In vitro method for screening, diagnosis and / or prognosis of liver cancer; or for predicting the response of subjects suffering from liver cancer to therapy

Abstract

The present invention refers to an in vitro method for screening, diagnosis and / or prognosis of liver cancer; or for predicting the response of subjects suffering from liver cancer to therapy.

Description

[0001]IN VITRO METHOD FOR SCREENING, DIAGNOSIS AND / OR PROGNOSIS OFLIVER CANCER; OR FOR PREDICTING THE RESPONSE OF SUBJECTSSUFFERING FROM LIVER CANCER TO THERAPY FIELD OF THE INVENTION The present invention refers to the medical field. Particularly, the present invention is directedto an in vitro method for screening, diagnosis and / or prognosis of liver cancer; or for predictingthe response of subjects suffering from liver cancer to therapy. STATE OF THE ART Liver cancer is a major health problem with more than 900,000 new cases annually worldwide. This neoplasm is currently a global health concern, and its incidence is expected to increase in the future. Hepatocellular carcinoma (HCC), the most common form (~90%) of liver cancer, ranks as the sixth most common neoplasm and the third leading cause of cancer-related death partly due to late diagnosis and a poor response to current treatments. The development of HCC is closely linked to the presence of cirrhosis and other chronic liver diseases. The worldwide incidence is heterogeneous because of the variable prevalence of risk factors. Hepatitis B and C virus, alcoholic liver disease, hemochromatosis, and several others represent the mainstay of risk factors for the development of HCC. Furthermore, recent studies have reported non- alcoholic fatty liver disease (NAFLD) to be an underlying cause of HCC in a number of cases even in the absence of cirrhosis. Importantly, NAFLD is a growing epidemic in developed regions, in part due to the increasing incidence of obesity and type 2 diabetes. As with many other cancers, HCC detected at an early stage has a better prognosis compared to advanced stage disease, in part due to the relative efficacy of local treatments compared with systemic therapies. However, a large percentage of subjects are diagnosed at advanced stages. The Barcelona Clinic Liver Cancer (BCLC) system establishes staging, prognosis, and treatment guidelines for hepatocellular carcinoma (HCC), serving as the widely used reference for decision-making in HCC worldwide. In general, and without taking into account the sub- classifications made within each BCLC group aimed at making completely individualized decisions, subjects in very early (BCLC 0) or early (BCLC A) stages might benefit from surgical resection, liver transplantation and radiofrequency ablation, reaching a 5-year survival rate of >75%. However, a high percentage of these subjects suffer from recurrences after these putative curative treatments. Subjects in intermediate stage (BCLC B) will be candidates for transarterial chemoembolization (TACE), with a median overall survival (OS) >2,5 years. Subjects with advanced-stage HCC (BCLC C) are treated with different systemic therapies. After nearly 20 years during which the use of sorafenib was the only option for these subjects, we are now experiencing a boom in new therapies based on immunotherapy. However, and despite significant advancements in survival outcomes compared to sorafenib, many subjects with advanced-stage HCC do not experience durable benefits from these treatment regimens. Finally, end-stage HCC subjects belong to the BCLC stage D, and symptomatic management and coordination of palliative care are the main goals. Conventional non-invasive diagnostic methods for the detection of HCC include image examination (ultrasonography, computed tomography and magnetic resonance tomography) in combination with the serological detection of alpha-fetoprotein (AFP) levels. However, these methods show limited sensitivity and specificity, and most subjects diagnosed with HCC are already in advanced stage. Alternatively, the still gold standard method for HCC diagnosis is based on biopsy specimen analyses. However, tissue pathology-based approaches are invasive procedures, suffer from high cost, limited tumour tissue accessibility and the fact that a tissue biopsy may not accurately reflect the intra-tumour heterogeneity.Therefore, there is an unmet medical need of finding reliable and minimally invasive tools forthe diagnosis and / or prognosis of liver cancer which offer high sensitivity and specificity.Moreover, there is an urgent need for the identification of new predictive biomarkers thatenable the accurate selection of subjects prior to treatment.The present invention is focused on solving this problem and an innovative in vitro method forthe diagnosis and / or prognosis of liver cancer; or for predicting the response of subjects suffering from liver cancer to therapy, is herein provided. DESCRIPTION OF THE INVENTION Brief