METHOD TO PERFORM DIAGNOSIS AND PROGNOSIS OF MELANOMA AND KIT

Abstract

The present invention refers to a method to perform a diagnosis of melanoma in a subject, as well as a method to perform a prognosis of a subject suffering from melanoma. Moreover, the present invention refers to a kit which allows to perform a non-invasive diagnosis of melanoma in a subject, and/or to perform a prognosis of a subject suffering from melanoma.

Claims

1. A method to perform a diagnosis of melanoma in a subject comprising the following steps: a) determining the expression levels of at least one gene, and/or of at least one protein encoded by at least one gene, from one or more samples taken and isolated from said subject; and b) comparing said expression levels determined in step a), with the standard or control expression levels of the same genes and/or the same proteins characterized in that said at least one gene is selected from the group of genes: ATG9A, BAG1, CAPN2, CHMP2B, GNAI3, ITGB4, KIAA0226, MLST8, PEX3, and WIPI1, and/or said at least one protein is a protein encoded by the genes selected from the group of genes: BAG1, PEX3, and WIPI1.

2. The method according to claim 1, characterized in that said at least one gene of which said expression levels are determined is selected from the group of genes: BAG1, CHMP2B, PEX3, and WIPI1.

3. The method to perform a prognosis for a subject suffering from melanoma comprising the following steps: a′) determining the expression levels of at least one gene, and/or of at least one protein encoded by a gene, from at least one sample taken and isolated from said subject; and b′) comparing said expression levels determined in step a′), with the standard or control expression levels of the same genes and/or the same proteins characterized in that said at least one gene is selected from the BAG1 and WIPI1 genes, and/or said at least one protein is a protein encoded by the genes selected from BAG1 and WIPI1.

4. The method according to any one of the preceding claims, characterized in that said at least one gene of which said expression levels are determined is WIPI1, and/or said protein of which said expression levels are determined is the protein encoded by the WIPI1 gene.

5. The method according to any one of the preceding claims, characterized in that said at least one isolated sample is a sample of isolated epidermis.

6. The method according to any one of the preceding claims, characterized in that said at least one isolated sample is a sample of isolated nevus.

7. The method according to claims 1 to 4, characterized in that said at least one isolated sample is a sample of isolated blood.

8. A kit comprising one or more agents suitable to measure the expression levels of at least one gene and/or at least one protein, characterized in that said at least one gene is selected from the group of genes: ATG9A, BAG1, CAPN2, CHMP2B, GNAI3, ITGB4, KIAA0226, MLST8, PEX3, and WIPI1, and/or said at least one protein is a protein encoded by the genes selected from the group of genes: BAG1, PEX3, and WIPI1.

9. The kit according to the preceding claim, characterized in that said at least one gene is selected from the group of genes: BAG1, CHMP2B, PEX3, and WIPI1.

10. The kit according to claim 9, characterized in that said at least one gene is selected from the BAG1 and WIPI1 genes, and/or said at least one protein is a protein encoded by the genes selected from BAG1 and WIPI1.

Description

DESCRIPTION OF THE FIGURES

[0052] FIG. 1 shows some plots where the area under the curve (AUC) of each gene differentially expressed in the melanoma with respect to controls, is reported as sensitivity percentage (Sensitivity %) vs specificity percentage (100-specificity %); the AUC calculated for each gene is reported with a p value <0.0001. The data refer to the expression measurements reported in the GEO public database (https://www.ncbi.nlm.nih.gov/gds)

[0053] FIG. 2 reports the gene expression according to the IST Online public database (http://ist.medisapiens.com/). In particular, this figure reports the differential expression levels in melanoma vs healthy skin of 208 melanoma biopsies and of 147 healthy skin biopsies. The dotted lines depict the discriminant level of 90% of the melanoma samples from the remaining 10% of the melanoma samples. The expression of PEX3, BAG1 and CHMP2B in the melanomas is lower with respect to the expression measured in the normal skin in at least 90% of the melanoma cases, while the expression of WIPI1 in the melanomas is higher than the expression measured in the healthy controls in at least 90% of the melanoma cases.

[0054] FIG. 3 is the expression of the proteins encoded by the BAG1, PEX3 and WIPI1 genes in the Human Protein Atlas public database (https://www.proteinatlas.org).

