CROSS REFERENCE TO RELATED APPLICATIONS

This application claims priority to U.S. Provisional Patent Application No. 61/835,743 filed June 17, 2013, incorporated herein by reference.

FIELD OF THE INVENTION

The invention relates to a micro-RNA signature for sarcoma.

BACKGROUND Sarcomas are cancers of mesenchymal origin which represent 2% of human malignancies ¹ . One of the most common STS subtype is the undifferentiated pleomorphic sarcoma (UPS) which is amongst the most aggressive STS with a high propensity for metastasis; associated with a dismal 5-year overall survival of 30-50% ^{2" 4} . The prognostic determinants in STS are grade, tumor size and surgical margin ⁵ , which are however not useful in determining who may benefit from chemotherapy ^6,7 . Thus, there is a need to develop novel biomarkers, which will provide both insights into the complex biology of UPS, and facilitate individualization of cancer therapy.

MicroRNAs (miRNA) are small non-coding RNA molecules of ~22-nucleotides that form one of the largest class of gene regulators by targeting up to 60% of the mRNAs to translational repression or degradation. There have been three studies describing miRNA expression patterns for a diverse group of sarcomas. These studies, which ranged from 27 to 270 samples of 5-22 different STS histologies, demonstrated that miRNA expression profiling can differentiate some STS histologies ^8"10 . Hisaoka ef al. further functionally characterized miRNAs specific to synovial sarcomas, demonstrating that modulating miR-let-7e and miR-99b levels affected downstream gene targets and suppressed cell proliferation in vitro, suggesting a potential role for these miRNAs in STS ⁹ . To date however, there have not been reports of miRNA profiling of STS in relation to clinical outcome. SUMMARY OF THE INVENTION

In an aspect, there is provided a method of prognosing or classifying a subject with undifferentiated pleomorphic sarcoma (UPS) comprising: (a) determining the expression of at least one biomarker in a test sample from the subject selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; and (b) comparing expression of the at least one biomarker in the test sample with expression of the at least one biomarker in a control sample; wherein a difference or similarity in the expression of the at least one biomarker between the control and the test sample is used to prognose or classify the subject with UPS into a low risk group or a high risk group of developing metastasis.

In an aspect, there is provided a method of selecting a therapy for a subject with UPS, comprising the steps: (a) classifying the subject with UPS into a high risk group or a low risk group according to the method described herein; and (b) selecting a more aggressive therapy, preferably adjuvant chemotherapy or radiation therapy, for the high risk group or a less aggressive therapy, preferably no adjuvant chemotherapy or no radiation therapy, for the low risk group.

In an aspect, there is provided a method of selecting a therapy for a subject with UPS, comprising the steps: (a) determining the expression of at least one biomarker in a test sample from the subject selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; (b) comparing expression of the at least one biomarker in the test sample with expression of the at least one biomarker in a control sample; (c) classifying the subject in a high risk group or a low risk group, wherein a difference or a similarity in the expression of the at least one biomarker between the control sample and the test sample is used to classify the subject into a high risk group or a low risk group; (d) selecting a more aggressive therapy, preferably adjuvant chemotherapy or radiation therapy, for the high risk group or a less aggressive therapy, preferably no adjuvant chemotherapy or no radiation therapy, for the low risk group.

In an aspect, there is provided a composition comprising a plurality of isolated nucleic acid sequences, wherein each isolated nucleic acid sequence hybridizes to: (a) Mir- 221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; and/or (b)a nucleic acid complementary to a), wherein the composition is used to measure the level of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132 expression. In an aspect, there is provided an array comprising, for each of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132, one or more polynucleotide probes complementary and hybridizable thereto.

In an aspect, there is provided a computer program product for use in conjunction with a computer having a processor and a memory connected to the processor, the computer program product comprising a computer readable storage medium having a computer mechanism encoded thereon, wherein the computer program mechanism may be loaded into the memory of the computer and cause the computer to carry out the method described herein. In an aspect, there is provided a computer implemented product for predicting a prognosis or classifying a subject with UPS comprising: (a) a means for receiving values corresponding to a subject expression profile in a subject sample; and (b) a database comprising a reference expression profile associated with a prognosis, wherein the subject biomarker expression profile and the biomarker reference profile each have at least one value representing the expression level of at least one biomarker selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; wherein the computer implemented product selects the biomarker reference expression profile most similar to the subject biomarker expression profile, to thereby predict a prognosis or classify the subject. In an aspect, there is provided a computer implemented product for determining therapy for a subject with UPS comprising: (a) a means for receiving values corresponding to a subject expression profile in a subject sample; and (b) a database comprising a reference expression profile associated with a prognosis, wherein the subject biomarker expression profile and the biomarker reference profile each have at least one value representing the expression level of at least one biomarker selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; wherein the computer implemented product selects the biomarker reference expression profile most similar to the subject biomarker expression profile, to thereby predict the therapy.

In an aspect, there is provided a computer readable medium having stored thereon a data structure for storing the computer implemented product described herein. In an embodiment, the data structure is capable of configuring a computer to respond to queries based on records belonging to the data structure, each of the records comprising: (a) a value that identifies a biomarker reference expression profile of at least one of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; (b) a value that identifies the probability of a prognosis associated with the biomarker reference expression profile.

In an aspect, there is provided a computer system comprising (a) a database including records comprising a biomarker reference expression profile of at least one of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132 associated with a prognosis or therapy; (b) a user interface capable of receiving a selection of expression levels of the at least one biomarker for use in comparing to the biomarker reference expression profile in the database; (c) an output that displays a prediction of prognosis or therapy according to the biomarker reference expression profile most similar to the expression levels of the at least one biomarker.

In an aspect, there is provided a kit comprising reagents for detecting the expression of any or all of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132 in a sample.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 shows application of the 6-miRNA prognostic signature on a) the "Validation Set" and analyzed for Distant Metastasis Free Survival (DMFS) based on their risk group "High" and "Low" and b) metastasis samples. Figure 2 shows association of a) individual miRNAs within the 6-miR signature with distant metastasis free survival (DMFS) and disease free survival (DFS) of the combined datasets (Training + Validation) on univariate analysis (log-rank) and multivariate analysis (Cox PH regression) and b) imiRNA-138 expression with DFS in the UPS samples. Values presented as p-value and hazard ratios (HR). Figure 3 shows summary of assays from different dataset and findings related to the 6- miR prognostic signature and RhoA gene expression.