description of the inventionThe present invention is directed to an in vitro method for screening, diagnosis and / or prognosisof liver cancer; or for predicting the response of subjects suffering from liver cancer to therapy.Particularly, the inventors of the present invention carried out a screening process to identifymethylation markers (Example 2.1, Figure 1, Figure 2, Figure 3 and Figure 4). Once themethylation markers were identified (HepaMeth Panel, Table 2), they were validated in silico(Example 2.2 and Figure 5) and in control and HCC tissues (Example 2.3 and Figure 6).Moreover, the diagnostic performance of the elected methylation markers was evaluated(Example 2.4 and Figure 7) and the diagnostic capacity of elected methylation markers wasassessed in plasma cfDNA compared to the gold-standard alpha-fetoprotein (Example 2.5 andFigure 8). Moreover, the performance of the elected methylation markers was assessed inplasma for the early diagnostic of HCC in comparison with the gold-standard alpha-fetoprotein(Example 2.6 and Figure 9). On the other hand, the prognostic capacity of the electedmethylation markers was evaluated in plasma cfDNA (Example 2.7 and Figure 10). Finally,the capacity of the elected methylation markers to predict the response to immunotherapy usingplasma cfDNA was assessed (Example 2.8 and Figure 11).Altogether, the results provided in Example 2 demonstrate that the individual methylationmarkers included in the HepaMeth Panel (see Table 2), or any combination thereof comprisingtwo, three or the four methylation markers, can be used for screening, diagnosis and / orprognosis of liver cancer, or for predicting the response of subjects suffering from liver cancerto therapy, particularly immunotherapy.Consequently, the first embodiment of the present invention refers to an in vitro method forscreening, diagnosis and / or prognosis of liver cancer, which comprises determining themethylation status of the gene MIXL1 in a biological sample obtained from the subject, whereina higher level of methylation of the gene MIXL1, as compared with a reference level of methylation of the gene MIXL1 measured in healthy control subjects, is an indication that thesubject is suffering from liver cancer and / or that the subject has a poor prognosis.In a preferred embodiment, the present invention refers to an in vitro method for screening,diagnosis and / or prognosis of liver cancer, which comprises determining the methylation statusof the gene MIXL1 in combination with at least a gene selected from: HIST3H2A, TSC22D1and / or RNF135, wherein a higher level of methylation of any of these genes, as compared with a reference level of methylation measured in healthy control subjects, is an indication that thesubject is suffering from liver cancer and / or that the subject has a poor prognosis.In a preferred embodiment, the present invention refers to an in vitro method for the diagnosisand / or prognosis of liver cancer, which comprises determining the methylation status of thefollowing combination of genes: MIXL1, HIST3H2A, TSC22D1 and RNF135, wherein a higherlevel of methylation of any of these genes, as compared with a reference level of methylationmeasured in healthy control subjects, is an indication that the subject is suffering from livercancer and / or that the subject has a poor prognosis.The second embodiment of the present invention refers to the in vitro use of the methylationstatus of the gene MIXL1, or of a kit comprising reagents for determining the methylation statusof the gene MIXL1, for screening, diagnosis and / or prognosis of liver cancer.In a preferred embodiment, the present invention refers to the in vitro use of the methylationstatus of the gene MIXL1 in combination with the methylation status of at least a gene selectedfrom: HIST3H2A, TSC22D1 and / or RNF135, or of a kit comprising reagents for determiningthe methylation status of the gene MIXL1 in combination with the methylation status of at leasta gene selected from: HIST3H2A, TSC22D1 and / or RNF135, for screening, diagnosis and / orprognosis of liver cancer.In a preferred embodiment, the present invention refers to the in vitro use of the methylationstatus of the following combination of genes: MIXL1, HIST3H2A, TSC22D1 and RNF135, orof a kit comprising reagents for determining the methylation status of the followingcombination of genes: MIXL1, HIST3H2A, TSC22D1 and RNF135, for screening, diagnosisand / or prognosis of liver cancer.The third embodiment of the present invention refers to a kit, suitable for screening, diagnosisand / or prognosis of liver cancer, which comprises: a) A pair of primers for the amplificationof the fragment chr1:226223965-226224144 of gene MIXL1, wherein at least one primer, preferably both, is selected from the forward primer TYGAATAGTTGTAGTTGTTGGAGTT(SEQ ID NO: 1) and the reverse primer CCRCAATCCAAACCTAAAATCC (SEQ ID NO:2). In a preferred embodiment, the kit of