[0055] The histological sections reported in the database have been obtained and converted in grayscale; the pixels have been then quantified as a function of their gray intensity according to an intensity scale ranging from the lightest gray (value of 255, corresponding to the lowest expression) to the darkest gray (0 value, corresponding to the greatest expression). The Figure shows that the distribution medians are shifted to the right in the BAG1, PEX3 and WIPI1 cases, suggesting for these proteins an expression in the melanoma greater than the controls.

Experimental Section

[0056] In the present experimental section will be reported, in the order as they have been performed, the tests and the analyses which allowed to identify the genes and the proteins differently expressed in a significant way in the melanomas with respect to the corresponding expression in the tissues of healthy subjects.

Example 1

Comparative Analysis of the Expression of Genes Correlated to the Autophagy (ARGs) in Melanoma Vs Normal Skin Samples

[0057] To identify potential genes and proteins involved in the development of melanoma and in its progression, 222 genes correlated to autophagy (ARGs) have been selected from the Human Autophagy Database (HADb) (http://www.autophagy.lu) to be studied.

[0058] The difference between the expression level of melanoma samples and of healthy epidermis samples of these 222 ARGs has been evaluated starting from 3 datasets (Talantov, Riker e Haqq) on the Oncomine database (www.oncomine.org) on the basis of 110 samples of melanoma and 15 samples of normal skin (125 total samples). Then the number of times where the mean value of the expression level of each of the 222 genes in the 110 melanoma samples is higher or lower than the mean value of the expression level of the same gene in the 15 samples of normal skin (fold change difference) has been calculated. Such calculation has been performed as follows:

FC difference=2.sup.d

wherein d=X.sub.2−X.sub.1; X.sub.2 is the mean of the values relative to the expression level of the gene of interest in the samples of condition 1 (melanoma); and X.sub.1 is the mean of the values relative to the expression level of the gene of interest in the samples of condition 2 (normal skin). In the present case, X.sub.2 is the mean of the values relative to the expression level of the gene of interest in the 110 samples of melanoma (condition 1) and X.sub.1 is the mean of the values relative to expression level of the gene of interest in the 15 control samples (condition 2).

[0059] The fold change difference highlighted that, in melanoma, 70 ARGs out of 222 show significant up-regulation or down-regulation, i.e. greater than +1.5 or lower than −1.5, with respect to the healthy skin. In Table 1 (depicted below) there is the list of the 222 analyzed ARGs and the fold change difference values; the 70 genes showing significant up-regulation or down-regulation are indicated by an x.

