WO2020104482A1 - Methods for predicting metastatic potential in patients suffering from sdhb-mutated paraganglioma - Google Patents

Abstract

Pheochromocytomas and paragangliomas (PPGL) are neuroendocrine tumors explained by germline or somatic mutations in about 70% of cases. Patients with SDHB mutations are at high- risk of developing a metastatic disease, yet no biomarkers are available to predict metastatic potential. Here, the inventors performed a comprehensive analysis of long non-coding RNAs (lncRNAs) using a mining approach of transcriptome data from a well-characterized series of 187 PPGL. They aimed at identifying lncRNAs specific for molecular groups and for metastatic progression of SDHB-mutated tumors. Consensus clustering analyses identified four lncRNA-based subtypes strongly correlated with mRNA expression clusters. This classification was validated in an independent series of 51 PPGL. Receiver operating characteristic curve analyses identified one putative lncRNAs (GenBank: BC063866) that accurately discriminates metastatic from benign tumors in patients at high-risk of progression. Expression of this transcript was validated by RT- qPCR in both discovery and validation series of PPGL. Moreover, cox proportional hazards regression analysis for metastasis-free survival (MFS) demonstrated that BC063866 is an independent risk factor associated with poor clinical outcome of SDHx carriers (log-rank test P = 2.29×10^-05). The findings extend the spectrum of transcriptional dysregulations in PPGL to lncRNAs and provide a novel biomarker that could be useful to identify potentially metastatic tumors in patients carrying SDHx mutations.

Description

WO 2020/104482 PCT/EP2019/081848

METHODS FOR PREDICTING METASTATIC POTENTIAL IN PATIENTS SUFFERING FROM SDHB-MUTATED PARAGANGLIOMA

FIELD OF THE INVENTION:

The present invention relates to methods for predicting metastatic potential in patients suffering from SDHB-mutated paraganglioma.

BACKGROUND OF THE INVENTION:

Pheochromocytomas and paragangliomas (PPGL) are tumours of the adrenal medulla or extra-adrenal paraganglia respectively, which exhibit a high degree of heritability. In fact,

40% of PPGL can be explained by germline mutations in one of 15 susceptibility genes comprising SDH A, SDHB, SDHC, SDHD, SDHAF2 (referred to as SDHx), FH, SLC25A11, MDH2, GOT2, VHL, KIF1B, RET, NF1, TMEM127 and MAX [1], whereas 35% of tumors harbor somatic mutations in some of these genes or in recently identified drivers [2] .

While most of these tumors are benign, metastatic disease occurs in up to 20% of PPGL with SDHB mutation carriers having a high-risk of malignant progression and worse prognosis [3, 4] Germline mutations in FH and SLC25A11 genes have also been associated with metastatic PPGL [5] [6]. However malignancy remains difficult to predict given the lack of robust histopathological criteria and tumor biomarkers to identify the metastatic potential of primary tumors [7] [8]. Hence, malignant tumors are still defined only by the occurrence of first metastases testified by the presence of PPGL tumor cells in non-chromaffin tissues.

Genomic analyses of PPGL established five molecular subtypes and showed that mutations in susceptibility genes are the main drivers of tumorigenesis [9]. Transcriptomic cluster CIA is characterized by high-risk tumors harboring mutations in SDHx/FH/SLC25Al 1/MDH2/GOT2 genes that participate in mitochondrial metabolism. These tumors display a DNA hypermethylated profile [10] and an EMT-like phenotype [11] Cluster C1B contains VHL- mutated tumors that display a glycolytic profile, and along with tumors from cluster CIA, expression signatures of hypoxia and angiogenesis [12] Cluster C2A includes tumors with germline or somatic mutations in RET/NF1/TMEM127/MAX/MET/FGFR1 and HRAS genes that upregulate the kinase signaling [12, 13], and clusters C2B and C2C are enriched in sporadic tumors [12] Additionally, the TCGA study reported that MAML3 fusion gene and CSDE1 somatic mutation define a Wnt-altered subtype [14]

While no recurrent alterations linked to metastatic progression were identified in those studies, genetic/epigenetic alterations affecting the TERT promoter region [15-17], as well as overexpression of miR-183 and miR-483 were reported in metastatic PPGL with SDHB mutations WO 2020/104482 PCT/EP2019/081848

[18]. Moreover, we recently identified a six-miRNA signature strongly associated with a short time to progression, regardless of the genotype [19]. These results highlight the importance of analyzing the noncoding region of the genome.

It has been suggested that up to 75% of the human genome gets transcribed into RNA, whereas only about 3% corresponds to exons of protein-coding genes [20]. Long non-coding RNAs (IncRNAs) are RNA molecules longer than 200 nucleotides that lack protein-coding capacity and are expressed in a tissue/developmental stage- specific manner. A large collection of IncRNAs are encoded in the human genome [21] and are involved in biological processes that are crucial for tumorigenesis such as cell cycle regulation, proliferation, apoptosis, migration, and genomic stability[22] . These molecules regulate gene expression of local or distal genes by controlling chromatin remodeling, RNA splicing and stability of mRNAs and miRNAs [23, 24] Regardless of the regulatory mechanisms, IncRNAs are becoming a new class of cancer biomarkers [25]. Notably, despite the overwhelming evidence about dysregulation of IncRNAs in many types of human cancers [25-27], very few is known regarding the expression of IncRNAs in PPGL.

SUMMARY OF THE INVENTION:

The present invention relates to methods for predicting tumour aggressiveness in patients suffering from SDHx-mutated paraganglioma. In particular, the present invention is defined by the claims.

DETAILED DESCRIPTION OF THE INVENTION:

Pheochromocytomas and paragangliomas (PPGL) are neuroendocrine tumors explained by germline or somatic mutations in about 70% of cases. Patients with SDHB mutations are at high- risk of developing a metastatic disease, yet no biomarkers are available to predict metastatic potential. Here, the inventors performed a comprehensive analysis of long non-coding RNAs (IncRNAs) using a mining approach of transcriptome data from a well-characterized series of 187 PPGL. They aimed at identifying IncRNAs specific for molecular groups and for metastatic progression of 577/7/1- mutated tumors. Consensus clustering analyses identified four IncRNA-based subtypes strongly correlated with mRNA expression clusters (chi² p-values from 1.38xl0 ³² to 1.07xl0 ⁶⁷). This classification was validated in an independent series of 51 PPGL. Co-expression analyses suggest that differentially expressed IncRNAs participate in regulating signaling pathways that characterize PPGL subtypes. Receiver operating characteristic curve analyses identified one putative IncRNA (GenBank: BC063866) that accurately discriminates metastatic from benign tumors in patients with SDHx mutations (AUC 0.95; P = 4.59xlO ⁰⁵). Moreover, cox proportional hazards regression analysis for metastasis-free survival (MFS) demonstrated that BC063866 is an WO 2020/104482 PCT/EP2019/081848 independent risk factor associated with poor clinical outcome of SDHx carriers (log-rank test P = 2.29xlO ⁰⁵). Notably this putative IncRNA have also been reported in aggressive neuroblastoma, another neural crest-derived tumor. The findings extend the spectrum of transcriptional dysregulations in PPGL to IncRNAs and provide a novel biomarker that could be useful to identify potentially metastatic tumors in patients carrying SDHx mutations.

Accordingly, the present invention relates to a method of identifying the metastatic potential of a tumor in a patient carrying at least one SDHB-mutation comprising i) determining the expression level of the putative long non-coding BC063866 in a tumor sample obtained from the patient, ii) comparing the expression level determined at step i) with a predetermined reference value and wherein detecting differential between the expression level determined at step i) and the predetermined reference value indicates whether the patient is at risk of developing a metastatic tumour.

As used herein, the term“paraganglioma” has its general meaning in the art and designates both functioning (catecholamine-secreting) and nonfunctioning tumors arising in the paraganglia outside or in the adrenal gland. As used herein, the term .V/l Z/i-mutated paraganglioma refers to paraganglioma associated with at least one SDHB mutation.

As used herein the term“SDHB” has its general meaning in the art and refers to the iron-sulfur protein (IP) subunit of succinate dehydrogenase (SDH) that is involved in complex II of the mitochondrial electron transport chain and is responsible for transferring electrons from succinate to ubiquinone (coenzyme Q). An exemplary amino acid sequence is represented by SEQ ID NO:l.