the invention further comprises at least one pair of primersselected from: b) A pair of primers for the amplification of the fragment chr1:228457730- 228457883 of gene HIST3H2A, wherein at least one primer, preferably both, is selected fromthe forward primer GGTTTAYGGGGAATTGTAGTTT (SEQ ID NO: 3) and the reverseprimer CTTCACTACCCTCTTATTTTTAATCT (SEQ ID NO: 4); c) A pair of primers for theamplification of the fragment chr13:44576100-44576218 of gene TSC22D1, wherein at least one primer, preferably both, is selected from the forward primerGTTTTTGGTTGAGGAGGAG (SEQ ID NO: 5) and the reverse primerAAAACRAATACAATTTCCTTCTACAC (SEQ ID NO: 6); and / or d) A pair of primers forthe amplification of the fragment chr17:30971079-30971222 of gene RNF135, wherein at least one primer, preferably both, is selected from the forward primer GGGTTTGGGTTTGGGTTT(SEQ ID NO: 7) and reverse primer CCCCACAAAACCTCCAAACAA (SEQ ID NO: 8).In a preferred embodiment, the kit comprises the four pair of primers a), b), c) and d). Morepreferably, the kit comprises both forward primer and reverse primer of each pair of primerscomprises a), b), c), and d) [SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 3, SEQ ID NO:4,SEQ ID NO: 5, SEQ ID NO: 6, SEQ ID NO: 7, and SEQ ID NO: 8].The fourth embodiment of the present invention refers to and in vitro method for predicting theresponse of subjects suffering from liver cancer to a treatment with immunotherapy whichcomprises determining the methylation status of the genes MIXL1, HIST3H2A, TSC22D1 andRNF135 in a biological sample obtained from the subject, wherein a higher level of methylationof at least one, preferably at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment; or wherein a lower level of methylation of at least one, preferably at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a better response to the treatment.The fifth embodiment of the present invention refers to an in vitro method for selecting subjectssuffering from liver cancer for a treatment with immunotherapy which comprises determiningthe methylation status of the genes MIXL1, HIST3H2A, TSC22D1 and RNF135 in a biologicalsample obtained from the subject, wherein a higher level of methylation of at least one, preferably at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment; or wherein a lower level of methylation of at least one, preferably at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a better response to the treatment. In a preferred embodiment of the invention, immunotherapy may comprise antibodies,preferably monoclonal antibodies, or any region thereof such as F(ab´)2, Fab, scFab, scFv, ora camelid variable heavy-chain-only domain, binding PD1 and / or PD-L1. For instance, thetherapy may comprise CAR-T cells, wherein chimeric antigen receptor (CAR) comprises anantibody, F(ab´)2, Fab, scFab, scFv with a light chain variable region (VL) and a heavy chainvariable region (VH) , or a camelid variable heavy-chain-only domain, binding PD1 and / orPD-L1. On the other hand, the therapy could be T cells comprising a TCR like antibody bindingPD1 and / or PD-L1 in the context of MHC Class I or II.The sixth embodiment of the present invention refers to an anti PD1 and / or PD-L1 antibody(preferably monoclonal antibody) for use in a method for the treatment of liver cancer, whereinthe method comprises predicting the response of subjects suffering from liver cancer to a treatment with immunotherapy by determining the methylation status of the genes MIXL1,HIST3H2A, TSC22D1 and RNF135 in a biological sample obtained from the subject, whereina higher level of methylation of at least one, preferably at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment; or wherein a lower levelof methylation of at least one, preferably at least two of the genes, as compared with a referencelevel of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a better response to the treatment.Alternatively, the sixth embodiment of the present invention refers to a method for treatingliver cancer which comprises the administration of a pharmaceutically effective amount of ananti PD1 and / or PD-L1 antibody (preferably monoclonal antibody), wherein the methodcomprises predicting the response of subjects suffering from liver cancer to a treatment with immunotherapy by determining the methylation status of the genes MIXL1, HIST3H2A,TSC22D1 and RNF135 in a biological sample obtained from the subject, wherein a higher levelof methylation of at least one, preferably at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment; or wherein a lower level of methylation ofat least one, preferably at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a better response to the treatment. In a preferred embodiment, the methylation status of the gene is determined in at least a CpG site of the gene; preferably a CpG site of the promoter region. In a preferred embodiment, the methylation status of the gene MIXL1 is