Figure 4 shows overall survival of patients from the "Training Sef(Black) and "Validation Sef'(Red) Figure 5 shows unsupervised hierarchical clustering using the Ward method of miRNA expression from 4 Normal tissues (Adipose, Carotid, Vein and Smooth Muscle), 4 primary UPS cell lines (STS48, STS93, STS109 and STS117) and 42 UPS samples from the "Training Set" (metastatic in red, non-metastatic in white). Figure 6 shows that for migration and invasion assays, 1.5 * 10 ⁵ cells were seeded inside the insert with medium containing 1% serum. High serum (20%) medium was then added to the bottom chamber of 24-well plates to serve as a chemo-attractant. Invasion index is calculated as (% Invasion of Test Cell)/(% Invasion of Control Cell) and depicted in a) following transfection with 10nM of pre-miR-138 or 50nM of Locked Nucleic Acid (LNA) of miR-138 and in b) for the morphological changes of the cells on the migration and invasion chambers.

Figure 7 shows clonogenic assay of STS117 cells following transfection with 10nM or 50nM of Locked Nucleic Acid (Control-Scrambled and mir-138).

Figure 8 shows selection of genes affected by the transfection of STS 1 17 cells by LNA-antimiR-138, LNA-antimiR-224 and pre-miR-375 while excluding genes modulated by LNA-antimiR-130a. Global mRNA profiling of STS117 cells 24-hours post-transfection were done using the Affymetrix Human Genome U133 plus 2.0 array that was processed with Affymetrix's WT Express protocol and 100 ng of starting material. The arrays were hybridized for 17 hrs at 45oC and washed and stained in fluidic station P450. Images were acquired with GeneChip scanner 3000 and preliminary analysis was carried out with Affymetrix gene expression console.

Figure 9 shows the quantification of a) RhoA mRNA expression in primary samples from non-metastatic (n=14) and metastatic patients (n=14), and metastatic samples (n=10). Significant differences in RhoA mRNA expression were observed between each group (Student t-test p __D.006). Data plotted as delta Ct normalized to the average values from the No-DM group (Lower number = lower expression), b) Protein level and activity downstream of miR-138 and RhoA.

Figure 10 shows miR-138-Rho-ROCK pathway schema (a), (b) proposed changes secondary to increased expression of miR-138 in combination with reduced RhoA and (c) in the potential convergence of targets from the other miRNA within the prognostic signature. Figure 11 shows distant metastasis free survival of 1056 breast cancer patients from 7 datasets dichotomized by the median expression of RhoA. Median follow-up was 238 months.

DESCRIPTION

In the following description, numerous specific details are set forth to provide a thorough understanding of the invention. However, it is understood that the invention may be practiced without these specific details.

A common and aggressive subtype of soft-tissue sarcoma, undifferentiated pleomorphic sarcoma (UPS) was examined to develop and validate a prognostic signature for distant metastasis-free survival (DMFS) composed of micro-RNAs (miRNA).

Following central pathology review, 110 fresh frozen UPS samples annotated prospectively with clinical data were split into independent training and validation cohorts. Following global miRNA profiling of the training set, multivariate regression model was fit on the miRNA expression values to yield a 6-miRNA signature model associated with DMFS. The expression of the 6 miRNAs were then measured in the validation set and metastatic samples to test the signature using Kaplan Meier and multivariate analysis adjusted for: patient age, gender, tumor grade, size, depth and radiotherapy use. Following cellular miRNA level modulation, in-vitro assays were used to derive biological understanding of miRNAs in promoting metastasis. Public breast cancer datasets (n=1912) were analyzed in silico to determine the prognostic value of the signature and RhoA in breast cancers.

Using the 6-miRNA training signature, patients from the validation set were successfully categorized into "High" and "Low" risk groups for DMFS (HR:2.2; p=0.05) and classified all metastatic samples as "High" risk. The signature is capable of predicting patient DMFS (HR:3.5; p=0.0001) after adjusting for other prognostic markers. In-vitro experiments suggest the involvement of RhoA/C-ROCK1/2-LIMK1/2 as downstream signature miRNA targets. The prognostic ability of the 6-miRNA signature (HR:1.8; p=0.01) and RhoA (HR:0.6; p=0.013) were demonstrated in the breast datasets. A prognostic 6-miRNA signature has been successfully developed and validated for UPS. This signature could help identify patients at "High" risk for distant metastasis, who might benefit from more aggressive systemic therapies. Common pathways that promote the development of metastasis in sarcomas and breast cancers may be present and are potential targets for future therapeutic investigations.

In an aspect, there is provided a method of prognosing or classifying a subject with undifferentiated pleomorphic sarcoma (UPS) comprising: (a) determining the expression of at least one biomarker in a test sample from the subject selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; and (b) comparing expression of the at least one biomarker in the test sample with expression of the at least one biomarker in a control sample; wherein a difference or similarity in the expression of the at least one biomarker between the control and the test sample is used to prognose or classify the subject with UPS into a low risk group or a high risk group of developing metastasis. The term "level of expression" or "expression level" as used herein refers to a measurable level of expression of the products of biomarkers, such as, without limitation, the level of micro-RNA, messenger RNA transcript expressed or of a specific exon or other portion of a transcript, the level of proteins or portions thereof expressed of the biomarkers, the number or presence of DNA polymorphisms of the biomarkers, the enzymatic or other activities of the biomarkers, and the level of specific metabolites.

As used herein, the term "control" refers to a specific value or dataset that can be used to prognose or classify the value e.g. expression level or reference expression profile obtained from the test sample associated with an outcome class. A person skilled in the art will appreciate that the comparison between the expression of the biomarkers in the test sample and the expression of the biomarkers in the control will depend on the control used.

The term "differentially expressed" or "differential expression" as used herein refers to a difference in the level of expression of the biomarkers that can be assayed by measuring the level of expression of the products of the biomarkers, such as the difference in level of micro-RNA or a portion thereof expressed . In a preferred embodiment, the difference is statistically significant. The term "difference in the level of expression" refers to an increase or decrease in the measurable expression level of a given biomarker, for example as measured by the amount of micro-RNA as compared with the measurable expression level of a given biomarker in a control.

The term "low risk" as used herein in respect of UPS refers to a lower risk of UPS r as compared to a general or control population. The term "sample" as used herein refers to any fluid, cell or tissue sample from a subject that can be assayed for biomarker expression products and/or a reference expression profile, e.g. genes differentially expressed in subjects.

In some embodiments, the level of gene expression is determined and compared.