SEQ ID NO:l >sp | P21912 | SDHB_HUMAN Succinate dehydrogenase

[ubiquinone] iron-sulfur subunit, mitochondrial OS=Homo sapiens

OX=9606 GN=SDHB PE=1 SV=3

MAAVVALSLRRRLPATTLGGACLQASRGAQTAAATAPRIKKFAIYRWDPDKAGDKPHMQT YEVDLNKCGPMVLDALIKIKNEVDSTLTFRRSCREGICGSCAMNINGGNTLACTRRIDTN LNKVSKIYPLPHMYVIKDLVPDLSNFYAQYKS IEPYLKKKDESQEGKQQYLQS IEEREKL DGLYECILCACCSTSCPSYWWNGDKYLGPAVLMQAYRWMIDSRDDFTEERLAKLQDPFSL YRCHTIMNCTRTCPKGLNPGKAIAEIKKMMATYKEKKASV

As used herein the term“ SDHB mutation” refers to any mutation that could occur in the SDHB gene and that is associated with paraganglioma progression. Any SDH mutation is encompassed by the invention, including point mutations, inversion, translocations, deletions, frame shifts... More than 150 mutations in the SDHB gene have been identified. Most of the inherited SDHB gene mutations change single protein building blocks (amino acids) in the SDHB protein sequence or result in a shortened protein. As a result, there is little or no SDH enzyme activity. Exemplary mutations are described in the literature [28] [29] [30] and in the LOYD Database (https://databases.lovd.nl/shared/genes/SDHB) and are encompassed in the WO 2020/104482 PCT/EP2019/081848 invention. Mutations in SDH may be identified by any suitable method in the art, but in certain embodiments the mutations are identified by one or more of polymerase chain reaction, sequencing or next- generation sequencing, and validated by SDHB immunohistochemistry

[31].

As used herein, the term "Risk" in the context of the present invention, relates to the probability (i.e. at least 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99% of risk) that an event will occur over a specific time period (e.g. 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 years), as in the conversion to metastatic tumor, and can mean a subject's "absolute" risk or "relative" risk. Absolute risk can be measured with reference to either actual observation post measurement for the relevant time cohort, or with reference to index values developed from statistically valid historical cohorts that have been followed for the relevant time period. Relative risk refers to the ratio of absolute risks of a subject compared either to the absolute risks of low risk cohorts or an average population risk, which can vary by how clinical risk factors are assessed. Odds ratios, the proportion of positive events to negative events for a given test result, are also commonly used (odds are according to the formula p/(l-p) where p is the probability of event and (1- p) is the probability of no event) to no- conversion. "Risk evaluation," or "evaluation of risk" in the context of the present invention encompasses making a prediction of the probability, odds, or likelihood that an event or disease state may occur, the rate of occurrence of the event or conversion to metastatic tumor. Risk evaluation can also comprise prediction of future clinical parameters, traditional laboratory risk factor values (e.g. primary tumor size (>5 cm) and extra-adrenal location), or other indices, either in absolute or relative terms in reference to a previously measured population. The methods of the present invention may be used to make continuous or categorical measurements of the risk of conversion to metastatic tumor, thus diagnosing and defining the risk spectrum of a category of subjects defined as being at risk of developing a metastatic tumor. In the categorical scenario, the invention can be used to discriminate between normal and other subject cohorts at higher risk of developing a metastatic tumor. Thus, the terms "high risk", "intermediate risk" and "low risk" refers to differences in the individual predisposition for developing a disease, disorder, complication or susceptibility therefor. Said high, intermediate or low risk can be statistically analyzed. Preferably, the differences between a subject or a group of subjects having a high, intermediate or low risk are statistically significant. This can be evaluated by well-known statistic techniques including Student's t-Test, Chi2-Test, Wilcoxon-Mann-Whitney Test, Kurskal- Wallis Test or Fisher's exact Test, log-rank test, logistic regression analysis, or Cox models. Most preferably, the risk groups are analyzed as described in the accompanied WO 2020/104482 PCT/EP2019/081848

Examples whereby explorative data analysis is carried out and the risk groups are formed with respect to the median, the 25% and the 75% percentiles. Differences in continuous variables of the groups are tested by Wilcoxon-Mann-Whitney Test or Kurskal- Wallis Test depending on the number of groups to be compared. For nominal or ordered categories, Fisher's exact or Chi2- Test for trend are applied. Without further ado, the person skilled in the art can carry out multivariant analysis with stratified versions of the aforementioned tests or Cox models in order to examine the independent impact of predictive factors and to establish the different risk groups.

The method of the present invention is also suitable for predicting the patient’s survival, in particular, the metastasis-free survival. As used herein, the term“metastasis-free survival” (MFS) or“distant metastasis-free survival” (DMFS) refers to the period after a curative treatment, when no disease can be detected, until a metastasis is detected. In the case of“distant metastasis-free survival”, such metastasis is detected in a tissue which is not primary tumor tissue. As used herein, the expression“short survival time” indicates that the patient will have a survival time that will be lower than the median (or mean) observed in the general population of patients suffering from said tumour. When the patient will have a short survival time, it is meant that the patient will have a“poor prognosis”. Inversely, the expression“long survival time” indicates that the patient will have a survival time that will be higher than the median (or mean) observed in the general population of patients suffering from said tumour. When the patient will have a long survival time, it is meant that the patient will have a“good prognosis”.

As used herein, the term“long non-coding RNA” or“IncRNA” refers to a non-protein coding RNA transcript longer than 200 nucleotides. The nature of the long non-coding RNAs of the present invention is based on bioinformatic predictions of protein-coding potential. The nucleic acid sequence for BC063866 is represented by SEQ ID NO:2 detailed below:

SEQ ID NO : 2 GenBank : BC063866.1

1 tttgattcct atctcgttca aatttttggt tcatcgtcac ctcaacctgg atttgggatg 61 tcaggggaag aactcagtga atctactcca gagcctcaaa aagaaatttc tgagtcattg 121 agtgtcacca gagaccagga tgaagatgat aaggctccag agccaacgtg ggctgatgat 181 ctgcctgcca ctacctcatc tgaggccacc accaccccca ggccactgct cagcacccct 241 gtggatgggg cagaggatcc tagatgtttg gaagccttga agcctggaaa ctgtggtgaa 301 tatgtggttc gatggtatta tgacaaacag gtcaactctt gtgcccgatt ttggttcagt 361 ggctgtaatg gctcaggaaa tagattcaac agtgaaaagg aatgtcaaga aacctgcatt 421 caaggatgag caagtaaatt ggcctgtctc tatcaaaagc atagaactcc ctaatttcca 481 catattcacc caatacaaat acagcactat atttgagtgt atactgagta tttacaactt 541 atacatgtaa ttgaattctc actacagccc taggatgtac atattattaa ccacttatat 601 aggtaagaaa gctgaggctc tgagaagttt agtaacttgt caactgtcac ccaactaaaa 661 agtttcagag ctgaggattt agacttagag ctgtgtaact tcaatacaca gactctatct 721 acttcacaac ctgcaatgtg attctgattc ctttaattcc tgttgtatgt actatgtcag 781 ctcaaacccc tacccctgtc cctgcccata cctccaccca ctcacctccc taacctcctt WO 2020/104482 PCT/EP2019/081848