determined in at least a CpG site of the promoter region located between the chromosomal positionschr1:226223965-226224144 (SEQ ID NO: 9), the methylation status of the gene HIST3H2A isdetermined in at least a CpG site of the promoter region located between the chromosomalpositions chr1:228457730-228457883 (SEQ ID NO: 11), the methylation status of the geneTSC22D1 is determined in at least a CpG site of the promoter region located between thechromosomal positions chr13:44576100-44576218 (SEQ ID NO: 13), and / or the methylationstatus of the gene RNF135 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr17: 30971079-30971222 (SEQ ID NO: 15) according to the human reference genome hg38.In a preferred embodiment of the invention, the methylation status of the gene MIXL1 isdetermined at the position chr1:226224014, the methylation status of the gene HIST3H2A isdetermined at the position chr1:228457781, the methylation status of the gene TSC22D1 isdetermined at the position chr13:44576127, and / or the methylation status of the gene RNF135 is determined at the position chr17:30971166.In a preferred embodiment of the invention, the liver cancer is hepatocellular carcinoma.In a preferred embodiment of the invention, the biological sample is a liquid biopsy selected from: plasma, blood, serum, urine or bile. In a preferred embodiment of the invention, immunotherapy refers to anti PD1 / PD-L1antibodies, preferably selected from: Nivolumab, pembrolizumab, dostarlimab, sintilimab,cemiplimab, atezolizumab, durvalumab, avelumab, toripalimab, camrelizumab, tislelizumab,zimberelimab, or prolgolimab. Alternatively, the present invention refers to:A method for identifying or detecting biomarker signatures for screening, diagnosis and / orprognosis of liver cancer; or for predicting the response of subjects suffering from liver cancer to therapy, which comprises assessing the methylation status of the genes, DNA fragments orCpG sites disclosed in Table 2, wherein the identification of a statically significant deviationof the methylation status is indicative that the methylation status of the genes, DNA fragmentsor CpG sites of Table 2 may be used for screening, diagnosis and / or prognosis of liver cancer;or for predicting the response of subjects suffering from liver cancer to therapy. In a preferred embodiment, the present invention is a computer-implemented invention,wherein a processing unit (hardware) and a software are configured to: a) Receive themethylation status of the genes, DNA fragments or CpG sites disclosed in Table 2, b) processthe methylation status for finding substantial variations or deviations, and c) provide an outputthrough a terminal display of the variation or deviation of the methylation status, wherein thevariation or deviation of the methylation status is indicative of the diagnosis and / or prognosisof liver cancer; or is predictive of the response of subjects suffering from liver cancer totherapy.Moreover, Example 3 shows the sensitivity of individual biomarkers, or combinations of two,three or four biomarkers for screening, diagnosis of liver cancer (n=117 HCCs) (see Table 3),Example 4 shows the prognostic value of the assayed biomarkers and Example 5 shows thecapacity of the assayed biomarkers to predict response to immunotherapy in subjects sufferingfrom liver cancer.The results provided by Example 3, Example 4 and Example 5 confirm that HIST3H2A, aloneor in combination with MIXL1, TSC22D1 and / or RNF135, is a reliable biomarker for screening,diagnosis and / or prognosis of liver cancer, or for predicting the response of subjects sufferingfrom liver cancer to a treatment with anti-PD1 or anti-PDL1 immunotherapy.Consequently, the present invention refers to an in vitro method for screening, diagnosis and / orprognosis of liver cancer, which comprises determining the methylation status of the geneHIST3H2A in a biological sample obtained from the subject, wherein a higher level ofmethylation of the gene HIST3H2A, as compared with a reference level of methylation of thegene HIST3H2A measured in healthy control subjects, is an indication that the subject is at riskof suffering from liver cancer, is suffering from liver cancer and / or has a poor prognosis ; or tothe use of HIST3H2A for screening, diagnosis and / or prognosis of liver cancer. In a preferred embodiment, the method or use according to the present invention comprisesdetermining the methylation status of the gene HIST3H2A in combination with at least a geneselected from: MIXL1, TSC22D1 and / or RNF135, wherein a higher level of methylation of anyof these genes, as compared with a reference level of methylation measured in healthy control subjects, is an indication that the subject is at risk of suffering from liver cancer, is suffering from liver cancer and / or has a poor prognosis. In a preferred embodiment, the method or use according to the present invention comprises determining the methylation