A person skilled in the art will appreciate that a number of methods can be used to detect or quantify the level of micro-RNA within a sample, including arrays, such as microarrays, RT-PCR (including quantitative RT-PCR), nanostring and various sequencing technologies. By way of non-limiting example, the RNA was quantified as follows in the disclosed exampels: Training Set: Total RNA from all tumor samples were extracted using the RNeasy kits (Qiagen). Global profiling of miRNA expression on the "Training Set" was performed using the TaqMan® Human Micro-RNA Array A (Applied Biosystems, Inc. CA, USA). Total RNA (300ng) was first reverse-transcribed with the Multiplex RT pool set, then quantitated using an Applied Biosystems 7900 HT Real-Time PCR system as previously describedl . Data were normalized using endogenous controls (RNU6B, RNU44 and RNU48) that were simultaneously quantified. Validation Set: Single well quantification of miRNA expressions was assessed by initially reverse-transcribing 200ng of total RNA with multiscribe reverse transcriptase and miR-specific primers (50nM), followed by qRT-PCR analysis using TaqMan microRNA Assays (Applied Biosystems). [Hui AB, Shi W, Boutros PC, et al: Robust global micro-RNA profiling with formalin-fixed paraffin-embedded breast cancer tissues. Lab Invest 89:597-606, 2009].

In various embodiments, the at least one biomarkers is two biomarkers, three biomarkers, four biomarkers, five biomarkers, or six biomarkers.

In some embodiments, overexpression of Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132 and underexpression of Mir-221 is indicative of high risk. In some embodiments, classification of the subject into a low or high risk group is based on a score = -O.IS^miR-IS -O^g^'miRNA-ISS'-O^I ^miR-MS'+O^Z^miR- 221 ^, -0.334 ^* 'miR-224'-0.35 ^*, miR-491-5p'.

In some embodiments, determining the biomarker expression level comprises use of quantitative PCR or an array, preferably sequencing technologies or nanostring.

In some embodiments, the sample comprises a tissue sample.

In an aspect, there is provided a method of selecting a therapy for a subject with UPS, comprising the steps: (a) classifying the subject with UPS into a high risk group or a low risk group according to the method described herein; and (b) selecting a more aggressive therapy, preferably adjuvant chemotherapy or radiation therapy, for the high risk group or a less aggressive therapy, preferably no adjuvant chemotherapy or no radiation therapy, for the low risk group.

In an aspect, there is provided a method of selecting a therapy for a subject with UPS, comprising the steps: (a) determining the expression of at least one biomarker in a test sample from the subject selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; (b) comparing expression of the at least one biomarker in the test sample with expression of the at least one biomarker in a control sample; (c) classifying the subject in a high risk group or a low risk group, wherein a difference or a similarity in the expression of the at least one biomarker between the control sample and the test sample is used to classify the subject into a high risk group or a low risk group; (d) selecting a more aggressive therapy, preferably adjuvant chemotherapy or radiation therapy, for the high risk group or a less aggressive therapy, preferably no adjuvant chemotherapy or no radiation therapy, for the low risk group.

In an aspect, there is provided a composition comprising a plurality of isolated nucleic acid sequences, wherein each isolated nucleic acid sequence hybridizes to: (a) Mir- 221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; and/or (b)a nucleic acid complementary to a), wherein the composition is used to measure the level of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132 expression.

The term "nucleic acid" includes DNA and RNA and can be either double stranded or single stranded. The term "hybridize" or "hybridizable" refers to the sequence specific non-covalent binding interaction with a complementary nucleic acid. In a preferred embodiment, the hybridization is under high stringency conditions. Appropriate stringency conditions which promote hybridization are known to those skilled in the art, or can be found in Current Protocols in Molecular Biology, John Wiley & Sons, N.Y. (1989), 6.3.1 6.3.6. For example, 6.0 x sodium chloride/sodium citrate (SSC) at about 45°C, followed by a wash of 2.0 x SSC at 50°C may be employed.

In an aspect, there is provided an array comprising, for each of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132, one or more polynucleotide probes complementary and hybridizable thereto.

The term "probe" as used herein refers to a nucleic acid sequence that will hybridize to a nucleic acid target sequence. In one example, the probe hybridizes to the RNA biomarker or a nucleic acid sequence complementary thereof. The length of probe depends on the hybridization conditions and the sequences of the probe and nucleic acid target sequence. In one embodiment, the probe is at least 8, 10, 15, 20, 25, 50, 75, 100, 150, 200, 250, 400, 500 or more nucleotides in length.

The term "primer" as used herein refers to a nucleic acid sequence, whether occurring naturally as in a purified restriction digest or produced synthetically, which is capable of acting as a point of synthesis when placed under conditions in which synthesis of a primer extension product, which is complementary to a nucleic acid strand is induced (e.g. in the presence of nucleotides and an inducing agent such as DNA polymerase and at a suitable temperature and pH). The primer must be sufficiently long to prime the synthesis of the desired extension product in the presence of the inducing agent. The exact length of the primer will depend upon factors, including temperature, sequences of the primer and the methods used. A primer typically contains 15-25 or more nucleotides, although it can contain less or more. The factors involved in determining the appropriate length of primer are readily known to one of ordinary skill in the art.

In an aspect, there is provided a computer program product for use in conjunction with a computer having a processor and a memory connected to the processor, the computer program product comprising a computer readable storage medium having a computer mechanism encoded thereon, wherein the computer program mechanism may be loaded into the memory of the computer and cause the computer to carry out the method described herein.

In an aspect, there is provided a computer implemented product for predicting a prognosis or classifying a subject with UPS comprising: (a) a means for receiving values corresponding to a subject expression profile in a subject sample; and (b) a database comprising a reference expression profile associated with a prognosis, wherein the subject biomarker expression profile and the biomarker reference profile each have at least one value representing the expression level of at least one biomarker selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; wherein the computer implemented product selects the biomarker reference expression profile most similar to the subject biomarker expression profile, to thereby predict a prognosis or classify the subject. Preferably, computer implemented product is for use with the method described herein.

In an aspect, there is provided a computer implemented product for determining therapy for a subject with UPS comprising: (a) a means for receiving values corresponding to a subject expression profile in a subject sample; and (b) a database comprising a reference expression profile associated with a prognosis, wherein the subject biomarker expression profile and the biomarker reference profile each have at least one value representing the expression level of at least one biomarker selected from Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; wherein the computer implemented product selects the biomarker reference expression profile most similar to the subject biomarker expression profile, to thereby predict the therapy. Preferably, computer implemented product is for use with the method described herein. In an aspect, there is provided a computer readable medium having stored thereon a data structure for storing the computer implemented product described herein. In an embodiment, the data structure is capable of configuring a computer to respond to queries based on records belonging to the data structure, each of the records comprising: (a) a value that identifies a biomarker reference expression profile of at least one of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132; (b) a value that identifies the probability of a prognosis associated with the biomarker reference expression profile. In an aspect, there is provided a computer system comprising (a) a database including records comprising a biomarker reference expression profile of at least one of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132 associated with a prognosis or therapy; (b) a user interface capable of receiving a selection of expression levels of the at least one biomarker for use in comparing to the biomarker reference expression profile in the database; (c) an output that displays a prediction of prognosis or therapy according to the biomarker reference expression profile most similar to the expression levels of the at least one biomarker.