841 atgtccctca cagtagcaag atgtaggtga taggaaggac ttcggtgtga gaattagaaa

901 tgatgtaaat gtttacgcag gagtgctggg ataggagtcg ggatggtgag ggtagttaga

961 tttttgcctc acttgccctg aaagtggtaa tagggagaaa ccaatctgaa ttacaattac

1021 ttaaatgtat cacagactgt cactttgtat tcctccaaca tgtttggtaa caagtgttta

1081 atgtatgtta aaacaaagaa ggtttttata cccttccatt aaaatatgtc agtgggccct

1141 tccattttat ggagtggaat gggaaggccc ttgacagcca ggaaccactt gaagttggca

1201 tccactcttg aacagtgtgt attaaagaca ggattcacac tgaaaagtga gccaccaaat

1261 tgaaatgcga gtaatggaga gtaacggaga attgactctg tcattttata atacttttgc

1321 aagatcctgt ctgggaagag cattagttct gagctgagtt cagacagcac ttgcatattt

1381 ctttcttttt tttttttttc ccccagacag agtcttgctc tgcactccag cctgggtgac

1441 acagcaagac tctgtctcaa aacaaaaaca aaaacaaaaa acaaaatact ttctccttct

1501 atcaggtcta ccctcattca attcaacagt gcaaaaaaga gcactcacgg aagaaggaga

1561 tggaagaata gccatcttgg cagaaataat gatatgtata atttttttat ttaattaaaa

1621 ttgtattatg taatatttga tactacaaaa gataatgcaa catgtttgtc aatttttttt

1681 attatacttt aagttttagg gtacatgtgc acaaagtaca gatttgttac atatgtattc

1741 atgtgccatg ttggtgtgct gcacccatta actcgtcatt taacattagg tatatttcct

1801 aatgctatcc ctccccactc ccccgccacc ccacaacagg ccctggtgtg tgatgttccc

1861 cttcctgtgt ccatgtgttc tcattgttca actcccacct atgagtgaga acatgcggtg

1921 tttggttttt tgtccttgtg acagtttgct gagaatgatg gtttccagct tcatccatgc

1981 ccctacaaag gacatgaact catcattttt tatggctgca tagtattctg tggtgtatat

2041 gtaacatgct tgtcaattta aaaacatgat aagatgaaca ccagtgaaca catcacctag

2101 agaatgaagc aaacatcaaa aattctattg cagcttcccg tgtggccctc cgccatccca

2161 gctctctatt ccatgccagg gatatcccat tttgaacttc gtatttacca tgtcaatgtg

2221 ctgttgacat ttataggtaa ttttacatgt aaaactctat atatctctat gtgttgccaa

2281 acaataaata tttagccttg cttattttac agttttaaaa aaaaaaaaaa aaa

As used herein, the term“tumor sample” refers to a sample obtained from the tumor of the patient. The tumor sample may be obtained from the patient by routine measures known to the person skilled in the art, i.e., biopsy taken by aspiration or punctuation, excision or by any other surgical method leading to biopsy or resected cellular material.

As used herein, the term "determining" refers to both quantitative and semi-quantitative determinations.

As used herein, the term "expression level" refers to the quantity of the long non-coding RNA. Such quantity may be expressed in the absolute terms, i.e., the total quantity of the polynucleotide in the tumor sample, or in the relative terms, i.e., the concentration of the polynucleotide in the sample.

The expression level of the IncRNAs can be detected or measured by a variety of methods including, an amplification assay, a hybridization assay, a sequencing assay, or an array. Non-limiting examples of such methods include reverse-transcription polymerase chain reaction (RT-PCR); quantitative real-time PCR (qRT-PCR); quantitative PCR, such as TaqMan®; Droplet digital PCR; Northern blotting; in situ hybridization assays; microarray analysis, e.g., microarrays from Nano String Technologies; multiplexed hybridization-based assays, e.g., QuantiGene 2.0 Multiplex Assay from Panomics; serial analysis of gene expression (SAGE); cDNA-mediated annealing, selection, extension, and ligation; nucleic acid immunoassay, direct sequencing or pyrosequencing; massively parallel sequencing; next WO 2020/104482 PCT/EP2019/081848 generation sequencing; high performance liquid chromatography (HPLC) fragment analysis; capillarity electrophoresis; mass spectrometry, including SELDI, MALDI; and other known methods.

Various methods involving amplification reactions and/or reactions in which probes are linked to a solid support and used to quantify RNA may be used. Alternatively, the RNA, or DNA copy of the RNA, may be linked to a solid support and quantified using a probe to the sequence of interest.

In some embodiments, the target RNA is first reverse transcribed and the resulting cDNA is quantified. In some embodiments, RT-PCR or other quantitative amplification techniques are used to quantify the target RNA. Amplification of cDNA using PCR is well known (see U.S. Patents 4,683,195 and 4,683,202; PCR PROTOCOLS: A GUIDE TO METHODS AND APPLICATIONS (Innis et al, eds, 1990)). Methods of quantitative amplification are disclosed in, e.g., U.S. Patent Nos. 6,180,349; 6,033,854; and 5,972,602, as well as in, e.g., Gibson et a , Genome Research 6:995-1001 (1996); DeGraves, et a , Biotechniques 34(1): 106-10, 112-5 (2003); Deiman B, et al, Mol Biotechnol. 20(2): 163-79 (2002). Alternative methods for determining the expression level of the IncRNA in a sample may involve other nucleic acid amplification methods such as ligase chain reaction (Barany (1991) Proc. Natl. Acad. Sci. USA 88: 189-193), self- sustained sequence replication (Guatelli et al. (1990) Proc. Natl. Acad. Sci. USA 87:1874-1878), transcriptional amplification system (Kwoh et al. (1989) Proc. Natl. Acad. Sci. USA 86:1173-1177), Q-Beta Replicase (Lizardi et al. (1988) Bio/Technology 6: 1197), rolling circle replication (U.S. Patent No. 5,854,033) or any other nucleic acid amplification method, followed by the detection of the amplified molecules using techniques well known to those of skill in the art.

In some embodiments, RNA (or a copy) is immobilized on a solid surface and contacted with a probe, e.g., in a microarray, dot blot or Northern format. A skilled artisan can readily adapt known RNA detection methods for use in detecting the expression level of the IncRNA.

In some embodiments, microarrays are employed. DNA microarrays provide one method for the simultaneous measurement of the expression levels of large numbers of genes. Each array consists of a reproducible pattern of capture probes attached to a solid support. Labeled RNA or DNA is hybridized to complementary probes on the array and then detected by laser scanning. Hybridization intensities for each probe on the array are determined and converted to a quantitative value representing relative gene expression levels. See, U.S. Patent Nos. 6,040,138, 5,800,992 and 6,020,135, 6,033,860, and 6,344,316. High-density oligonucleotide arrays are particularly useful for determining the gene expression profile for a WO 2020/104482 PCT/EP2019/081848 large number of R A's in a sample. Techniques for the synthesis of these arrays using mechanical synthesis methods are described in, e.g., U.S. Patent No. 5,384,261. Although a planar array surface is often employed the array may be fabricated on a surface of virtually any shape or even a multiplicity of surfaces. Arrays may be peptides or nucleic acids on beads, gels, polymeric surfaces, fibers such as fiber optics, glass or any other appropriate substrate, see U.S. Patent Nos. 5,770,358, 5,789,162, 5,708,153, 6,040,193 and 5,800,992. Arrays may be packaged in such a manner as to allow for diagnostics or other manipulation of an all-inclusive device. [0065] In some embodiments, gene-specific probes and/or primers are used in hybridization assays to detect RNA expression. The probes and/or primers may be labeled with any detectable moiety or compound, such as a radioisotope, fluorophore, chemiluminescent agent, and enzyme.

Probes and primers can be selected using know algorithms that takes into account binding energies, base composition, sequence complexity, cross-hybridization binding energies, and secondary structure. See, e.g., PCT Patent Publication WO 01/05935, published Jan. 25, 2001; Hughes et al, Nat. Biotech. 19:342-7 (2001).

The probes and primers necessary for practicing the present invention can be synthesized and labeled using well known techniques. Oligonucleotides used as probes and primers may be chemically synthesized according to the solid phase phosphoramidite triester method first described by Beaucage and Caruthers, Tetrahedron Letts., 22: 1859-1862, 1981, using an automated synthesizer, as described in Needham- Van Devanter et al, Nucleic Acids Res. 12:6159-6168, 1984.

In some embodiments, probes can be obtained, e.g., by polymerase chain reaction (PCR) amplification of genomic DNA or RNA or cloned sequences. PCR primers are selected based on a known sequence of the genome that will result in amplification of specific fragments of genomic DNA. Computer programs that are well known in the art are useful in the design of primers with the required specificity and optimal amplification properties, such as Oligo version 5.0 (National Biosciences). Typically each probe is between 10 bases and 50,000 bases, usually between 300 bases and 1,000 bases in length. It will be apparent to one skilled in the art that controlled robotic systems are useful for isolating and amplifying nucleic acids.

The expression level of the IncRNA can be normalized to a reference level for a control gene. The control value can be predetermined, determined concurrently, or determined after a sample is obtained from the subject. The standard can be run in the same assay or can be a known standard from a previous assay. In some embodiments, a normalized expression level of the IncRNA can be transformed into a score for likelihood of progression. WO 2020/104482 PCT/EP2019/081848

Typically, the expression level of the IncRNA is determined as described in the EXAMPLE.