status of the following combination of genes: MIXL1, HIST3H2A,TSC22D1 and RNF135, wherein a higher level of methylation of any of these genes, ascompared with a reference level of methylation measured in healthy control subjects, is an indication that the subject is at risk of suffering from liver cancer, is suffering from liver cancer and / or has a poor prognosis. In a preferred embodiment, the methylation status of the gene is determined in at least a CpG site of the gene; preferably a CpG site of the promoter region. In a preferred embodiment, the methylation status of the gene MIXL1 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr1:226223965-226224144, the methylation status of the gene HIST3H2A is determined in at least a CpG site of the promoter region located between the chromosomal positions chr1:228457730-228457883, the methylation status of the gene TSC22D1 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr13:44576100-44576218, and / or the methylation status of the gene RNF135 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr17: 30971079-30971222 according to the human reference genome hg38.The present invention also refers to a kit, suitable for screening, diagnosis and / or prognosis ofliver cancer, which comprises: a) a pair of primers for the amplification of the fragmentchr1:228457730-228457883 of gene HIST3H2A, wherein at least one primer, preferably both, is selected from the forward primer GGTTTAYGGGGAATTGTAGTTT (SEQ ID NO: 3) and the reverse primer CTTCACTACCCTCTTATTTTTAATCT (SEQ ID NO: 4). In a preferred embodiment, the kit further comprises at least one pair of primers selected from:b) a pair of primers for the amplification of the fragment chr1:226223965-226224144 of geneMIXL1, wherein at least one primer, preferably both, is selected from the forward primer TYGAATAGTTGTAGTTGTTGGAGTT (SEQ ID NO: 1) and the reverse primerCCRCAATCCAAACCTAAAATCC (SEQ ID NO: 2); c) a pair of primers for theamplification of the fragment chr13:44576100-44576218 of gene TSC22D1, wherein at least one primer, preferably both, is selected from the forward primerGTTTTTGGTTGAGGAGGAG (SEQ ID NO: 5) and the reverse primerAAAACRAATACAATTTCCTTCTACAC (SEQ ID NO: 6); and / or d) a pair of primers forthe amplification of the fragment chr17:30971079-30971222 of gene RNF135, wherein at least one primer, preferably both, is selected from the forward primer GGGTTTGGGTTTGGGTTT (SEQ ID NO: 7) and reverse primer CCCCACAAAACCTCCAAACAA (SEQ ID NO: 8).In a preferred embodiment, the kit comprises: a) a pair of primers for the amplification of thefragment chr1:228457730-228457883 of gene HIST3H2A, wherein at least one primer, preferably both, is selected from the forward primer GGTTTAYGGGGAATTGTAGTTT (SEQ ID NO: 3) and the reverse primer CTTCACTACCCTCTTATTTTTAATCT (SEQ IDNO: 4), b) a pair of primers for the amplification of the fragment chr1:226223965-226224144of gene MIXL1, wherein at least one primer, preferably both, is selected from the forward primer TYGAATAGTTGTAGTTGTTGGAGTT (SEQ ID NO: 1) and the reverse primerCCRCAATCCAAACCTAAAATCC (SEQ ID NO: 2), c) a pair of primers for theamplification of the fragment chr13:44576100-44576218 of gene TSC22D1, wherein at least one primer, preferably both, is selected from the forward primerGTTTTTGGTTGAGGAGGAG (SEQ ID NO: 5) and the reverse primerAAAACRAATACAATTTCCTTCTACAC (SEQ ID NO: 6); and d) a pair of primers for theamplification of the fragment chr17:30971079-30971222 of gene RNF135, wherein at least one primer, preferably both, is selected from the forward primer GGGTTTGGGTTTGGGTTT (SEQ ID NO: 7) and reverse primer CCCCACAAAACCTCCAAACAA (SEQ ID NO: 8).The present invention also refers to an in vitro method for predicting the response of subjectssuffering from liver cancer to a treatment with anti-PD1 or anti-PDL1 immunotherapy whichcomprises determining the methylation status of the gene HIST3H2A in a biological sampleobtained from the subject, wherein a higher level of methylation, as compared with a reference level of methylation of the gene measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.In a preferred embodiment, the method comprises determining the methylation status of thegene HIST3H2A in combination with at least a gene selected from: MIXL1, TSC22D1 and / orRNF135 in a biological sample obtained from the subject, wherein a higher level ofmethylation, as compared with a reference level of methylation of the gene measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment. In a preferred embodiment, the method comprises determining the methylation status of thegenes MIXL1, HIST3H2A, TSC22D1 and RNF135 in a biological sample obtained from thesubject, wherein a higher level of methylation of at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.The present invention also refers to an in vitro method for selecting subjects suffering fromliver cancer for a treatment with anti-PD1 or anti-PDL1 immunotherapy which