TABLE-US-00001 TABLE 1 Name of Symbol of Fold change # the gene the gene difference  1 autophagy/beclin-1 regulator 1 AMBRA1 1.41  2 apolipoprotein L, 1 APOL1 −0.11  3 aryl hydrocarbon receptor nuclear ARNT 0.70 translocator  4 arylsulfatase A ARSA 0.67  5 arylsulfatase B ARSB 0.11  6 activating transcription factor 4 ATF4 0.50  7 activating transcription factor 6 ATF6 0.25  8 ATG10 autophagy related 10 ATG10 −0.11 homolog (S. cerevisiae)  9 ATG12 autophagy related 12 ATG12 1.11 homolog (S. cerevisiae)  10 ATG16 autophagy related ATG16L1 0.91 16-like 1 (S. cerevisiae)  11x ATG16 autophagy related ATG16L2 −2.21 16-like 2 (S. cerevisiae)  12 ATG2 autophagy related 2 ATG2A −1.19 homolog A (S. cerevisiae)  13 ATG2 autophagy related 2 ATG2B −0.27 homolog B (S. cerevisiae)  14 ATG3 autophagy related 3 ATG3 0.58 homolog (S. cerevisiae)  15 ATG4 autophagy related 4 ATG4A homolog A (S. cerevisiae)  16 ATG4 autophagy related 4 ATG4B 1.44 homolog B (S. cerevisiae)  17 ATG4 autophagy related 4 ATG4C −0.08 homolog C (S. cerevisiae)  18 ATG4 autophagy related 4 ATG4D 1.03 homolog D (S. cerevisiae)  19 ATG5 autophagy related 5 ATG5 0.95 homolog (S. cerevisiae)  20 ATG7 autophagy related 7 ATG7 0.60 homolog (S. cerevisiae)  21 ATG9 autophagy related 9 ATG9A 0.61 homolog A (S. cerevisiae)  22 ATG9 autophagy related 9 ATG9B −0.63 homolog B (S. cerevisiae)  23 5-aminoimidazole-4- ATIC 1.08 carboxamide ribonucleotide formyltransferase/IMP cyclohydrolase  24 BCL2-associated athanogene BAG1 −1.43  25 BCL2-associated athanogene 3 BAG3 0.04  26 BCL2-antagonist/killer 1 BAK1 0.89  27x BCL2-associated X protein BAX 8.16  28 B-cell CLL/lymphoma 2 BCL2 0.71  29x BCL2-like 1 BCL2L1 1.86  30x beclin 1, autophagy related BECN1 1.58  31 BH3 interacting domain death BID 0.80 agonist  32x baculoviral IAP repeat-contain- BIRC5 3.57 ing 5  33 baculoviral IAP repeat-contain- BIRC6 −0.39 ing 6  34 BCL2/adenovirus E1B 19 kDa BNIP1 1.24 interacting protein 1  35 BCL2/adenovirus E1B 19 kDa BNIP3 −0.11 interacting protein 3  36 BCL2/adenovirus E1B 19 kDa BNIP3L 0.00 interacting protein 3-like  37 chromosome 12 open reading C12orf44 0.67 frame 44  38 chromosome 17 open reading C17orf88 1.01 frame 88  39 calcium binding and CALCOCO2 −0.35 coiled-coil domain 2  40 calcium/calmodulin- CAMKK2 1.49 dependent protein kinase kinase 2, beta  41x calnexin CANX 1.97  42 calpain 1, (mu/I) large CAPN1 −0.82 subunit  43 calpain 10 CAPN10 0.49  44x calpain 2, (m/II) large CAPN2 1.50 subunit  45x calpain, small subunit 1 CAPNS1 3.29  46 caspase 1, apoptosis- CASP1 0.84 related cysteine peptidase (interleukin 1, beta, convertase)  47 caspase 3, apoptosis-related CASP3 0.76 cysteine peptidase  48 caspase 4, apoptosis-related CASP4 0.58 cysteine peptidase  49 caspase 8, apoptosis-related CASP8 0.47 cysteine peptidase  50 chemokine (C-C motif) CCL2 −0.57 ligand 2  51 chemokine (C-C motif) CCR2 receptor 2  52 CD46 molecule, complement CD46 1.39 regulatory protein  53 cyclin-dependent kinase CDKN1A 1.15 inhibitor 1A (p21, Cip1)  54 cyclin-dependent kinase CDKN1B −0.64 inhibitor 1B (p27, Kip1)  55x cyclin-dependent kinase CDKN2A 2.36 inhibitor 2A (melanoma, p16, inhibits CDK4)  56 CASP8 and FADD-like apoptosis CFLAR 1.17 regulator  57 chromatin modifying protein 2B CHMP2B −0.17  58 chromatin modifying protein 4B CHMP4B −1.12  59x ceroid-lipofuscinosis, CLN3 1.51 neuronal 3  60x cathepsin B CTSB 7.17  61 cathepsin D CTSD 0.86  62 cathepsin L1 CTSL1 0.89  63x chemokine (C-X3-C motif) CX3CL1 −1.96 ligand 1  64x chemokine (C-X-C motif) CXCR4 2.79 receptor 4  65 death-associated protein DAPK1 1.15 kinase 1  66x death-associated protein DAPK2 −1.56 kinase 2  67 DNA-damage-inducible DDIT3 0.23 transcript 3  68x DIRAS family, GTP-binding DIRAS3 −1.84 RAS-like 3  69x deleted in liver cancer 1 DLC1 1.68  70 DnaJ (Hsp40) homolog, DNAJB1 0.82 subfamily B, member 1  71 DnaJ (Hsp40) homolog, DNAJB9 1.47 subfamily B, member 9  72 DNA-damage regulated DRAM1 0.11 autophagy modulator 1  73 ER degradation enhancer, EDEM1 0.42 mannosidase alpha-like 1  74 eukaryotic translation EEF2 −0.29 elongation factor 2  75x eukaryotic elongation EEF2K −3.22 factor-2 