A predetermined reference value can be relative to a number or value derived from population studies, including without limitation, such subjects having similar clinical profile. Such predetermined reference values can be derived from statistical analyses and/or risk prediction data of populations obtained from mathematical algorithms and computed indices of metabolic syndrome. In some embodiments, the predetermined reference values are derived from the expression level in a control sample derived from one or more subjects who were not subjected to the event. Furthermore, retrospective measurement of the expression level in properly banked historical subject samples may be used in establishing these predetermined reference values. The predetermined reference value has to be determined in order to obtain the optimal sensitivity and specificity according to the function of the test and the benefit/risk balance (clinical consequences of false positive and false negative). Typically, the optimal sensitivity and specificity (and so the predetermined reference value) can be determined using a Receiver Operating Characteristic (ROC) curve based on experimental data. For example, after determining the level of the marker in a group of reference, one can use algorithmic analysis for the statistic treatment of the measured levels of the marker in samples to be tested, and thus obtain a classification standard having significance for sample classification. The full name of ROC curve is receiver operator characteristic curve, which is also known as receiver operation characteristic curve. It is mainly used for clinical biochemical diagnostic tests. ROC curve is a comprehensive indicator that reflects the continuous variables of true positive rate (sensitivity) and false positive rate (1 -specificity). It reveals the relationship between sensitivity and specificity with the image composition method. A series of different cut-off values (thresholds or critical values, boundary values between normal and abnormal results of diagnostic test) are set as continuous variables to calculate a series of sensitivity and specificity values. Then sensitivity is used as the vertical coordinate and specificity is used as the horizontal coordinate to draw a curve. The higher the area under the curve (AUC), the higher the accuracy of diagnosis. On the ROC curve, the point closest to the far upper left of the coordinate diagram is a critical point having both high sensitivity and high specificity values. The AUC value of the ROC curve is between 1.0 and 0.5. When AUC>0.5, the diagnostic result gets better and better as AUC approaches 1. When AUC is between 0.5 and 0.7, the accuracy is low. When AUC is between 0.7 and 0.9, the accuracy is moderate. When AUC is higher than 0.9, the accuracy is quite high. This algorithmic method is preferably done with a computer. Existing software or systems in the art may be used for the drawing of the ROC curve, such as: MedCalc WO 2020/104482 PCT/EP2019/081848

9.2.0.1 medical statistical software, SPSS 9.0, ROCPOWER.SAS, DESIGNROC.FOR, MULTIREADER POWER.SAS, CREATE-ROC.SAS, GB STAT VIO.O (Dynamic Microsystems, Inc. Silver Spring, Md., USA), etc.

Typically, when the expression level of BC063866 is lower than its corresponding predetermined reference value, it is concluded that the patient is at risk of developing a metastatic tumour.

In some embodiments, the method of the invention comprises the use of a classification algorithm typically selected from Linear Discriminant Analysis (LDA), Topological Data Analysis (TDA), Neural Networks, Support Vector Machine (SVM) algorithm and Random Forests algorithm (RF) such as described in the Example. In some embodiments, the method of the invention comprises the step of determining the subject response using a classification algorithm. As used herein, the term "classification algorithm" has its general meaning in the art and refers to classification and regression tree methods and multivariate classification well known in the art such as described in US 8,126,690; WO2008/156617. As used herein, the term “support vector machine (SVM)” is a universal learning machine useful for pattern recognition, whose decision surface is parameterized by a set of support vectors and a set of corresponding weights, refers to a method of not separately processing, but simultaneously processing a plurality of variables. Thus, the support vector machine is useful as a statistical tool for classification. The support vector machine non-linearly maps its n-dimensional input space into a high dimensional feature space, and presents an optimal interface (optimal parting plane) between features. The support vector machine comprises two phases: a training phase and a testing phase. In the training phase, support vectors are produced, while estimation is performed according to a specific rule in the testing phase. In general, SVMs provide a model for use in classifying each of n subjects to two or more disease categories based on one k-dimensional vector (called a k-tuple) of biomarker measurements per subject. An SVM first transforms the k-tuples using a kernel function into a space of equal or higher dimension. The kernel function projects the data into a space where the categories can be better separated using hyperplanes than would be possible in the original data space. To determine the hyperplanes with which to discriminate between categories, a set of support vectors, which lie closest to the boundary between the disease categories, may be chosen. A hyperplane is then selected by known SVM techniques such that the distance between the support vectors and the hyperplane is maximal within the bounds of a cost function that penalizes incorrect predictions. This hyperplane is the one which optimally separates the data in terms of prediction (Vapnik, 1998 Statistical Learning Theory. New York: Wiley). Any new observation is then classified as belonging to any one of WO 2020/104482 PCT/EP2019/081848 the categories of interest, based where the observation lies in relation to the hyperplane. When more than two categories are considered, the process is carried out pairwise for all of the categories and those results combined to create a rule to discriminate between all the categories. As used herein, the term "Random Forests algorithm" or "RF" has its general meaning in the art and refers to classification algorithm such as described in US 8,126,690; WO2008/156617. Random Forest is a decision-tree-based classifier that is constructed using an algorithm originally developed by Leo Breiman (Breiman L, "Random forests," Machine Learning 2001, 45:5-32). The classifier uses a large number of individual decision trees and decides the class by choosing the mode of the classes as determined by the individual trees. The individual trees are constructed using the following algorithm: (1) Assume that the number of cases in the training set is N, and that the number of variables in the classifier is M; (2) Select the number of input variables that will be used to determine the decision at a node of the tree; this number, m should be much less than M; (3) Choose a training set by choosing N samples from the training set with replacement; (4) For each node of the tree randomly select m of the M variables on which to base the decision at that node; (5) Calculate the best split based on these m variables in the training set. In some embodiments, the score is generated by a computer program. The algorithm of the present invention can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The algorithm can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application- specific integrated circuit).

In some embodiments, the method of the present invention comprises a) determining the expression level of the putative IncRNA BC063866; b) implementing a classification algorithm on data comprising the determined expression level so as to obtain an algorithm output; c) determining the probability that the patient is at risk of developing a metastatic tumour.

Once it is concluded that the patient presents a high risk of developing a metastatic disease, a lifelong annual follow up by a multidisciplinary team with appropriate expertise must be prescribed [32] It has indeed been shown that a close follow-up of high risk patients allows diagnosing metastasis at an early stage and subsequently improves patients’ survival [33]. This follow-up should encompass specialized biological and imaging exams [34], such as, ¹⁸FDG- PET/CT imaging, which is the most sensitive imaging tool to detect SDHB-related metastases [35]. A such as management will give the opportunity to treat the patient early, at the time of WO 2020/104482 PCT/EP2019/081848 the first metastasis diagnosis, before the spreading of the metastatic disease, which is the main point for a favorable outcome [36].

The invention will be further illustrated by the following figures and examples. However, these examples and figures should not be interpreted in any way as limiting the scope of the present invention.

FIGURES:

Figure 1. Analysis of IncRNAs predictive of metastatic PPGL. (A) ROC analysis for the best probeset that discriminates metastatic from benign tumors within the cluster CIA. The AUC and Wilcoxon test p-value are shown. (B) Box plots show relative expression levels of BC063866 normalized against housekeeping genes (-Delta Ct values) as determined by RT- qPCR in tumors from discovery (on the left) and validation (on the right) series. Asterisks (*) correspond to two-tailed t-test P values ( ***P < 0.001).

Figure 2. Prognostic value of noncoding transcript BC063866. (A) Forest plot of the univariate cox model for metastasis-free survival. The hazard ratio and Wald test p-value for putative IncRNA BC063866 as well as for known risk factors of metastatic PPGL are indicated. Orange highlights p-values <0.01. (B) Forest plot of the multivariate cox model for metastasis- free survival showing the hazard ratio and Wald test p-values for the most significant covariates. (C) Survival curve analysis shows the association of the expression of putative IncRNA BC063866, discretized by the 60% quantile cutoff, with the MFS within the CIA subgroup.