comprisesdetermining the methylation status of the genes HIST3H2A in a biological sample obtainedfrom the subject, wherein a higher level of methylation of the gene, as compared with a reference level of methylation measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.In a preferred embodiment, the method comprises determining the methylation status of thegenes HIST3H2A in combination with at least a gene selected from: MIXL1, TSC22D1 and / orRNF135, in a biological sample obtained from the subject, wherein a higher level of methylation of the genes, as compared with a reference level of methylation measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.In a preferred embodiment, the method comprises determining the methylation status of thegenes MIXL1, HIST3H2A, TSC22D1 and RNF135 in a biological sample obtained from thesubject, wherein a higher level of methylation of at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment. The present invention also refers to anti PD1 and / or PD-L1 antibody for use in a method for the treatment of liver cancer, wherein the method comprises predicting the response of subjects suffering from liver cancer to a treatment with anti PD1 and / or anti PD-L1 immunotherapy by determining the methylation status of the gene HIST3H2A, in a biological sample obtained from the subject, wherein a higher level of methylation of the gene, as compared with a reference level of methylation of the genes measured in healthy control subjects, is anindication that the subject will have a worse response to the treatment. In other words, themethod comprises selecting subjects suffering from liver cancer who are not classified as having a worse response to a treatment with anti PD1 and / or anti PD-L1 immunotherapy by implementing the above method for selecting subjects suffering from liver cancer for a treatment with anti-PD1 or anti-PDL1 immunotherapy.In a preferred embodiment, the method comprises predicting the response of subjects sufferingfrom liver cancer to a treatment with anti PD1 and / or anti PD-L1 immunotherapy by determining the methylation status of the gene HIST3H2A in combination with at least a geneselected from: MIXL1, TSC22D1 and / or RNF135, in a biological sample obtained from thesubject, wherein a higher level of methylation of the gene, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.In a preferred embodiment, the method comprises predicting the response of subjects sufferingfrom liver cancer to a treatment with immunotherapy by determining the methylation status ofthe genes MIXL1, HIST3H2A, TSC22D1 and RNF135 in a biological sample obtained from thesubject, wherein a higher level of methylation of at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment. For the purpose of the present invention, the following terms are defined: ^“CpG site”: The CpG sites refers to dinucleotides formed by a cytosine followed by aguanine. Cytosines in CpG dinucleotide can be methylated to form 5-methylcytosines. ^“CpG island” (CGI): CpG islands (or CG islands) are regions with a high frequency ofCpG sites. The usual formal definition is a region with at least 200 bp and a GC percentage greater than 50%.^ “Beta-value”: The beta-value provides a quantitative measure of methylation at aspecific locus. It is calculated as the ratio of methylated signal intensity to the total signal intensity (M / M+U) at a specific cytosine site. Its value ranges from 0 (unmethylated) to 1 (fully methylated).^ “Bisulfite treatment”: Treatment of DNA with sodium bisulfite converts unmethylatedcytosine to uracil, which is subsequently converted to thymine during PCR amplification, while methylcytosine remains unchanged, as cytosine.^ “Promoter region”: The promoter region of the gene comprises the sequences of the5’UTR and the first exon of the gene.^ As used herein, the term "methylation" will be understood to mean the presence of amethyl group added by the action of a DNA methyl transferase enzyme to a cytosine base followed by a guanine (CpG) nucleotide of a nucleic acid.^ Accordingly, the term, “methylation status” as used herein refers to the presence orabsence of methylation in a specific nucleic acid region.^ The expression “higher level of methylation” refers to an increase in the relative amountof methylation of a nucleic acid, as compared with the subject used as control that, in this case, are healthy subjects or subjects not suffering from liver cancer. Thus, in the present disclosure, the “higher level of methylation” is generally determined with reference to a baseline level represented by the methylation status of a given genomicregion in a sample obtained from control subjects. For example, “higher level of methylation” may be at least 2% greater than the baseline level of methylation, for example at least 5% greater than the baseline level of methylation, or at least 10% greater than the baseline level of methylation, or at least 15% greater than the baseline level of methylation, or at least 20% greater than the baseline level of methylation, or at least 25% greater than the baseline level of methylation, or at least 30% greater than the baseline level of methylation, or at least 40% g greater than the baseline level of methylation, or at least 50% greater than the baseline level of methylation, or at least 60% greater than the baseline level of methylation, or at least 70% greater than the baseline level of methylation, or at least 80% greater than the baseline level of methylation, or at least 90% greater than the baseline level of methylation.