In an aspect, there is provided a kit comprising reagents for detecting the expression of any or all of Mir-221 , Mir-491-5p, Mir-224, Mir-138, Mir-143 and Mir-132 in a sample.

The advantages of the present invention are further illustrated by the following examples. The examples and their particular details set forth herein are presented for illustration only and should not be construed as a limitation on the claims of the present invention.

EXAMPLES

Methods and Materials

Patient information and tissues

Institutional Research Ethics Board approval was obtained from the University Health Network (UHN) and Mount-Sinai Hospital in Toronto for this study. Samples are collected from multiple Canadian institutions and stored as fresh frozen samples within the Clinical Core and Tumor Bank at the Mount-Sinai Hospital, which also prospectively annotate the samples with clinical data. The "Training Set" of UPS was comprised of 42 fresh frozen samples that had undergone central pathology review at the Mount-Sinai Hospital and were obtained from Stage l-lll patients diagnosed from 1988-1999. Following central pathology review of 100 other sarcoma samples from patients diagnosed from 2000-2010, 68 fresh frozen samples were confirmed to be UPS from Stage l-lll patients and formed the "Validation Set". Samples were collected from patients prior to any chemotherapy or radiotherapy. Corresponding fresh frozen metastatic samples (n=10) from lung metastatectomies of 6 patients from the "Validation Set" were obtained from the UHN biobank. Pathology review of the metastasis corresponded with the original diagnosis of the primary tumors.

Four RNA samples originating from non-cancerous human tissues of mesenchymal origin were purchased from Clontech Laboratories, Inc. (Smooth muscle), Applied Biosystem, Inc. (Adipose tissue) and Agilent Technologies, Inc. (Carotid Artery and Vein).

RNA purification from UPS samples

Total RNA from the tumors were extracted using the RNeasy kits (Qiagen). Samples were assayed randomly, with clinical outcome unknown, to avoid experimental bias. Cell lines and reagents

Three primary cell lines (STS48, STS93, STS117) from patients diagnosed with UPS were used for in-vitro experiments. The cells were incubated at 37°C under 5% C0 ₂ in DMEM:F12 1 :1 media with 10% bovine serum. Lipofectamine 2000 (Invitrogen) was used to transfect cells with pre-miRs (Invitrogen), Locked-Nucleic-Acid (LNA) antimiRs (Exiqon) and siRNAs (Qiagen). Total RNA was isolated using the Total RNA Purification Kit (Norgen, Inc.), according the manufacturer's protocol. Recovered RNA concentrations and quality were measured using the Nanodrop 1000A spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA). The cytotoxic effects of transfections were investigated in cells using clonogenic assays ¹¹ . Cell-cycle analysis of pre-miRNA-138 and LNA-antimiRNA-138 transfected cells were done using propidium iodide staining as previously described ¹¹ . Analyses were made using the BD FACScalibur using the FL-2 channel. The flow cytometry data were analyzed using FlowJo software (Tree Star, Inc.). Invasion and migration of cells were assayed using the BD BioCoat Matrigel Invasion Chambers and Control Inserts (BD Bioscience) as per the manufacturer's instructions.

Quantification of miRNA and mRNA MiRNA profiling

Global profiling of miRNA expression on the "Training Set" was performed using the TaqMan ^® Human Micro-RNA Array A (Applied Biosystems, Inc. CA, USA). Total RNA (300ng) was first reverse-transcribed with the Multiplex RT pool set, then quantitated using an Applied Biosystems 7900 HT Real-Time PCR system as previously described ¹² . Data was normalized using endogenous controls (RNU6B, RNU44 and RNU48) that were simultaneously quantified. The resulting ACt values were used for hierarchical clustering and signature derivation. Clustering was done using JMP 10 (SAS institute, Cary, NC, USA).

Real-time Quantification of RNAs

Single well quantification of miRNA expressions was assessed by initially reverse- transcribing 200ng of total RNA with multiscribe reverse transcriptase and miRNA- specific primers (50nM), followed by qRT-PCR analysis using TaqMan microRNA Assays (Applied Biosystems) ¹² . Quantitative RT-PCR was also utilized to analyze mRNA expression of: DICER, RHOA, RHOC, ROCK1 , ROCK2, LEPR, LIMK1 and GAPDH. Following reverse transcription of 200ng of total RNA using Superscript II I Reverse Transcriptase (Invitrogen), qRT-PCR was done using SYBR Green PCR Master Mix (Applied Biosystems). Primers for PCR amplifications were designed using Primer 3 Input (version 0.4.0) (Table 1 ). Relative mRNA levels were calculated using the 2 ^{~ ACt} method ¹³ .

Table 1 : Primer sequences used for quantitative RT-PCR:

Affymetrix Human Genome U133 plus 2.0

Global gene expression was profiled using the Affymetrix Human Genome U133 plus 2.0 array (Affymetrix, Inc). The arrays were run at Ontario Cancer Institute Genomics Centre using 100ng of total RNA. Data were pre-processed using the RMA method and 3X median for background removal.

Western blot analysis

Total protein extracts were harvested from cell lines 48 hours post-transfection and prepared for immunoblotting as previously described ¹⁴ . Membranes were probed with anti-RhoA, anti-RhoC, anti-ROCK1 , anti-ROCK2, anti-LIMK1 , anti-LIMK2, anti- phosphoLIMKI , anti-phosphoLIMK2 (Cell Signaling Technology, Inc.) and anti-GAPDH (glyceraldehyde-3-phosphate dehydrogenase) mAbs (1 :15,000 dilution; Abeam, Inc.), followed by secondary antibodies conjugated to horseradish peroxidase (1 :2,000 dilution; Abeam, Inc.) GAPDH protein levels were used as loading controls. Western blots were quantified with the Adobe Photoshop Pixel Quantification Plug-In (Richard Rosenman Advertising & Design.