EXAMPLE:

Material & Methods

PPGL cohort

Tumor samples (n=187) from 176 patients with PPGL, collected by the French‘Cortico et Medullosurrenale: les Tumeurs Endocrines’ network [37] [38] [38], were analyzed in this study. Written informed consent for the sample collection and subsequent analyses was obtained from each patient after a full explanation of the purpose and nature of the procedures used. This study was approved by the institutional review board [Comite de Protection des Personnes (CPP) he de France III, June 2012] Genomic characterization of the cohort was previously reported [9]. Mutation status (germline or somatic) for the main PPGL susceptibility genes was recently updated by using a next- generation sequencing approach [39].

Computational analyses

A computational pipeline that annotated probes of the HG-U133 Plus 2.0 Affymetrix microarray targeting known and putative IncRNAs [40] was used for analyses. This approach was applied to transcriptome data corresponding to 187 PPGL generated for a previous study WO 2020/104482 PCT/EP2019/081848

(ArrayExpress entry E-MTAB-733) and on a public gene expression dataset of 51 samples (GEO, GSE67066). GENCODE gene annotation vl5 was used to determine the IncRNA types [41]. Unsupervised classification of IncRNAs was performed using ConsensusClusterPlus R package [42] Each consensus clustering used a specific quantile of the most variant probesets from 10 to 99%. The Median Absolute Deviation (MAD) was used as a variation criterion. The final consensus clusters were obtained from clustering on the co-classification matrix. Pearson was used as distance, Ward.D2 was used as inner linkage and Complete was used as a final linkage. Several metrics were used to determine the number of clusters (dispersion, silhouette width, cophenetic correlation coefficient). In silico analysis of RNA:DNA triplexes were performed using when the LongTarget algorithm [43]. Expression of transcripts (targeted by informative probesets) in normal and cancer tissues, was interrogated using the Genevestigator database [44]

Statistical analyses

An independent component analysis (ICA) was used to find a component related to the metastatic status within the samples of the cluster CIA [45]. ANOVA tests were applied for multigroup comparison to identify differentially expressed IncRNAs between molecular groups using limma R package. To control for multiple testing we measured the local false discovery rate by the Benjamini and Hochberg method. We consider that Affymetrix expression values <5 correspond most likely to noise. Consequently, only mean expression values >5 in the groups for comparisons were taken into account and RT-qPCR was used to better assess the expression differences. Enrichment analyses of biological pathways were performed by applying hypergeometric tests on the lists of mRNAs showing positive (r > 0.5) or negative (r < -0.5) correlations with the lists of IncRNAs up- or down-regulated in each molecular group (ANOVA qvalue < 0.05 and fold-change > 2.0). In total, 13,963 biological pathways collected from KEGG, GO and Biocarta (and related genes) were tested. Receiver operating characteristic curve analysis was applied to identify the best discriminators of metastasis. For continuous variables, univariate and multivariate cox regression models were performed (function coxph, R-package survival). Metastasis-free survival (MFS) curves were calculated according to the Kaplan-Meier method (function Surv, R-package survival) and differences between curves were assessed using the log-rank test (function survdiff, R-package survival).

RT-qPCR and cell fractionation

To validate the expression of the IncRNAs, 1 pg RNA (DNase treated) was subject to retrotranscription with Superscript III kit (Thermo fisher scientific). Quantitative PCR of cDNA preparations were performed with iTaq Universal SYBR Green Supermix (Bio-Rad) and WO 2020/104482 PCT/EP2019/081848 carried out using a CFX96 Real-Time machine (Bio-Rad) by applying the following cycling parameters: 95°C for 5 min, 50 cycles of 95°C for 10 s, 60°C for 20 s, 72°C for 20 s. A melt curve (65-98°C) was generated at the end of each run to verify specificity. Relative quantifications were performed by normalization of IncRNAs Ct expression values against the mean of GAPDH and RPL38 Ct expression values used as reference genes. To determine the subcellular localization of IncRNAs, cytoplasmic and nuclear RNAs were purified (Norgen Biotek Corp kit) from NCTH295 and 293T cells according to the manufacturer's instructions. The sequences of primers (5’-3’) used in this study are as follow:

BC063866 Forward: AGCCACCAAATTGAAATGCGA (SEQ ID N0:3) BC063866 Reverse: TGCAAGTGCTGTCTGAACTC (SEQ ID N0:4) GAPDH Forward: CATGTTCGTCATGGGTGTGAACCA (SEQ ID N0:5) GAPDH Reverse: ATGGCATGGACTGTGGTCATGAGT (SEQ ID N0:6) RPL28 Forward: TGCTGCTTGCTGTGAGTGTCT (SEQ ID N0:7) RPL38 Reverse: CGCGGACCAGGACCTTT (SEQ ID NO: 8)

RESULTS:

IncRNA-based classification of PPGL

The expression profiles of IncRNAs in 187 PPGL from 176 patients were obtained by repurposing 1586 probes corresponding to annotated and potential IncRNA transcripts present on the Affymetrix microarray previously used for transcriptome profiling [12, 40]. This series of tumors has been extensively characterized at the clinical, genetic and genomic levels [9]. The main characteristics of the analyzed cohort are summarized in Table 1.

A consensus clustering analysis using the 10% to 99% most variable probesets, segregated PPGL into two robust subtypes (lncl and lnc2) (data not shown). Of these, cluster lncl can be divided into three clinically relevant subtypes (lncl A, Inc IB, and Inc 1C). The IncRNA subtypes were strongly associated with mRNA expression clusters (chi² p-values from 1.38xl0 ³² to 1.07xl0 ⁶⁷) (data not shown). Specifically, lnclA was associated with mRNA expression cluster CIA characterized by the presence of SDHx mutations, whereas the Inc IB subtype was associated with gene expression cluster C1B associated with VEIL mutations. The cluster Inc 1C aggregates a few C2B and C2C samples. Regarding the robust IncRNA cluster lnc2, it was associated with gene expression cluster C2, albeit the IncRNA expression profiles did not separate gene expression clusters C2A, C2B, and C2C, characterized by NF1-, RET-, HRAS, TMEM127-, MET-, FGFR1, MAX- mutated, and sporadic tumors. Given the strong correlation between IncRNA and mRNA expression clusters and their corresponding clinical WO 2020/104482 PCT/EP2019/081848 features, we observed that lnclA subtype exhibits, like the CIA mRNA cluster, the worse prognosis (data not shown).

Total PPGL patients (n=176)

Male _ 68 (38.6%)

Female _ 108 (61.3%)

Mean age at diagnosis 42.34 (7-82)

(range)

Metastatic 23 (13%)

Median follow-up (years) 7.7

Total tumor samples (n=187)

Pheochromocytoma 153 (81.8%)

Paraganglioma 28 (14.9%)

Metastases 6 (3.2%)

Known driver^* _ 150 (80.2%)

Sporadic 37 (19.8%)

Transcriptomic cluster

Cl A (High-risk) tumors _ (n=24)

Metastatic cases_ 12/22 (54.5%)

C1 B tumors (n=42)

Metastatic cases 4/37 (10.8%)

C2A tumors _ _ (n=78) _

Metastatic cases_ 4/76 (5.26%)

C2B/C2C tumors _ (n=43) _

Metastatic cases _ 3/41 (7.31 %)

Table 1. Clinical and genetic characteristics of the analyzed cohort.

To validate this IncRNA classification, we analyzed a publicly available dataset of 51 PPGL (GEO, GSE67066). A predictor model constructed in the discovery set based on the most differentially expressed genes, revealed the existence of the four IncRNA subtypes in the validation cohort. Remarkably, despite the small number of samples, and the fact that the validation cohort is enriched in sporadic tumors (68%), the IncRNA classification of the two datasets showed similar tendencies (data not shown). WO 2020/104482 PCT/EP2019/081848

Differentially expressed IncRNAs and pathways between PPGL subtypes

We then performed a comparison of IncRNAs across the five mRNA expression clusters. We found that 1289/1586 (81%) of IncRNAs analyzed were differentially expressed (qval anova <0.05) and 16% of the significant IncRNAs had a high fold-change (>2-fold up or down) between the mRNA clusters (data not shown). Some of these transcripts are known IncRNAs and were annotated according to GENCODE vl5, whereas the majority correspond to putative IncRNA transcripts (data not shown). Notably, the percentages of dysregulated IncRNAs and protein-coding genes are similar for each tumor subtype (data not shown). For instance, subtype CIA displays a clear tendency towards the down-regulation of IncRNAs compared to the other molecular groups. This result is likely due to the distinctive hypermethylation phenotype driven by SDHx mutations in these tumors [10]. Indeed, we found that an absolute correlation of IncRNA expression and methylome data was significantly higher than the correlation of IncRNA expression with copy number data (data not shown).