kinase  76 epidermal growth factor EGFR −1.47 receptor  77 eukaryotic translation EIF2AK2 1.44 initiation factor 2-alpha kinase 2  78 eukaryotic translation EIF2AK3 1.10 initiation factor 2-alpha kinase 3  79 eukaryotic translation EIF2S1 0.19 initiation factor 2, subunit 1 alpha, 35 kDa  80x eukaryotic translation EIF4EBP1 2.78 initiation factor 4E binding protein 1  81x eukaryotic translation EIF4G1 3.67 initiation factor 4 gamma, 1  82 v-erb-b2 erythroblastic ERBB2 −0.16 leukemia viral oncogene homolog 2, neuro/glioblas- toma derived oncogene homolog (avian)  83 endoplasmic reticulum to ERN1 1.21 nucleus signaling 1  84x ERO1-like (S. cerevisiae) ERO1L 1.74  85 Fas (TNFRSF6)-associated FADD 1.47 via death domain  86 family with sequence FAM48A 0.35 similarity 48, member A  87 Fas (TNF receptor super- FAS 1.35 family, member 6)  88x FK506 binding protein 1A, FKBP1A 1.82 12 kDa  89x FK506 binding protein 1B, FKBP1B 1.76 12.6 kDa  90x FBJ murine osteosarcoma FOS −5.84 viral oncogene homolog  91 forkhead box O1 FOXO1 −0.52  92 forkhead box O3 FOXO3 −0.67  93x glucosidase, alpha; acid GAA 2.02  94 GABA(A) receptor-associated GABARAP 0.91 protein  95 GABA(A) receptor-associated GABARAPL1 0.57 protein like 1  96 GABA(A) receptor-associated GABARAPL2 0.54 protein-like 2  97x glyceraldehyde-3-phosphate GAPDH 3.17 dehydrogenase  98 guanine nucleotide binding GNAI3 −0.26 protein (G protein), alpha inhibiting activity polypeptide 3  99 guanine nucleotide binding GNB2L1 1.48 protein (G protein), beta polypeptide 2-like 1 100 golgi-associated PDZ and GOPC 1.41 coiled-coil motif containing 101x glutamate receptor, ionotropic, GRID1 1.93 delta 1 102 glutamate receptor, ionotropic, GRID2 0.11 delta 2 103 histone deacetylase 1 HDAC1 1.43 104 histone deacetylase 6 HDAC6 1.44 105 hepatocyte growth factor- HGS 1.34 regulated tyrosine kinase substrate 106 hypoxia inducible factor 1, HIF1A 0.15 alpha subunit (basic helix- loop-helix transcription factor) 107x heat shock protein 90 kDa alpha HSP90AB1 2.64 (cytosolic), class B member 1 108x heat shock 70 kDa protein 5 HSPA5 2.08 (glucose-regulated protein, 78 kDa) 109x heat shock 70 kDa protein 8 HSPA8 4.53 110x heat shock 22 kDa protein 8 HSPB8 −3.99 111 interferon, gamma IFNG not found 112x inhibitor of kappa light IKBKB 2.00 polypeptide gene enhancer in B-cells, kinase beta 113 inhibitor of kappa light IKBKE 1.24 polypeptide gene enhancer in B-cells, kinase epsilon 114x interleukin 24 IL24 2.08 115 immunity-related GTPase IRGM −1.41 family, M 116x integrin, alpha 3 (antigen ITGA3 5.18 CD49C, alpha 3 subunit of VLA-3 receptor) 117x integrin, alpha 6 ITGA6 4.31 118x integrin, beta 1 (fibronectin ITGB1 2.63 receptor, beta polypeptide, antigen CD29 includes MDF2, MSK12) 119x integrin, beta 4 ITGB4 1.84 120 inositol 1,4,5-triphosphate ITPR1 0.06 receptor, type 1 121 KIAA0226 KIAA0226 0.07 122 KIAA0652 KIAA0652 1.30 123 KIAA0831 KIAA0831 −1.13 124 kinesin family member 5B KIF5B −0.51 125x kelch-like 24 (Drosophila) KLHL24 1.95 126x lysosomal-associated membrane LAMP1 2.28 protein 1 127 lysosomal-associated membrane LAMP2 0.63 protein 2 128 microtubule-associated protein 1 MAP1LC3A −1.10 light chain 3 alpha 128x microtubule-associated protein 1 MAP1LC3B 1.94 light chain 3 beta 130 microtubule-associated protein 1 MAP1LC3C −1.22 light chain 3 gamma 131 mitogen-activated protein kinase MAP2K7 0.82 kinase 7 132x mitogen-activated protein kinase 1 MAPK1 1.52 133 mitogen-activated protein kinase 3 MAPK3 −0.08 134 mitogen-activated protein kinase 8 MAPK8 0.88 135 mitogen-activated protein kinase 8 MAPK8IP1 0.96 interacting protein 1 136 mitogen-activated protein kinase 9 MAPK9 1.41 137x membrane-bound transcription MBTPS2 1.98 factor peptidase, site 2 138x MTOR associated protein, LST8 MLST8 5.05 homolog (S. cerevisiae) 139 myotubularin related protein 14 MTMR14 0.12 140 mechanistic target of rapamycin MTOR −0.03 (serine/threonine kinase) 141x v-myc myelocytomatosis viral MYC 2.60 oncogene homolog (avian) 142x nuclear assembly factor 1 homolog NAF1 −1.56 (S. cerevisiae) 143 nicotinamide phosphoribosyl- NAMPT 1.11 transferase 144 neighbor of BRCA1 gene 1 