Statistical analyses

Development and validation of miRNA signature:

Univariate analyses were conducted on the "Training Set" using Cox proportional hazard (PH) regression model. Potentially associated miRNAs (p-value <0.05) were applied into multivariate models while adjusting for multiple clinical factors. Cox PH regression models were built for time to event outcomes: Distant-metastasis-free- survival (DMFS), OS, disease-free-survival (DFS), and local-recurrence-free-survival. Stepwise-selection algorithm was implemented for model selection to select the 6 most-significantly associated miRNAs for the DFMS signature. Hazard ratios (HRs) and 95% confidence intervals (Cls) were estimated for significant predictors. Statistical significance level for multivariate analysis was <0.05. The signature score was based on the weighted combination of the miRNAs with the estimated regression coefficient of the Cox PH regression model as the weight ^15,16 . Statistics were performed using SAS version 9.1 (SAS institute, Cary, NC, USA) and the R package (http://CRAN.R- proiect.org, R Foundation, Vienna, Austria). All in-vitro experiments were conducted at least 3 independent times, with the data presented as the mean ± SEM. The statistical differences between treatment groups were determined using a Student f-test when comparing 2 treatment groups.

Breast cancer dataset - TCGA breast cancer (BRCA) dataset: for the 856 samples in which corresponding miRNA and mRNA were profiled, miRNA counts-per-million- reads and mRNA RPKM from the dataset were converted into z-scores. The sarcoma signature scoring was then applied onto the z-scores to dichotomize patients into risk groups (High and Low). MRNA expressions were dichotomized by the median Z-score for univariate and multivariate analysis. The Desmedt (n=198) ¹⁷ , Kao (n=327) ¹⁸ , Loi1 (n=87) ¹⁹ , Loi3 (n=77) ²⁰ , Minn2 (n=99) ²¹ , Schmidt (n=200) ²² , Symmansl (n=195) ²³ and Symmans2 (n=103) ²³ breast cancer datasets were used for assessment of mRNA (RhoA, LIMK1 and Dicer) expressions and outcomes. These datasets were selected for their use of Affymetrix mRNA profiling platforms with annotated clinical outcome. All mRNA expressions were converted into z-scores within each dataset, and then combined together for analysis.

RESULTS AND DISCUSSION

Sample and Patient Characteristics:

Descriptive statistics of the patients in the "Training Set" and "Validation Set" are provided in Table 2. As patients from the "Validation Set" were treated more recently than patients in the "Training Set", the "Validation Set" had shorter follow up (p<0.0001 ), larger tumors (p=0.03) and older patient age (p=0.0004). Despite the aforementioned differences, the 5-year OS probabilities of the 2 cohorts were similar (54% vs. 58%; Log-rank p=0.83) (Figure 4).

Table 2: Patient, disease and treatment characteristics of the UPS "Training Set" and "Validation Set"

Factors Training set Validation set P-value

N=42 N=68

Gender M:F 23:18 41:27 0.67

Stage l/ll 40% 38% 0.82

III 60% 62%

Grade 2 26% 21% 0.73

3 74% 79%

Adjuvant Chemo No 98% 98% 0.72

Yes 2% 2%

Adjuvant RT No 29% 32% 0.54

Yes 71% 68%

Median age (range) 64 (35-95) 68 (32-90) 0.03

Median size (range) 6.25 (1.3-28) 11.5(2.2-28) 0.0004

Median follow-up mos (range) 123 (18-225) 28.5(0-116) <0.0001

Development and validation of the miRNA signature:

Global miRNA profiling of the "Training Set" showed that 179 (47.3%) of the mean miRNA expressions were significantly (t-test p<0.00013) reduced in UPS in comparison to normal tissues. None of the miRNAs were significantly over-expressed in UPS in comparison to normal tissues. Based on the univariate and multivariate modeling of the miRNA expressions from the "Training Set" (Table 3) and adjusting for clinical factors, a prognostic signature score for DMFS consisting of 6 miRNAs was developed: Score = -0.15*'miR-132'-0.299 ^*, miRNA-138 ^, -0.217 ^*, miR-143'+0.427* ^, miR- 221 ^, -0.334 ^* 'miR-224'-0.35* ^, miR-491-5p'. The patients were stratified into "Low risk" and "High risk" categories according to their signature score value: low risk (score <median); high risk (risk score >median). The hazard ratio (HR) for DMFS was 16.0 in the training set (p<0.0001). Table 3: List of miRNAs associated with clinical outcomes of patients in the "Training Set". MiRNAs within the final signature prognostic of Distant Metastasis Free survival are in red. Distant Metastasis (DM), Local Recurrence (LR), Disease Free Survival (DFS) and Overall Survival (OS) DM LR DFS OS hsa-miR-10a hsa-let-7g hsa-miR-125b hsa-miR-224 hsa-miR-93 hsa-miR-139-5p hsa-miR-128 hsa-miR-491-5p hsa-miR-106b hsa-miR-211 hsa-miR-130a hsa-miR-128 hsa-miR-449a hsa-miR-138 hsa-miR-130a hsa-miR-512-3p hsa-miR-139-5p hsa-miR-132 hsa-miR-181a hsa-miR-143 hsa-miR-197 hsa-miR-221 hsa-miR-224 hsa-miR-224 hsa-miR-331-3p hsa-miR-34a hsa-miR-339-3p hsa-miR-374a hsa-miR-433 hsa-miR-455-3p hsa-miR-886-5p hsa-miR-491-5p hsa-miR-744

To verify the signature, the score for each sample from the "Validation Set" (n=68) was calculated using the miRNA expression measurements from single-well qRT-PCR and the above formula. In the validation set, "High-Risk" patients were 2.2 times more likely to develop metastasis (p = 0.05) (Figure 1a) than "Low Risk" patients on univariate analysis. After adjusting for patient age, gender, tumor grade, maximum size, depth, and use of adjuvant radiotherapy, the signature remained capable of discriminating patients at "High" and "Low" risk of developing DM (HR: 3.5; p=0.0001) in the combined groups of patients (Training + Validation set) (Table 4a). Only tumor depth persisted to be significantly associated with DMFS on multivariate analysis (HR: 0.3; p=0.04).

Table 4a: Multivariate analysis of the 6-miRNA signature score for its ability to predict distant metastasis free survival in the combined UPS cohort of "Training Set" and "Validation Set".

Application of the signature scoring on the metastatic samples classified all metastatic samples as "High" risk, with 9 of the 10 metastasis having higher risk scores than their corresponding primaries (Figure 1 b).