^ The term "comprising" means "including", but not limited to what follows the term"comprising". Thus, the use of the term "comprising" indicates that the elements listed are necessary or mandatory, but that other elements are optional and may or may not be present. ^The term "consisting of" means "including" but is limited to what follows the term"consisting of". Thus, the term "consists of" indicates that the elements listed are mandatory, and that other elements may not be present. ^The term “screening” refers to the systematic application of the test or procedure toidentify individuals who may have liver cancer in a population of asymptomatic or at- risk individuals. The goal of screening is early detection to facilitate timely intervention and improve outcomes. Screening typically precedes diagnostic testing and does not provide a definitive diagnosis. For example, liver cancer screening might involve imaging tests like ultrasound or blood tests for markers such as alpha-fetoprotein (AFP). So, the subject may be a subject having a risk factor of developing liver cancer; e.g., arisk factor such as chronic infection with hepatitis B virus (HBV) or hepatitis C virus (HCV), cirrhosis, non-alcoholic fatty liver disease, steato-hepatitis, hereditary hemochromatosis or other hereditary diseases prone to develop cirrhosis, heavy alcoholand / or tobacco use, obesity, type 2 diabetes, certain rare disease (e.g., Tyrosinemia, Alpha1-antitrypsin deficiency, Porphyria cutanea tarda, Glycogen storage diseases, Wilson disease), and / or Aflataxin B1, and the present method can be advantageously used for screening and early detect liver cancer in a subject being at risk of developingliver cancer, and in particular in programs for periodically screening of those subjects having any of the mentioned risks factors. ^The term “diagnosis” refers to the process of determining the presence or absence of aspecific disease or condition in an individual, typically following the presentation of symptoms or an abnormal screening result. Diagnosis involves the integration of clinical evaluation, diagnostic tests, imaging, and sometimes invasive procedures (e.g., biopsy) to confirm the disease and its characteristics (e.g., staging or subtype in liver cancer). Brief description of the figuresFigure 1. HCC tissue methylomes (GSE56588, GSE54503 and TCGA-LIHC) used for the insilico identification of HepaMeth Panel.Figure 2. Workflow followed to reduce the list of 3862 hypermethylated CpGs in HCCcompared to control tissues to a smaller panel of 39 hypermethylated CpGs.Figure 3. Methylation value of the 39 CpG candidates in leukocytes (n=656; GSE40279),control livers (n=135) and HCC tissue (n=670) of the three studies used for the identificationof DNA hypermethylated markers (GSE56588, GSE54503 and TCGA-LIHC).Figure 4. Heatmap showing the methylation value of the four CpGs of HepaMeth Panel incontrol livers (n=135) and HCC tissue (n=670) of the three studies used for the identificationof the Panel (GSE56588, GSE54503 and TCGA-LIHC). The in silico performance ofHepaMeth Panel is shown.Figure 5. Heatmap showing the methylation value of the four CpGs of HepaMeth Panel incontrol livers and HCC tissue of the different studies (GSE60753, GSE89852 andGSE157341). The in silico performance of HepaMeth Panel in each study is shown.Figure 6. Heatmap showing the methylation value of the four CpGs of HepaMeth Panel in 19HCC tissue. The empirical performance of HepaMeth Panel is shown. White color reflects nomethylation.Figure 7. Heatmap comparing the results of HepaMeth considering positive (grey) when theCpG initially identified was methylated with those of HepaMeth considering the methylationvalue of all the CpGs presented in each amplified DNA fragment. For this comparison, a groupof HCC samples with a negative result was included in our previous HepaMeth analyses. Thesensitivity and statistics are provided.Figure 8. Number of HCC subjects positive (black) and negative (grey) for HepaMeth Paneland AFP (>15 ng / mL).Figure 9. Number of early (BCLC 0 / A) HCC subjects positive (black) and negative (grey) forHepaMeth Panel and AFP (>15 ng / mL).Figure 10. Overall survival of HCC subjects divided according to positive or negativeHepaMeth Panel.Figure 11. Overall survival of HCC subjects divided according