NBR1 1.23 145 NCK-associated protein 1 NCKAP1 −0.38 146 nuclear factor (erythroid-derived NFE2L2 0.55 2)-like 2 147 nuclear factor of k light polypep- NFKB1 −0.54 tide gene enhancer in B-cells 1 148 NK2 transcription factor related, NKX2-3 0.26 locus 3 (Drosophila) 149x NLR family, CARD domain NLRC4 1.84 containing 4 150x Niemann-Pick disease, type C1 NPC1 1.93 151 neuregulin 1 NRG1 0.60 152 neuregulin 2 NRG2 0.31 153 neuregulin 3 NRG3 −1.14 154x prolyl 4-hydroxylase, beta P4HB 3.67 polypeptide 155 Parkinson disease (autosomal PARK2 1.28 recessive, juvenile) 2, parkin 156x poly (ADP-ribose) polymerase 1 PARP1 2.61 157x phosphoprotein enriched in astro- PEA15 2.26 cytes 15 158 proline, glutamate and leucine PELP1 −1.18 rich protein 1 159 peroxisomal biogenesis factor 14 PEX14 1.36 160 peroxisomal biogenesis factor 3 PEX3 0.20 161 phosphoinositide-3-kinase, class 3 PIK3C3 0.63 162 phosphoinositide-3-kinase, PIK3R4 0.82 regulatory subunit 4 163 PTEN induced putative kinase 1 PINK1 −0.77 164 protein phosphatase 1, regulatory PPP1R15A 1.09 (inhibitor) subunit 15A 165 protein kinase, AMP-activated, PRKAB1 0.42 beta 1 non-catalytic subunit 166x protein kinase, cAMP-dependent, PRKAR1A 2.74 regulatory, type I, alpha 167x protein kinase C, delta PRKCD 3.81 168 protein kinase C, theta PRKCQ 1.23 169x phosphatase and tensin homolog PTEN 2.01 170x PTK6 protein tyrosine kinase 6 PTK6 −7.21 171 RAB11A, member RAS oncogene RAB11A −0.55 family 172x RAB1A, member RAS oncogene RAB1A 2.31 family 173 RAB24, member RAS oncogene RAB24 −0.20 family 174x RAB33B, member RAS oncogene RAB33B −1.83 family 175x RAB5A, member RAS oncogene RAB5A 1.60 family 176x RAB7A, member RAS oncogene RAB7A 2.06 family 177 ras-related C3 botulinum toxin RAC1 1.37 substrate 1 (rho family, small GTP binding protein Rac1) 178 v-raf-1 murine leukemia viral RAF1 −0.41 oncogene homolog 1 179 retinoblastoma 1 RB1 1.32 180 RB1-inducible coiled-coil 1 RB1CC1 −0.37 181 v-rel reticuloendotheliosis viral RELA 0.60 oncogene homolog A (avian) 182 regulator of G-protein signaling RGS19 1.45 19 183x Ras homolog enriched in brain RHEB 2.28 184 ribosomal protein S6 kinase, RPS6KB1 0.29 70 kDa, polypeptide 1 185x regulatory associated protein of RPTOR −1.66 MTOR, complex 1 186 SAR1 homolog A (S. cerevisiae) SAR1A 1.35 187 serpin peptidase inhibitor, SERPINA1 −0.86 clade A (alpha-1 antiproteinase, antitrypsin), member 1 188 sestrin 2 SESN2 1.17 189 SH3-domain GRB2-like endophilin SH3GLB1 0.47 B1 190x sirtuin (silent mating type SIRT1 −1.63 information regulation 2 homolog) 1 (S. cerevisiae) 191 sirtuin (silent mating type SIRT2 0.52 information regulation 2 homolog) 2 (S. cerevisiae) 192x sphingosine kinase 1 SPHK1 1.51 193 spinster homolog 1 (Drosophila) SPNS1 −1.37 194x sequestosome 1 SQSTM1 6.26 195 suppression of tumorigenicity 13 ST13 0.83 (colon carcinoma) 196 serine/threonine kinase 11 STK11 −0.04 197 TANK-binding kinase 1 TBK1 1.02 198 transmembrane 9 superfamily TM9SF1 1.22 member 1 199x transmembrane protein 49 TMEM49 2.89 200x transmembrane protein 74 TMEM74 2.10 201 tumor necrosis factor (ligand) TNFSF10 −1.40 superfamily, member 10 202 tumor protein p53 TP53 −0.17 203 tumor protein p53 inducible TP53INP2 −0.09 nuclear protein 2 204x tumor protein p63 TP63 −1.64 205x tumor protein p73 TP73 −1.66 206 tuberous sclerosis 1 TSC1 −1.30 207x tuberous sclerosis 2 TSC2 −1.80 208 tumor suppressor candidate 1 TUSC1 −1.34 209 unc-51-like kinase 1 (C. elegans) ULK1 −1.14 210x unc-51-like kinase 2 (C. elegans) ULK2 1.51 211 unc-51-like kinase 3 (C. elegans) ULK3 −1.41 212 ubiquitin specific peptidase 10 USP10 1.01 213 UV radiation resistance associated UVRAG −0.32 gene 214 vesicle-associated membrane VAMP3 1.04 protein 3 (cellubrevin) 215 vesicle-associated membrane VAMP7 0.51 protein 7 216 vascular endothelial growth VEGFA 0.35 factor A 217 WD repeat and FYVE domain WDFY3 −0.05 containing 3 218 WD repeat domain 45 WDR45 1.08 219 WDR45-like WDR45L −1.24 220x WD repeat domain, phosphoino- WIPI1 5.28 sitide interacting 1 221x WD repeat domain, phosphoino- WIPI2 2.01 sitide interacting 2 222 zinc finger, FYVE domain ZFYVE1 1.00 containing 1