MiRNA-138 promotes invasion in sarcoma cells

Three primary cell lines (STS48, STS93, and STS117) derived from patients diagnosed with UPS were used to study the biologic functions of miRNAs in UPS. Global miRNA profiling of the "Training Set", 4 primary UPS cell lines (STS48, STS93, STS109 and STS117) and 4 RNA samples from normal mesenchymal tissues (Adipose, Carotid, Vein and Smooth Muscle) demonstrated that the expression of miRNAs differed between normal and sarcoma samples (Figure 5). To select candidate miRNAs for more in depth studies, screening migration/invasion and clonogenic assays were performed following modulation of cellular miRNA levels (Table 5). These experiments suggested that increased expression of miRNA-138 promoted cell invasion without affecting survival ( Figures 6a and 7) or cell cycle (data not shown). Of the 6 miRNAs within the signature, elevated expression of miRNA-138 was best associated with shorter DMFS (log-rank p=0.60) and DFS (p=0.017) on Kaplan-Meier univariate analysis and Cox PH multivariate analysis (DMFS: HR=2.3; p=0.008) (Figure 2a and b) of the combined UPS datasets. Table 5: List of miRNAs selected for functional screening using data from the global profiling of the "Training Set" and 4 normal mesenchymal tissue RNAs. STS117 cells were transfected with 50nM of Locked Nucleic Acid antimir-128, 130a, 138, 139-5p and 224, and pre-miR-375 to evaluate the effect of miRNA modulation on the migration, invasion and clonogenic survival of the cells.

To determine potential pathways related to increased metastatic potentials in UPS, global mRNA expression analysis was performed on STS117 cells, which spontaneously develop lung metastasis in NOD-SCID mice xenografts, following their transfections with LNA - antimiR-130a, antimiRNA-138, antimiR-224 and pre-miR-375. As transfection with LNA-miR-130a was cytotoxic and had no effect on cellular invasion, while LNA-antimiRNA-138, 224 and pre-miR-375 reduced the migration and invasion of cells, we identified common genes affected by the transfection of LNA- antimiRNA-138, 224 and pre-miR-375, but not by LNA-antimiRNA-130a (Figure 8). In addition to validated targets of miRNA-138 such RhoC and ROCK2 ²⁴ ' ²⁵ , these genes are potentially important to the observed increase in invasiveness of UPS cells secondary miRNA-138 over-expression.

In addition, global mRNA profiling of the "Training Set" (unpublished data from JW and IA) identified the under-expression of RhoA to be associated with increased odds of developing metastasis. To validate this finding, RhoA mRNA expression was measured in 28 samples from the "Validation Set" (14 patients who developed metastasis and 14 patients without metastasis) and the 10 metastatic samples. The association between metastasis and reduced expression of RhoA mRNA was validated in both the "Validation Set" and the metastatic samples (p ^).006) (Figure 9a).

Pathway analysis (DAVID, g-profiler) of validated miRNA-138 target genes and the above identified genes suggested that the cofilin metastatic pathway is modulated by the expression of miRNA-138 and potentially other miRNA within the signature. QRT- PCR experiments confirmed findings from prior publication that the overexpression of miRNA-138 reduced the mRNA level of RhoC and ROCK2 by a mean of 2.6 (SEM:0.08) and 3.1 (SEM:0.22) folds respectively. Western blot analysis demonstrated the predicted down-regulation of RhoC with the transfection of miRNA- 138 (Figure 9b). However, paradoxically increased ROCK1 and ROCK2 levels were observed in miRNA-138 transfected cells. The activities of ROCKs were also increased as observed by the phosphorylation of their downstream target, LIMK1 and LIMK2 (Figure 9b, 10a and b).

In silico analysis of the prognostic value of signature in other cancers:

Application of the signature was explored in the breast cancers dataset (n=856) due to 1) the similarity between the observed morphological changes in UPS cells and previously described changes in breast cancer cells following miRNA-138 and Rho modulation ²⁶ , 2) prior success in cross-validating the "cinsarc" sarcoma-derived mRNA signature ²⁷ and 3) the breast dataset being the largest dataset containing miRNA, mRNA and clinical data. The 6-miR signature demonstrated a trend (HR: 1.5, 95%Confidence Interval: 0.9-2.5; p=0.13) for its ability to predict for overall survival after adjusting for patient age, gender and disease stage in melanoma (Table 4b).

Table 4b: Cox-regression proportional hazard analysis of the 6-miR signature adjusting for clinical factors in Melanoma TCGA database

N=129; 67 deaths

In breast cancers, the signature was prognostic for patient overall survival (HR: 1.8, 95% Confidence Interval: 1.2-2.7; p=0.005) after adjusting for patient age, disease stage and ER status. We then postulated that the expressions of the genes (RhoA, RhoC, ROCK1 , ROCK2, LIMK1 , LIMK2, CFL1 and CFL2) within the cofilin pathway (Table 4c; Figure 10) and DICER (miRNA processing) may also be correlated with patient survival.

Table 4c: Cox-regression proportional hazard analysis of the 6-miR signature adjusting for clinical factors in Breast cancers from the TCGA database.

N=856; 105 deaths Indeed, they were all significantly (p<0.042) associated with overall survival on univariate analysis. Following multivariate analysis and backward selection modeling, 6-miR signature score (HR:1.8; p=0.01), LIMK1(HR:0.6; p=0.012), RhoA (HR:0.6, p=0.013) and DICER (HR:1.8; p=0.01) remain associated with overall survival after adjusting for patient age, disease stage and ER status (Table 6). The associations of the 3 genes (RhoA, LIMK1 and DICER) with breast cancer patient DMFS were further tested using 1056 previously profiled breast samples from 7 datasets. Low RhoA expression was significantly (univariate Log-rank p=0.01) associated with shorter DMFS (Figure 11). No multivariate analysis was performed due to lack of clinical annotations. Table 6: Cox proportional hazard regression analysis of the 6-miR signature adjusting for clinical factors in Breast cancers from the TCGA database.

N=856; 105 deaths; med an u: 7.7 mont s

Two studies had previously described prognostic molecular signatures for STS composed of 3 hypoxia related mRNA expressions and a 177 mRNA gene signature. Yet, neither of these signatures has been validated ^28,29 . Chibon et al. published the "cinsarc" signature composed of 67 differentially expressed mRNAs related to cell cycle progression and chromosomal stability ²⁷ that is able to dichotomize patients with STS, breast cancers and lymphoma into high and low risk groups for the development of metastasis. While the "cinsarc" signature demonstrated that increasing genomic complexity of tumors is associated with higher risk of metastasis, the genes involved in the signature were selected from pathways related to mitosis control and chromosomal integrity and are thus not specific to STS or likely to be druggable targets ^30,31 .