Next, we aimed to estimate possible functional roles for differentially expressed IncRNAs by each tumor subtype. At present, there are no bioinformatics tools to do this directly but co-expression analysis with mRNAs is a relevant metric to infer the biological function of IncRNAs [46]. Therefore, we evaluated the signaling pathways of protein-coding genes that showed strong positive (Pearson >0.5) or negative (Pearson <-0.5) correlations with differentially expressed IncRNAs. Regarding positive correlations, up-regulated IncRNAs in .S'/l/Tr-mutated tumors (cluster CIA) were co-expressed with genes involved in cell cycle, whereas down-regulated IncRNAs correlated with genes of mitochondria pathway (data not shown). In addition, up-regulated IncRNAs in VT L-mutated tumors (cluster C1B), as well as in sporadic tumors (cluster C2C), correlated with genes involved in angiogenesis/hypoxia. Regarding negative correlations, down-regulated IncRNAs in cluster CIA were correlated with up-regulated genes involved in most of the pathways (data not shown), whereas the opposite was found in tumors from the cluster C2A. These results suggest that IncRNAs might synergistically participate with protein-coding genes in regulating signaling pathways that characterize PPGL subtypes.

Analysis of IncRNAs predictive of metastatic PPGL

Next, we investigated whether IncRNAs are associated with metastatic PPGL. We did not find differentially expressed IncRNAs between metastatic vs non-metastatic cases. We reasoned that this is probably due to the marked genetic heterogeneity of PPGL, and the fact that there were few metastatic cases within each molecular group. Thus, we decided to restrain this study to 24 samples of cluster CIA (21 primary tumors and 3 metastatic tissues) from 22 WO 2020/104482 PCT/EP2019/081848 patients. This cluster corresponds to transcriptionally homogenous SDHx- mutated tumors, and includes the highest proportion of patients with metastatic disease (12/22, 54.5%) (Table 1).

Half of these patients presented with metastases at diagnosis, and the remaining during the follow-up period (mean 37.78 months; range 0-112).

Using receiver operating characteristic curve (ROC) analyses, and applying stringent criteria (AUC>0.9 and fold-change >2.0), we identified one probeset (239921_at) targeting a putative IncRNA (GenBank, BC063866) that accurately discriminates metastatic from benign tumors (Fig 1A). Using RT-qPCR, we confirmed that the expression of this transcript was dramatically decreased (up to 10-fold) in metastatic tumors from the discovery and validation series of .SY/T/v-mutated carriers (Fig IB).

We then tested the impact of the expression levels of this transcript on the Metastasis Free Survival (MFS) of affected SDHx mutation carriers. For this analysis, we included known clinical factors of poor prognosis such as primary tumor size (>5 cm) and extra-adrenal location [7], as well as two recently suggested biomarkers, telomerase activation and ATRX mutation [47] Univariate cox models in SDHx mutation carriers show that increased expression of BC063866 was associated with lower risk of developing metastasis (H.R, 0.17; 95% C.I, 0.052- 0.56; P=0.0035), whereas telomerase (+) status and ATRX mutations were associated with higher risk of metastasis (H.R. from 2.4 to 6.1; p-values from 0.0043 to 0.014) (Fig 2A).

Remarkably, multivariate analysis on these remaining covariates revealed transcript BC063866 as an independent risk factor (hazard ratio, 0.106; P=0.01). (Fig 2B). The partition of the patients according to the expression levels of BC063866 (up- or down-regulated) separates patients with the opposite outcome (log-rank test P = 2.29x10 ⁰⁵) (Fig 20.

In silico analyses of transcript BC063866

ENCODE ChiP-seq data for RNA polymerase II and marks of active chromatin in normal adrenal tissue, indicate the presence of regulatory elements nearby the probeset 239921_at. In addition, RNA-seq signals suggest active though weak transcription of the putative IncRNA BC063866 located in the 3’UTR of its neighboring gene COL28A1 (data not shown). We noticed also that this noncoding transcript is preferentially expressed in the peripheral nervous system and becomes dysregulated in tumors such as ependymoma, ganglioneuroblastoma, and astrocytoma (data not shown). To explore a possible functional role for this putative IncRNA, we carried out co-expression analyses. Towards this aim, we first identified protein-coding genes that are differentially expressed between metastatic and benign tumors from cluster CIA (data not shown). We found that transcript BC063866 was highly correlated not only with its neighboring gene COL28A1 but also with genes at distant locations WO 2020/104482 PCT/EP2019/081848 that belong to the metastatic signature (data not shown). Strikingly, some of these genes are involved in neural crest and peripheral glial development including SOX10, ERBB3, CDH19, and PLP1. Moreover, RNA:DNA triplexes of transcript BC063866 with some of the correlated genes were also suggested by bioinformatics predictions using the LongTarget software (data not shown). Finally, we found that this transcript is predominantly localized in the nuclear fraction of NCI-H295 and 293T human cell lines (data not shown). Altogether, these observations indicate that BC063866 might participate in the molecular circuitry underlying metastasis in lT/v-mutated PPGL.

DISCUSSION

It is now widely accepted that the transcription of the non-coding genome generates functional RNAs [48]. Compelling evidence suggests that IncRNAs act as oncogenes or tumor suppressor genes by promoting wide expression changes at transcriptional and post- transcriptional levels [26, 27] Previous studies regarding the genomic characterization of PPGL have been devoted to the analysis of protein-coding genes and miRNAs specific to molecular groups, whereas only a few studies have addressed the role of IncRNAs.

Here, we found IncRNAs expression profiles that clearly distinguish PPGL subtypes. This result reinforces the concept that IncRNAs expression is highly tumor-type specific as previously shown for other types of human cancers [25-27, 49]. Moreover, IncRNA subtypes were highly correlated with mRNA expression clusters, indicating that both protein-coding and non-coding genes likely share the same regulatory elements [50]. Altered expression of IncRNAs in PPGL seems to be mostly explained by changes in DNA methylation rather than gain or losses of IncRNA gene loci, which is also consistent with Pan-Cancer data [26]. A recent study using RNA-seq TCGA data attempted to elucidate relationships between 23 IncRNAs, 22 mRNAs, and 6 miRNAs that were differentially expressed between normal and tumor samples [51]. Here we found fewer differentially expressed IncRNAs than protein coding genes in each molecular group. Although such co-expression analyses suggest that IncRNAs might play regulatory roles, functional studies are necessary to ascertain the specific mechanisms through which IncRNAs participate in signaling pathways that characterize PPGL subtypes.

Previous studies reported that maternally expressed IncRNAs MEG3 and HI 9 play roles in PPGL tumorigenesis of sporadic, VHL and SDHD subtypes via hypermethylation [52-56]. In line with these observations, we had reported the silencing of the imprinted DLK1-MEG3 locus in a subgroup of sporadic tumors of the present cohort [9]. Here, we found that HI 9 was less expressed in metastatic vs benign tumors with SDHx mutations by a loss of imprinting WO 2020/104482 PCT/EP2019/081848 mechanism, but it was not a good discriminator (AUC 0.73) (data not shown). It is important to note that the DLK1-MEG3 locus encompasses the expression of more than 50 miRNAs and that the miR-675 is embedded in the same transcription unit of H19 (first exon). Thus, such a genetic organization makes difficult to determine the specific roles of MEG 3 and HI 9 in PPGL tumorigenesis.

Regarding IncRNAs as prognostic biomarkers in PPGL, it was reported that expression of C9orfl47 and BSN-AS2 is associated with overall survival [51]. Here we identified a putative IncRNA (BC063866) that accurately distinguished metastatic from benign tumors with SDHx mutations and appeared as an independent risk factor of metastasis in this tumor subtype.

Although IncRNA studies that analyzed RNA-Seq datasets or transcriptome data (Affymetrix platforms) proved informative, it is worth noting that these approaches mainly capture protein-coding genes and are not optimal for IncRNAs profiling. It is estimated that in both cases, the coverage is over 10-16% of known human IncRNAs. Therefore, we expect that more IncRNAs than what is reported here, as well as other classes of ncRNAs linked to PPGL, will be uncovered by performing more comprehensive analyses.