[0060] From the analysis of the present example it has been concluded that about 30% of the analyzed ARGs (70 out of 222) are differentially expressed in the melanoma samples significantly with respect to normal skin samples. It is interesting to note that 10 genes (BAX, CTSB, FOS, HSPA8, ITGA3, ITGA6, MLST8, PTK6, SQSTM1 and WIPI1) show 4-fold greater up- or down-regulation in melanoma samples with respect to the normal skin samples (i.e. fold change difference greater than +4 or lower than −4).

Example 2

Comparative Analysis of the Expression of ARGs in Melanoma Vs Nevi Samples

[0061] The expression levels of the 222 ARGs evaluated in 45 samples of melanoma and 18 samples of nevi have been taken from the dataset GDS1375 in the GEO database (https://www.ncbi.nlm.nih.gov/gds). Once such expression levels have been obtained, the comparison between the expression levels of the ARGs in melanoma samples and nevi samples has been carried out by means of mean calculation, t test and ROC analysis. The ROC analysis is the most recognized method in binary tests and calculates the area under the curve (AUC) which denotes how effective is the expression of a given gene to discriminate the healthy biopsies (in this case, nevi biopsies) from the melanoma biopsies.

[0062] In Table 2 the 42 genes out of the analyzed 222 are reported which resulted to be differentially expressed in a significant relevant way, i.e. with an AUC greater than or equal to 0.85 and p<0.0001.