Our current study used a homogeneous subtype of STS, the UPS to 1) investigate the use of miRNAs to prognosticate patients and 2) derive biological understanding of the mechanisms by which STS metastasize. The current study developed and validated a 6-miR signature that predicts for DMFS (HR: 3.5) in UPS independently from other prognostic factors such as patient age, gender, tumor size, grade, depth and use of adjuvant treatment (Table 4a). The validity of the signature and the potential biological roles of the 6-miRNAs were further supported by the higher risk scores observed in metastatic samples when compared to the risk scores of the corresponding primaries (Figure 1 b). The increasing risk score from "Low-risk" primaries to "High-risk" primaries to metastasis supports a biological selection for cells with a pattern of miRNA expression that promotes UPS metastasis.

Phenotypic screening of candidate miRNAs (Table 5 and Figure 3a) and the correlation between higher expression of miRNA-138 with increased risk of metastasis (Figure 4) encouraged further investigation on the biological functions of miRNA-138. Examination of potential downstream targets of miRNA-138 combined with the validation that reduced RhoA (Figure 2a) is associated with metastasis suggests that the cofilin pathway ^32,33 is involved in the promotion of UPS metastasis (Figure 10a). Functional assays in-vitro confirmed the inhibitory effect of miRNA-138 on RhoC, disinhibiting RhoA to activate downstream effector ROCKs and LIMKs (Figure 2b). These molecular changes resulted in observed losses in the spindled cell shape of pre-miRNA-138 transfected sarcoma cells (Figure 6b), corroborating with previously observed morphological changes secondary to increased RhoA-ROCK activity in prostate and breast cancer cells ²⁶ .

Given the similarity between the phenotypic changes secondary to miRNA-138 modulation in UPS (Figure 6b) and breast cancer cells ²⁶ , we postulated that the 6-miR signature may be prognostic in breast cancer and explored the signature's value using the TCGA breast cancer dataset. Remarkably, the 6-miR signature along with 3 genes from the cofilin pathway (RhoA and LIMK1) and miRNA machinery (DICER) were prognostic for breast cancer patient OS after adjusting for patient age, disease stage and ER status (Table 4c). The association between the 3 genes and breast cancer patient outcome was further tested using 1056 publically available mRNA profiles from breast cancers annotated with patient DMFS from 7 studies ^17,19"23 . Indeed, RhoA low- expression was significantly (log-rank p=0.01) associated with worsened DMFS (Figure 11), as found in UPS (Figure 9a). These results suggest a common mechanism involving miRNA and RhoA may exist in UPS and breast cancer in promoting metastasis. Contrary to prior findings in which RhoA and RhoC were thought to be prometastatic and involved in epithelial-mesenchymal transition (EMT) · , the current results derived from sarcoma and breast cancer clinical and in-vitro data suggest that reduced expression of RhoA and RhoC is associated with higher metastatic rates. Our unanticipated finding is supported by 2 recent publications that demonstrated the importance of cellular plasticity in cancer cells to undergo EMT followed by a reversion through mesenchymal-epithelial transition to colonize the target metastatic environment and grow ^40,41 . Furthermore, given the complexity of the metastatic process, miRNA-138 probably works in partnership with other molecular changes, such as RhoA and the other 5-miRNAs within the UPS prognostic signature to induce and/or promote biological changes in patients (Figure 10c).

Nair et al.'s systematically reviewed publications dedicated to study the role of miRNAs in predicting clinical outcomes in cancers and identified 41 studies on 20 human cancers with at least 10 samples ⁴² . Among these studies, the median study size was 65 samples, and only 6 (13%) of the studies proceeded to validate their findings using independent cohorts, with 3(7%) of the studies using multivariate adjustments in their analysis. While the current study's sample size (n=110) is at par with other published miRNA biomarker studies, the robustness of our investigation was optimized through the use of central pathology review to ensure a homogeneous subtype (UPS) of a rare cancer (STS) was profiled than validated using an independent sets of UPS with prospectively annotated clinical data. After adjusting for other potential prognostic factors, the signature remains significantly associated with DMFS. The findings were further corroborated using metastatic samples, functional assays and independent breast cancer datasets from 1912 samples.

Although preferred embodiments of the invention have been described herein, it will be understood by those skilled in the art that variations may be made thereto without departing from the spirit of the invention or the scope of the appended claims. All documents disclosed herein are incorporated by reference.

References:

1. Society AC: Cancer Facts & Figures 2013. Atlanta, American Cancer Society, 2013

2. Canter RJ, Beal S, Borys D, et al: Interaction of histologic subtype and histologic grade in predicting survival for soft-tissue sarcomas. Journal of the American College of Surgeons 210:191-198 e2, 2010

3. Folpe AL, Inwards CY: Bone and soft tissue pathology. Philadelphia, PA, Saunders/Elsevier, 2010

4. Gutierrez JC, Perez EA, Franceschi D, et al: Outcomes for soft-tissue sarcoma in 8249 cases from a large state cancer registry. J Surg Res 141 :105-14,

2007

5. Lehnhardt M, Daigeler A, Homann HH, et al: MFH revisited: outcome after surgical treatment of undifferentiated pleomorphic or not otherwise specified (NOS) sarcomas of the extremities ~ an analysis of 140 patients. Langenbecks Arch Surg 394:313-20, 2009

6. Le Cesne AVG, M; Woll, P. J.; Bramwell, V. H.; Casali.P. G.;

Hoekstra.H. J.; Reichardt, P.; Hogendoorn, P. C; Hohenberger, P.; Blay, J. Y. : The end of adjuvant chemotherapy (adCT) era with doxorubicin-based regimen in resected high-grade soft tissue sarcoma (STS): Pooled analysis of the two STBSG-EORTC phase III clinical trials. . Presented at the ASCO annual meeting, Chicago, 2008

7. Woll PJ, Reichardt P, Le Cesne A, et al: Adjuvant chemotherapy with doxorubicin, ifosfamide, and lenograstim for resected soft-tissue sarcoma (EORTC 62931): a multicentre randomised controlled trial. Lancet Oncol 13:1045-54, 2012

8. Cheng H, Dodge J, Mehl E, et al: Validation of immature adipogenic status and identification of prognostic biomarkers in myxoid liposarcoma using tissue microarrays. Hum Pathol 40:1244-51 , 2009

9. Hisaoka M, Matsuyama A, Nagao Y, et al: Identification of altered MicroRNA expression patterns in synovial sarcoma. Genes, chromosomes & cancer 50:137-45, 2011

10. Subramanian S, Lui WO, Lee CH, et al: MicroRNA expression signature of human sarcomas. Oncogene 27:2015-26, 2008

11. Hui AB, Yue S, Shi W, et al: Therapeutic efficacy of seliciclib in combination with ionizing radiation for human nasopharyngeal carcinoma. Clin Cancer Res 15:3716-24, 2009

12. Hui AB, Shi W, Boutros PC, et al: Robust global micro-RNA profiling with formalin-fixed paraffin-embedded breast cancer tissues. Lab Invest 89:597-606, 2009