We acknowledge also that a formal distinction between putative IncRNAs and transcripts of unknown coding potential [21] targeted by the probesets of the Affymetrix platform should be addressed. In fact, it remains to be determined whether transcript BC063866 corresponds to either a 3’UTR associated IncRNA or an alternative 3’-terminal exon of COL28A1 [57] Regardless of the nature of this transcript, it appeared as a good prognostic marker that may help to discriminate .SY/T/v-mutated tumors that progress towards metastasis from those that remain indolent. Nevertheless, this marker should be replicated in a large prospective cohort of patients with PPGL to definitely assess its real clinical value. Molecular tumor markers predictive of metastatic PPGL would be particularly helpful for the pediatric group who has a higher risk of metastatic disease [58] . A plethora of such markers were initially proposed based on a candidate approach but several technical, analytical and statistical limitations precluded reliable conclusions [59]. More recently, novel markers such as telomerase activation and ATRX mutations [47], overexpression of miR-21-3p/miR-183-5p [19], and RDBP hypermethylation [60] have emerged upon the arrival of unbiased genomic studies of large PPGL cohorts. The development of a molecular assay, based on the combined analysis of these promising biomarkers in tumoral tissues, will be a worthwhile objective for the next future. However, such clinical assay would have limitations in cases with synchronous metastasis [61], multiple tumors and/or intra-tumor heterogeneity [62, 63]. The assessment of prognostic markers in liquid biopsies will be the ultimate goal to circumvent these limitations. WO 2020/104482 PCT/EP2019/081848

In this regard, detection of circulating miRNAs [19] and of neuroendocrine gene transcripts [64] have already shown significant promise.

In conclusion, our findings extend the spectrum of transcriptional dysregulations in PPGL to IncRNAs and provide new insights into the roles of these transcripts in tumorigenesis and metastasis. Remarkably, we identified a putative IncRNA that could serve as a potential prognostic marker to predict the metastatic risk of patients carrying SDHB mutations

REFERENCES:

Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.

1. Group, N.G.S.i.P.S., et ah, Consensus Statement on next-generation-sequencing-based diagnostic testing of hereditary phaeochromocytomas and paragangliomas. Nat Rev Endocrinol, 2017. 13(4): p. 233-247.

2. Castro-Vega, L.J., et a , Rethinking pheochromocytomas and paragangliomas from a genomic perspective. Oncogene, 2016. 35(9): p. 1080-9.

3. Amar, L., et a , Genetic testing in pheochromocytoma or functional paraganglioma. J Clin Oncol, 2005. 23(34): p. 8812-8.

4. Amar, L., et a , Succinate dehydrogenase B gene mutations predict survival in

patients with malignant pheochromocytomas or paragangliomas. J Clin Endocrinol Metab, 2007. 92(10): p. 3822-8.

5. Castro-Vega, L.J., et a , Germline mutations in FH confer predisposition to malignant pheochromocytomas and paragangliomas. Hum Mol Genet, 2014. 23(9): p. 2440-6.

6. Buffet, A., et a , Germline Mutations in the Mitochondrial 2-Oxoglutarate/Malate Carrier SLC25A11 Gene Confer a Predisposition to Metastatic Paragangliomas. Cancer Res, 2018. 78(8): p. 1914-1922.

7. Nicolas, M. and P. Dahia, Predictors of outcome in phaeochromocytomas and

paragangliomas. FlOOORes, 2017. 6: p. 2160.

8. Lloyd RV, O.R., Kloppel G, Rosai J, WHO classification of tumours: pathology and genetics of tumours of endocrine organs. 4th ed. 2017: Lyon: IARC.

9. Castro-Vega, L.J., et a , Multi-omics analysis defines core genomic alterations in pheochromocytomas and paragangliomas. Nat Commun, 2015. 6: p. 6044.

10. Letouze, E., et a , SDH mutations establish a hypermethylator phenotype in

paraganglioma. Cancer Cell, 2013. 23(6): p. 739-52.

11. Loriot, C., et a , Epithelial to mesenchymal transition is activated in metastatic

pheochromocytomas and paragangliomas caused by SDHB gene mutations. J Clin Endocrinol Metab, 2012. 97(6): p. E954-62.

12. Bumichon, N., et a , Integrative genomic analysis reveals somatic mutations in

pheochromocytoma and paraganglioma. Hum Mol Genet, 2011. 20(20): p. 3974-85.

13. Toledo, R.A., et a , Recurrent Mutations of Chromatin-Remodeling Genes and Kinase Receptors in Pheochromocytomas and Paragangliomas. Clin Cancer Res, 2016.

22(9): p. 2301-10.

14. Fishbein, L., et a , Comprehensive Molecular Characterization of Pheochromocytoma and Paraganglioma. Cancer Cell, 2017. 31(2): p. 181-193. WO 2020/104482 PCT/EP2019/081848

15. Papathomas, T.G., et al., Telomerase reverse transcriptase promoter mutations in tumors originating from the adrenal gland and extra-adrenal paraganglia. Endocr Relat Cancer, 2014. 21(4): p. 653-61.

16. Svahn, F., et al., Telomerase reverse transcriptase promoter hypermethylation is

associated with metastatic disease in abdominal paraganglioma. Clin Endocrinol (Oxf), 2018. 88(2): p. 343-345.

17. Dwight, T., et al., TERT structural rearrangements in metastatic pheochromocytomas.

Endocr Relat Cancer, 2018. 25(1): p. 1-9.

18. Patterson, E., et al., The microRNA expression changes associated with malignancy and SDHB mutation in pheochromocytoma. Endocr Relat Cancer, 2012. 19(2): p. 157- 66.

19. Calsina B, C.V.L., Torres-Perez R, Inglada-Perez L, Curras-Freixes M, Roldan- Romero JM, Mancikova V, Leton R, Remacha L, Santos M, Bumichon N, Lussey- Lepoutre C, Rapizzi E, Grana O, Alvarez-Escola C, de Cubas AA, Lanillos J, Cordero-Barreal A, Martinez-Montes AM, Bellucci A, Amar L, Femandes-Rosa FL, Calatayud M, Aller J, Lamas C, Sastre-Marcos J, Canu L, Korpershoek E, Timmers HJ, Lenders J, Beuschlein F, Fassnacht-Capeller M, Eisenhofer G, and A.-S.F.

Mannelli M, Favier J, Rodriguez- Antona C, Cascon A, Montero-Conde C, Gimenez- Roqueplo AP, Robledo M, Integrative multi-omics analysis identifies a prognostic miRNA signature and a targetable miR-21 -3p/TSC2/mTOR axis in metastatic pheochromocytoma/paraganglioma. Theranostics, 2019. 9(17): p. 4946-4958.

20. Consortium, E.P., An integrated encyclopedia ofDNA elements in the human genome.

Nature, 2012. 489(7414): p. 57-74.

21. Iyer, M.K., et al., The landscape of long noncoding RNAs in the human transcriptome.

Nat Genet, 2015. 47(3): p. 199-208.

22. Huarte, M., The emerging role oflncRNAs in cancer. Nat Med, 2015. 21(11): p. 1253- 6E

23. Ulitsky, I. and D.P. Bartel, lincRNAs: genomics, evolution, and mechanisms. Cell, 2013. 154(1): p. 26-46.

24. Marchese, F.P., I. Raimondi, and M. Huarte, The multidimensional mechanisms of long noncoding RN A function. Genome Biol, 2017. 18(1): p. 206.

25. Ching, T., et al., Pan-Cancer Analyses Reveal Long Intergenic Non-Coding RNAs Relevant to Tumor Diagnosis, Subtyping and Prognosis. EBioMedicine, 2016. 7: p. 62-72.

26. Yan, X., et al., Comprehensive Genomic Characterization of Long Non-coding RNAs across Human Cancers. Cancer Cell, 2015. 28(4): p. 529-540.

27. Chiu, H.S., et al., Pan-Cancer Analysis oflncRNA Regulation Supports Their

Targeting of Cancer Genes in Each Tumor Context. Cell Rep, 2018. 23(1): p. 297-312 el2.

28. Saxena, N., et al., SDHB-Deficient Cancers: The Role of Mutations That Impair Iron Sulfur Cluster Delivery. J Natl Cancer Inst, 2016. 108(1).

29. Curras-Freixes, M., et al., Recommendations for somatic and germline genetic testing of single pheochromocytoma and paraganglioma based on findings from a series of 329 patients. J Med Genet, 2015. 52(10): p. 647-56.