TABLE-US-00002 TABLE 2 Mean Mean Correlation Symbol of expression in expression in P value between the gene the gene the melanoma the nevus AUC of AUC and melanoma  1 ATF4 5611 7598 0.87 <0.0001 Yes *  2 ATG4B 1016 739 0.88 <0.0001 Yes **  3x ATG9A 1014 735 0.89 <0.0001 No  4x BAG1 727 1891 1 <0.0001 No  5 BAG3 1098 1831 0.87 <0.0001 Yes **  6 BAX 498 186 0.93 <0.0001 Yes *  7 BCL2 1409 180 0.99 <0.0001 Yes **  8 BCL2L1 1244 375 0.99 <0.0001 Yes **  9 BIRC5 590 230 0.92 <0.0001 Yes * 10x CAPN2 9236 5096 0.96 <0.0001 No 11 CAPNS1 8826 4392 0.93 <0.0001 Yes ** 12 CDKN1A 2790 1397 0.90 <0.0001 Yes ** 13 CDKN2A 650 335 0.86 <0.0001 Yes * 14 CFLAR 472 796 0.85 <0.0001 Yes ** 15x CHMP2B 282 561 0.87 <0.0001 No 16 CTSB 16713 1655 0.99 <0.0001 Yes * 17 CTSD 2115 1029 0.89 <0.0001 Yes * 18 CX3CL1 266 627 0.91 <0.0001 Yes * 19 EGFR 184 1976 0.98 <0.0001 Yes ** 20 EIF2AK3 566 282 0.93 <0.0001 Yes * 21 EIF2S1 16903 9247 0.90 <0.0001 Yes * 22 ERBB2 2107 1695 0.90 <0.0001 Yes ** 23 FAS 338 681 0.89 <0.0001 Yes * 24 FOXO1 482 1055 0.96 <0.0001 Yes ** 25x GNAI3 193 362 0.94 <0.0001 No 26 HDAC1 1614 1146 0.86 <0.0001 Yes * 27 HSPA5 3830 2390 0.86 <0.0001 Yes * 28 HSPB8 200 947 0.94 <0.0001 Yes ** 29 ITGA3 2436 497 0.95 <0.0001 Yes * 30x ITGB4 206 944 0.87 <0.0001 No 31x KIAA0226 186 104 0.85 <0.0001 No 32 MAPK1 730 1339 0.86 <0.0001 Yes * 33x MLST8 833 453 0.90 <0.0001 No 34 NFE2L2 1410 2622 0.91 <0.0001 Yes * 35 PARP1 2212 975 0.99 <0.0001 Yes * 36 PEA15 5307 3477 0.94 <0.0001 Yes ** 37x PEX3 343 670 0.93 <0.0001 No 38 PTK6 63 556 0.96 <0.0001 Yes ** 39 SQSTM1 4197 2636 0.95 <0.0001 Yes * 40 TP63 131 1067 0.93 <0.0001 Yes * 41 TP73 578 785 0.89 <0.0001 Yes * 42x WIPI1 3043 374 0.99 <0.0001 No

[0063] Through the analysis by means of the Chilibot software on Pubmed, it turned out that 10 genes out of the 42 reported in Table 2 have never been correlated to the diagnosis or prognosis of melanomas. Such genes are: ATG9A, BAG1, CAPN2, CHMP2B, GNAI3, ITGB4, KIAA0226, MLST8, PEX3, WIPI1 (indicated by an x in the Table 2 shown above).

[0064] The ROC curves of each of these 10 genes are depicted in FIG. 1; in particular, in each plot in FIG. 1, it is possible to note on the x axis the specificity %, and on the y axis the sensitivity %.

