylated microRNA-duplexes. bioRxiv. 2014.
36. Martin H C, Wani S, Steptoe A L, Krishnan K, Nones K, Nourbakhsh E, et al. Imperfect centered miRNA binding sites are common and can mediate repression of target mRNAs. Genome Biol. 2014;15:R51.
37. Maertens O, Johnson B, Hollstein P, Frederick D T, Cooper Z A, Messiaen L, et al. Elucidating distinct roles for NF1 in melanomagenesis. Cancer Discov. 2013;3:338-49.
38. Whittaker S R, Theurillat J P, Van Allen E, Wagle N, Hsiao J, Cowley G S, et al. A genome-scale RNA interference screen implicates NF1 loss in resistance to RAF inhibition. Cancer Discov. 2013;3:350-62.
39. Shalem O, Sanjana N E, Hartenian E, Shi X, Scott D A, Mikkelsen T S, et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science. 2014;343:84-7.
40. Nissan M H, Pratilas C A, Jones A M, Ramirez R, Won H, Liu C, et al. Loss of NF1 in cutaneous melanoma is associated with RAS activation and MEK dependence. Cancer Res. 2014.
41. Levy C, Khaled M, Iliopoulos D, Janas M M, Schubert S, Pinner S, et al. Intronic miR-211 assumes the tumor suppressive function of its host gene in melanoma. Mol Cell. 2010;40:841-9.
42. Yokoyama S, Woods S L, Boyle G M, Aoude L G, MacGregor S, Zismann V, et al. A novel recurrent mutation in MITF predisposes to familial and sporadic melanoma. Nature. 2011;480:99-103.
43. Lito P, Rosen N, Solit D B. Tumor adaptation and resistance to RAF inhibitors. Nat Med. 2013;19:1401-9.
44. Khan A A, Betel D, Miller M L, Sander C, Leslie C S, Marks D S. Transfection of small RNAs globally perturbs gene regulation by endogenous microRNAs. Nat Biotechnol. 2009;27:549-55.
45. Hodis E, Watson I R, Kryukov G V, Arold S T, Imielinski M, Theurillat J P, et al. A landscape of driver mutations in melanoma. Cell. 2012;150:251-63.
46. Krauthammer M, Kong Y, Ha B H, Evans P, Bacchiocchi A, McCusker J P, et al. Exome sequencing identifies recurrent somatic RAC1 mutations in melanoma. Nat Genet. 2012;44:1006.-14.
47. van Kempen L C, van den Hurk K, Lazar V, Michiels S, Winnepenninckx V, Stas M, et al. Loss of microRNA-200a and c, and microRNA-203 expression at the invasive front of primary cutaneous melanoma is associated with increased thickness and disease progression. Virchows Arch. 2012;461:441-8.
48. Poell J B, van Haastert R J, de Gunst T, Schultz U. Gommans W M, Verheul M, et al. A functional screen identifies specific microRNAs capable of inhibiting human melanoma cell viability. PLoS One. 2012;7:e43569.
49. Dar A A, Majid S, de Semir D, Nosrati M, Bezrookove V, Kashani-Sabet M. miRNA-205 suppresses melanoma cell proliferation and induces senescence via regulation of E2F1 protein. J Biol Chem. 2011;286:16606-14.
50. Xu Y, Brenn T, Brown E R, Doherty V, Melton D W. Differential expression of microRNAs during melanoma progression: miR-200c, miR-205 and miR-211 are downregulated in melanoma and act as tumour suppressors. Br J Cancer. 2012;106:553-61.
51. Liu S, Tetzlaff M T, Liu A, Liegl-Atzwanger B, Guo J, Xu X. Loss of microRNA-205 expression is associated with melanoma progression. Lab Invest. 2012;92:1084-96.
52. Hanna J A, Hahn L, Agarwal S, Rimm D L. In situ measurement of miR-205 in malignant melanoma tissue supports its role as a tumor suppressor microRNA. Lab Invest. 2012;92:1390-7.
53. Noguchi S, Iwasaki J, Kumazaki M, Mori T, Maruo K, Sakai H, et al. Chemically modified synthetic microRNA-205 inhibits the growth of melanoma cells in vitro and in vivo. Molecular therapy : the journal of the American Society of Gene Therapy. 2013;21:1204-11.

Claims

1. A method of determining whether or not a subject has melanoma, including:

determining an expression level of one or more miRNA biomarkers in a biological sample from a subject, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, or a fragment or variant thereof, and wherein melanoma is detected if said expression level of one or more miRNA biomarkers is altered or modulated in the biological sample.

2. A method of determining the prognosis of a subject with melanoma, including:

determining an expression level of one or more miRNA biomarkers in a biological sample obtained from the subject, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, or a fragment or variant thereof, to thereby evaluate the prognosis of melanoma in the subject.

3. The method of claim 2, wherein if the expression level of said one or more miRNA biomarkers is altered or modulated in the biological sample, the prognosis may be negative or positive.

4. The method of claim 3, wherein the prognosis is used, at least in part, to determine whether the subject would benefit from treatment of melanoma.

5. The method of claim 3, wherein the prognosis is used, at least in part, to develop a treatment strategy for the subject.

6. The method of claim 2, wherein the prognosis is used, at least in part, to determine disease progression in the subject.

7. The method of claim 2, wherein prognosis is defined as an estimated time of survival.

8. The method of claim 2, further including determining suitability of the subject for treatment based, at least in part, on the prognosis.

9. The method of claim 1, further including determining a disease stage and/or grade for melanoma based on, at least in part, the expression level of the one or more miRNA biomarkers.

10. The method of claim 1, wherein the expression level of the one or more miRNA biomarkers is determined before, during and/or after treatment.

11. A method of treating melanoma in a subject including;

determining an expression level of one or more miRNA biomarkers in a biological sample from the subject, before, during and/or after treatment of melanoma, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, or a fragment or variant thereof, and based on the determination made, initiating, continuing, modifying or discontinuing a treatment of melanoma.

12. A method of evaluating treatment efficacy of melanoma in a subject including:

determining an expression level of one or more miRNA biomarkers in a biological sample from the subject before, during and/or after treatment, wherein the one or more miRNA biomarkers are selected from the group consisting of miRNA-4487, miRNA-4706 and miRNA-4731, or a fragment or variant thereof; and

determining whether or not the treatment is efficacious according to whether said expression level of one or more miRNA biomarkers is altered or modulated in the subject's biological sample.

13. The method of claim 11, further including selecting a treatment for melanoma based on the expression level of the one or more miRNA biomarkers.

14. The method of claim 11, further including measuring an expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p, or a fragment or variant thereof.

15-20. (canceled)

21. The method of claim 1, further including measuring an expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p, or a fragment or variant thereof.

22. The method of claim 2, further including determining a disease stage and/or grade for melanoma based on, at least in part, the expression level of the one or more miRNA biomarkers.

23. The method of claim 2, wherein the expression level of the one or more miRNA biomarkers is determined before, during and/or after treatment.

24. The method of claim 2, further including measuring an expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p, or a fragment or variant thereof.

25. The method of claim 12, further including selecting a treatment for melanoma based on the expression level of the one or more miRNA biomarkers.

26. The method of claim 12, further including measuring an expression level of one or more additional miRNA biomarkers selected from the group consisting of miRNA-16, miRNA-211, miRNA-509-3p and miRNA-509-5p, or a fragment or variant thereof.