30. Jochmanova, L, et al., SDHB-related pheochromocytoma and paraganglioma

penetrance and genotype -phenotype correlations. J Cancer Res Clin Oncol, 2017. 143(8): p. 1421-1435.

31. Papathomas, T.G., et al., SDHB/SDHA immunohistochemistry in pheochromocytomas and paragangliomas: a multicenter interobserver variation analysis using virtual WO 2020/104482 PCT/EP2019/081848 microscopy: a Multinational Study of the European Network for the Study of Adrenal Tumors (ENS@T). Mod Pathol, 2015. 28(6): p. 807-21.

32. Lenders, J.W., et al., Pheochromocytoma and paraganglioma: an endocrine society clinical practice guideline. J Clin Endocrinol Metab, 2014. 99(6): p. 1915-42.

33. Buffet, A., et al., Positive Impact of Genetic Test on the Management and Outcome of Patients With Paraganglioma and/or Pheochromocytoma. J Clin Endocrinol Metab, 2019. 104(4): p. 1109-1118.

34. Plouin, P.F., et al., European Society of Endocrinology Clinical Practice Guideline for long-term follow-up of patients operated on for a phaeochromocytoma or a

paraganglioma. Eur J Endocrinol, 2016. 174(5): p. G1-G10.

35. Timmers, H.J., et al., Superiority of fluorodeoxy glucose positron emission tomography to other functional imaging techniques in the evaluation of metastatic SDHB- associated pheochromocytoma and paraganglioma. J Clin Oncol, 2007. 25(16): p. 2262-9.

36. Baudin, E., et al., Therapy of endocrine disease: treatment of malignant

pheochromocytoma and paraganglioma. Eur J Endocrinol, 2014. 171(3): p. R111-22.

37. Gimenez-Roqueplo, A.P., et al., Mutations in the SDHB gene are associated with

extra-adrenal and/or malignant phaeochromocytomas. Cancer Res, 2003. 63(17): p. 5615-21.

38. Plouin, P.F., A.P. Gimenez-Roqueplo, and X. Bertagna, [COMETE, a network for the study and management of hypersecreting adrenal tumors]. Bull Acad Natl Med, 2008. 192(1): p. 73-82; discussion 83-5.

39. Ben Aim, L., et al., Targeted next- generation sequencing detects rare genetic events in pheochromocytoma and paraganglioma. J Med Genet, 2019.

40. Yarmishyn, A. A., et al., HOXD-AS1 is a novel IncRNA encoded in HOXD cluster and a marker of neuroblastoma progression revealed via integrative analysis of noncoding transcriptome. BMC Genomics, 2014. 15 Suppl 9: p. S7.

41. Derrien, T., et al., The GENCODE v7 catalog of human long noncoding RNAs:

analysis of their gene structure, evolution, and expression. Genome Res, 2012. 22(9): p. 1775-89.

42. Wilkerson, M.D. and D.N. Hayes, ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics, 2010. 26(12): p. 1572-3.

43. He, S., et al., LongTarget: a tool to predict IncRNA DNA-binding motifs and binding sites via Hoogsteen base-pairing analysis. Bioinformatics, 2015. 31(2): p. 178-86.

44. Hruz, T., et al., Genevestigator v3: a reference expression database for the meta

analysis of transcriptomes. Adv Bioinformatics, 2008. 2008: p. 420747.

45. Biton, A., et al., Independent component analysis uncovers the landscape of the

bladder tumor transcriptome and reveals insights into luminal and basal subtypes.

Cell Rep, 2014. 9(4): p. 1235-45.

46. Huarte, M., et al., A large intergenic noncoding RNA induced by p53 mediates global gene repression in the p53 response. Cell, 2010. 142(3): p. 409-19.

47. Job, S., et al., Telomerase Activation andATRX Mutations Are Independent Risk

Factors for Metastatic Pheochromocytoma and Paraganglioma. Clin Cancer Res, 2019. 25(2): p. 760-770.

48. Djebali, S., et al., Landscape of transcription in human cells. Nature, 2012. 489(7414): p. 101-8.

49. Reon, B.J., et al., Expression oflncRNAs in Low-Grade Gliomas and Glioblastoma Multiforme: An In Silico Analysis. PLoS Med, 2016. 13(12): p. el002192. WO 2020/104482 PCT/EP2019/081848

50. Cabili, M.N., et al., Integrative annotation of human large intergenic noncoding RNAs reveals global properties and specific subclasses. Genes Dev, 2011. 25(18): p. 1915- 27.

51. Liang, Y.C., et al., Building a Competing Endogenous RNA Network to Find Potential Long Non-Coding RNA Biomarkers for Pheochromocytoma. Cell Physiol Biochem, 2018. 51(6): p. 2916-2924.

52. Astuti, D., et al., Epigenetic alteration at the DLK1-GTL2 imprinted domain in human neoplasia: analysis of neuroblastoma, phaeochromocytoma and Wilms' tumour. Br J Cancer, 2005. 92(8): p. 1574-80.

53. Margetts, C.D., et al., Epigenetic analysis ofHICl, CASP8, FLIP, TSPI, DCR1,

DCR2, DR4, DR5, KvDMRl, H19 and preferential llpl5.5 maternal-allele loss in von Hippel-Lindau and sporadic phaeochromocytomas. Endocr Relat Cancer, 2005. 12(1): p. 161-72.

54. Yeap, P.M., et al., Molecular analysis of pheochromocytoma after maternal

transmission of SDHD mutation elucidates mechanism of parent-of-origin effect. J Clin Endocrinol Metab, 2011. 96(12): p. E2009-13.

55. Nielsen, H.M., et al., Copy number variations alter methylation and parallel IGF2 overexpression in adrenal tumors. Endocr Relat Cancer, 2015. 22(6): p. 953-67.

56. Hoekstra, A.S., et al., Loss of maternal chromosome 11 is a signature event in

SDHAF2, SDHD, and VHL-related paragangliomas, but less significant in SDHB- related paragangliomas. Oncotarget, 2017. 8(9): p. 14525-14536.

57. Dutertre, M., et al., A recently evolved class of alternative 3'-terminal exons involved in cell cycle regulation by topoisomerase inhibitors. Nat Commun, 2014. 5: p. 3395.

58. Pamporaki, C., et al., Characteristics of Pediatric vs Adult Pheochromocytomas and Paragangliomas. J Clin Endocrinol Metab, 2017. 102(4): p. 1122-1132.

59. Goncalves, J., et al., Emerging molecular markers of metastatic pheochromocytomas and paragangliomas. Ann Endocrinol (Paris), 2019. 80(3): p. 159-162.

60. de Cubas, A.A., et al., DNA Methylation Profiling in Pheochromocytoma and

Paraganglioma Reveals Diagnostic and Prognostic Markers. Clin Cancer Res, 2015. 21(13): p. 3020-30.

61. Hamidi, O., et al., Outcomes of patients with metastatic phaeochromocytoma and paraganglioma: A systematic review and meta-analysis. Clin Endocrinol (Oxf), 2017. 87(5): p. 440-450.

62. Crona, J., et al., Spatiotemporal Heterogeneity Characterizes the Genetic Landscape of Pheochromocytoma and Defines Early Events in Tumorigenesis. Clin Cancer Res, 2015. 21(19): p. 4451-60.

63. Backman, S., et al., Global DNA Methylation Analysis Identifies Two Discrete clusters of Pheochromocytoma with Distinct Genomic and Genetic Alterations. Sci Rep, 2017. 7: p. 44943.

64. Peczkowska, M., et al., The clinical utility of circulating neuroendocrine gene

transcript analysis in well-differentiated paragangliomas and pheochromocytomas. Eur J Endocrinol, 2017. 176(2): p. 143-157.

Claims

WO 2020/104482 PCT/EP2019/081848 CLAIMS:

1. A method of identifying the metastatic potential of a tumor in a patient carrying at least one SDHx-mutation comprising i) determining the expression level of the putative long non-coding RNA BC063866 in a tumor sample obtained from the patient, ii) comparing the expression level determined at step i) with a predetermined reference value and wherein detecting differential between the expression level determined at step i) and the predetermined reference value indicates whether the patient is at risk of developing a metastatic tumour.

2. Use of the method of claim 1 for predicting patient’s survival, in particular, metastasis- free survival.