13. Livak KJ, Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25:402-8, 2001 14. Hui AB, Bruce JP, Alajez NM, et al: Significance of dysregulated metadherin and microRNA-375 in head and neck cancer. Clinical cancer research : an official journal of the American Association for Cancer Research 17:7539-50, 2011

15. Reeves GK, Travis RC, Green J, et al: Incidence of breast cancer and its subtypes in relation to individual and multiple low-penetrance genetic susceptibility loci. JAMA 304:426-34, 2010

16. Wang Y, Jatkoe T, Zhang Y, et al: Gene expression profiles and molecular markers to predict recurrence of Dukes' B colon cancer. J Clin Oncol 22:1564-71 , 2004

17. Desmedt C, Piette F, Loi S, et al: Strong time dependence of the 76- gene prognostic signature for node-negative breast cancer patients in the TRANSBIG multicenter independent validation series. Clin Cancer Res 13:3207-14, 2007

18. Kao KJ, Chang KM, Hsu HC, et al: Correlation of microarray-based breast cancer molecular subtypes and clinical outcomes: implications for treatment optimization. BMC Cancer 11 :143, 2011

19. Loi S, Haibe-Kains B, Desmedt C, et al: Definition of clinically distinct molecular subtypes in estrogen receptor-positive breast carcinomas through genomic grade. J Clin Oncol 25:1239-46, 2007

20. Loi S, Haibe-Kains B, Desmedt C, et al: Predicting prognosis using molecular profiling in estrogen receptor-positive breast cancer treated with tamoxifen. BMC Genomics 9:239, 2008

21. Minn AJ, Gupta GP, Siegel PM, et al: Genes that mediate breast cancer metastasis to lung. Nature 436:518-24, 2005

22. Schmidt M, Bohm D, von Tome C, et al: The humoral immune system has a key prognostic impact in node-negative breast cancer. Cancer Res 68:5405-13,

2008

23. Symmans WF, Hatzis C, Sotiriou C, et al: Genomic index of sensitivity to endocrine therapy for breast cancer. J Clin Oncol 28:4111-9, 2010

24. Liu X, Wang C, Chen Z, et al: MicroRNA-138 suppresses epithelial- mesenchymal transition in squamous cell carcinoma cell lines. The Biochemical journal 440:23-31 , 2011

25. Jiang L, Liu X, Kolokythas A, et al: Downregulation of the Rho GTPase signaling pathway is involved in the microRNA-138-mediated inhibition of cell migration and invasion in tongue squamous cell carcinoma. Int J Cancer 127:505-12, 2010

26. Vega FM, Fruhwirth G, Ng T, et al: RhoA and RhoC have distinct roles in migration and invasion by acting through different targets. The Journal of cell biology 193:655-65, 2011

27. Chibon F, Lagarde P, Salas S, et al: Validated prediction of clinical outcome in sarcomas and multiple types of cancer on the basis of a gene expression signature related to genome complexity. Nature medicine 16:781-7, 2010 28. Francis P, Namlos HM, Muller C, et al: Diagnostic and prognostic gene expression signatures in 177 soft tissue sarcomas: hypoxia-induced transcription profile signifies metastatic potential. BMC genomics 8:73, 2007

29. Hoffmann AC, Danenberg KD, Taubert H, et al: A three-gene signature for outcome in soft tissue sarcoma. Clinical cancer research : an official journal of the

American Association for Cancer Research 15:5191-8, 2009

30. Dixon SJ, Stockwell BR: Identifying druggable disease-modifying gene products. Curr Opin Chem Biol 13:549-55, 2009

31. Verdine GL, Walensky LD: The challenge of drugging undruggable targets in cancer: lessons learned from targeting BCL-2 family members. Clin Cancer Res 13:7264-70, 2007

32. Wang W, Eddy R, Condeelis J: The cofilin pathway in breast cancer invasion and metastasis. Nature reviews. Cancer 7:429-40, 2007

33. Nanjappa V, Raju R, Muthusamy B, et al: A Comprehensive Curated Reaction Map of Leptin Signaling Pathway. J Proteomics Bioinform 4:184-189, 2011

34. Bellovin Dl, Simpson KJ, Danilov T, et al: Reciprocal regulation of RhoA and RhoC characterizes the EMT and identifies RhoC as a prognostic marker of colon carcinoma. Oncogene 25:6959-67, 2006

35. Gjorevski N, Boghaert E, Nelson CM: Regulation of Epithelial- Mesenchymal Transition by Transmission of Mechanical Stress through Epithelial Tissues. Cancer microenvironment : official journal of the International Cancer Microenvironment Society 5:29-38, 2012

36. Ma L, Liu YP, Zhang XH, et al: Relationship of RhoA signaling activity with ezrin expression and its significance in the prognosis for breast cancer patients. Chinese medical journal 126:242-7, 2013

37. Pille JY, Denoyelle C, Varet J, et al: Anti-RhoA and anti-RhoC siRNAs inhibit the proliferation and invasiveness of MDA-MB-231 breast cancer cells in vitro and in vivo. Molecular therapy : the journal of the American Society of Gene Therapy 11 :267-74, 2005

38. Struckhoff AP, Rana MK, Worthylake RA: RhoA can lead the way in tumor cell invasion and metastasis. Frontiers in bioscience : a journal and virtual library 16:1915-26, 2011

39. Thiery JP, Acloque H, Huang RY, et al: Epithelial-mesenchymal transitions in development and disease. Cell 139:871-90, 2009

40. Ocana OH, Corcoles R, Fabra A, et al: Metastatic colonization requires the repression of the epithelial-mesenchymal transition inducer Prrxl . Cancer cell 22:709-24, 2012

41. Tsai JH, Donaher JL, Murphy DA, et al: Spatiotemporal regulation of epithelial-mesenchymal transition is essential for squamous cell carcinoma metastasis. Cancer cell 22:725-36, 2012 42. Nair VS, Maeda LS, loannidis JP: Clinical outcome prediction by microRNAs in human cancer: a systematic review. J Natl Cancer Inst 104:528-40, 2012

43. Ma L, Reinhardt F, Pan E, et al: Therapeutic silencing of miR-10b inhibits metastasis in a mouse mammary tumor model. Nature biotechnology 28:341-7, 2010

44. Stenvang J, Petri A, Lindow M, et al: Inhibition of microRNA function by antimiR oligonucleotides. Silence 3:1 , 2012

45. Valastyan S, Chang A, Benaich N, et al: Activation of miR-31 function in already-established metastases elicits metastatic regression. Genes & development

25:646-59, 2011