Example 3

Validation of the Data of Gene Expression on the IST Online Database

[0065] The expression of the 10 genes indicated by an x in Table 2 has been validated by means of a set of expression data of 208 melanoma samples and 147 samples of healthy skin taken from the IST Online database (http://ist.medisapiens.com/), which expresses the expression data as scatter plots. From the analysis of the dataset of IST Online, it has been noted that 4 genes out of the validated 10 have high expression differences in the melanoma with respect to the healthy skin: BAG1, CHMP2B, PEX3 and WIPI1. The scatter plots obtained for BAG1, CHMP2B, PEX3 and WIPI1 have been de novo graphically replotted and are depicted in FIG. 2. In particular, in FIG. 2 it is possible to observe, for each of the 4 genes, the gene expression data detected on melanoma biopsies on the left, and the corresponding gene expression data detected on biopsies of healthy skin on the right. From the analysis of the present example, the gene expression of BAG1, CHMP2B and PEX3 in the melanoma is lower than that of the healthy skin in at least 90% of the melanoma cases, while the gene expression of WIPI1 in the melanoma is greater than that in the healthy skin in at least 90% of the melanoma cases.

Example 4

Evaluation of the Expression Levels of the BAG1, CHMP2B, PEX3 and WIPI1 Proteins on Histological Sections

[0066] The expression level of the proteins of the 4 validated genes BAG1, CHMP2B, PEX3 and WIPI1 has been analyzed on the basis of the histochemical data reported in the Human Protein Atlas (HPA) website (https://www.proteinatlas.org), which contains images of histological sections wherein the proteins are highlighted by means of a bond with labeled antibodies. The expression of the proteins deriving from the 4 identified genes in the previous example has been evaluated thanks to the analysis of 80 images from histological sections (in particular, from 47 images of melanoma and 33 images of healthy skin). The 80 images of the histological sections of HPA have been converted in grey scale by means of the photo editing GNU Image Manipulation Program (GIMP) software, which allowed to obtain quantification of the pixel distribution.

[0067] In FIG. 3 the plots relative to the expression level of each of the 4 proteins analyzed in melanoma and healthy skin samples are depicted. In each of such plots, the dark curves refer to the melanoma samples and the light curves refer to the control samples. In three cases the medians of the dark curves are shifted to the right with respect to the medians of the light curves, thus suggesting a greater expression in the melanoma, for the BAG1, PEX3 and WIPI1 proteins. Such plots have been obtained starting from the quantification of the pixel distribution, and in particular have been obtained by carrying out the following steps: [0068] 1) selection of the image; [0069] 2) removal of the background (by means of the conversion of the background from white into transparent); [0070] 3) conversion of the RGB image color into greyscale; and [0071] 4) production of histograms.

[0072] The information referring to the mean, pixels, median and standard deviation present in FIG. 3 has been obtained by selecting, in the image menu of GIMP: Color.fwdarw.Information.fwdarw.Histogram.

[0073] The expression of the CHMP2B protein results to be unchanged in the melanoma with respect to the normal tissues.

[0074] It should be noted that the gene expression levels of BAG1 and PEX3 (detected in the validation of Example 3) and the total level of the proteins encoded by them (evaluated in the present Example) do not correspond, i.e. the gene expression of BAG1 and PEX3 is lower in the melanoma with respect to the healthy skin, however the expression of the proteins encoded by them is greater in the melanoma with respect to the healthy skin. Such contradiction is only apparent, indeed the gene expression levels depend on the transcription mechanism, while the protein expression levels are based on the reactive adaptation of the synthesis/degradation mechanism of the proteins in response to the modified gene expression levels. On the other hand the expression levels of WIPI1 are increased in the melanoma samples with respect to the healthy skin samples, both at the level of gene transcription (see Example 3) and of protein expression (Example 4).

Example 5

Evaluation of the Prognostic Value of the Considered Genes

[0075] The prognostic value of the BAG1, CHMP22B, PEX3 and WIPI1 genes has been evaluated by means of the data reported in HPA relative to patients with melanoma with 3 years-survival data. Such patients have been stratified based on the high or low expression level of the 4 genes under evaluation, and then the progression, positive or negative, of the tumoral pathology (based on the survival of the patient) has been associated to each so-stratified patient. This way, it has been possible to associate the high or low expression level of each of the 4 genes with a higher or lower survival of the patients.

[0076] From the evaluation of the present example, the genes with prognostic value, among the four considered genes, are BAG1 and WIPI1. In particular: the analysis of the survival in 102 patients with melanoma showed a prognostic value indicating a greater survival in the patients in which the BAG1 expression level in the melanoma is low (p=0.04). The survival analysis in 102 patients with melanoma showed that WIPI1 is a favorable prognostic marker in melanoma (p<0.0001).