to positive (2 or more markerspositive) or negative HepaMeth Panel.Figure 12, 13, 14 and 15. Overall survival of HCC subjects divided according to positive ornegative methylation of the assayed biomarkers.Detailed description of the invention The present invention is illustrated by means of the Examples set below without the intention of limiting its scope of protection. Example 1. MATERIAL AND METHODS Example 1.1. Protocols to evaluate the DNA methylation status of the biomarkers ^Sample: Any sample containing DNA could be used. Therefore, the starting materialincluded tissue, biopsy, plasma, serum, bile, urine and the like. Here we provide examples using liver tissue and plasma. ^DNA: Genomic DNA and fragmented DNA, including circulating cell-free DNA, couldbe used. Here we provide examples using both genomic DNA and circulating cell-free DNA. ^Isolation of DNA: Any protocol to isolate DNA from a sample could be used.^ Methylation analysis: Any method to analyze the DNA methylation status could beused, including those requiring bisulfite treatment and those that do not require it, including methylation sensitive restriction enzyme treatment, MeDIP method, MSP, nanopore real time sequencing, microarray analysis, mass spectrometry, and the like. We provide examples using bisulfite treatment, amplification by PCR with specific primers and sequencing both by Sanger and targeted multiplex NGS.Example 1.2. Protocols used related to the results provided in Example 2Example 1.2.3. Sample preparation: Isolation of DNA: -Cell-free DNA (cfDNA) from plasma samples. Purified plasma (1-2 mL) was slowly thawedin ice, and cfDNA was extracted using the QIAamp Circulating Nucleic Acid kit (Qiagen,Hilden, Germany), following manufacturer instructions. -Genomic DNA (gDNA) from liver tissue. Genomic DNA from frozen tissues was isolated using the Maxwell® RSC Cultured Cells DNA Purification Kit with a Maxwell® RSC 48 instrument (Promega, Madison, WI, USA; AS1620). cfDNA QC: -cfDNA concentration was quantified using the QuantiFluor® dsDNA Sample Kit (Promega) and cfDNA size distributions was analyzed using Agilent cell-free Screen Tape and reagents on the Agilent TapeStation System (Agilent Technologies, Santa Clara, CA). -gDNA purity and concentration were measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Bisulfite conversion: 15 ng of isolated cfDNA or gDNA was modified upon bisulfite treatment using the EZ DNA Methylation Kit (Zymo Research, Irvine, CA, USA) following manufacturer instructions, and clean up with EZ DNA Methylation Lightning Kit (Zymo Research, Irvine, CA, USA). Target enrichment: Individual PCRs are carried out with 2.5 ng of converted DNA. Annealing temperature was57°C for HIST3H2A and 59°C for MIXL1, TSC22D1 and RNF135. Primers are listed in thetable below. Table 1 Example 1.3. Protocol based on SANGER sequencing Electrophoresis. PCR products were visualized in 1.8% agarose gel electrophoresis. DNA fragments were excise from the agarose gel with a clean, sharp scalpel. PCR products purification. DNA fragments were isolated from the gel and purified using the QIAquick Gel Extraction Kit (Qiagen), following manufacturer instructions. Sanger sequencing. Purified PCR products are sequenced by Sanger. The methylated status of each CpG is evidenced as C (methylated) or T (unmethylated) in the corresponding CpG dinucleotides. Example 1.4. Protocol based on targeted bisulfite sequencing Post-Target Enrichment. QC analyses with D1000 ScreenTape System and reagents on the Agilent TapeStation System (Agilent Technologies, Santa Clara, CA) to verify amplification products. Amplicons were pooled for each sample at equal concentration based on reported molarity from TapeStation. Barcoding. Pools were used in barcoding PCR reaction (Zymo Research Company). Post-Barcoding. Barcoded products were verified by QC analyses with D1000 ScreenTape System and reagents on the Agilent TapeStation System (Agilent Technologies, Santa Clara, CA). All libraries were pooled, cleaned-up and QC.Sequencing and Analysis. Sequencing was carried out on MiSeq (Illumina) at 2x 150bp pair-end. Read alignment and methylation calling were performed. In the examples provided the mean methylation value of each fragment was calculated. Example 1.5. Detailed information of the four CpG of interest, the primers designed to amplify the region in bisulfite converted DNA, the whole amplified fragment and the surrounding CpGs within the fragmentAll CpGs within the fragment are underlined. The CpG of interest is underlined and writtenin bold. Fragment of interest: chr1:226223965-226224144Gene: MIXL1 > Genomic Amplicon Sequence: CCGAACAGCTGCAGCTGCTGGAGCTCGTCTTCCGCCGGACCCGGTACCCCGACA TCCACTTGCGCGAGCGCCTGGCCGCGCTCACCCTGCTCCCCGAGTCCAGGATCCA GGTGAGGGCCCGCTGCGTTCGCAAGTGCGCGCTGGAGCGGAGGCGCTGCGGACTCTAGGTCTGGACTGCGG (SEQ ID NO: 9).> Bisulfite Amplicon Sequence with primer: TCGAATAGTTGTAGTTGTTGGAGTTCGTTTTTCGTCGGATTCGGTATTTCGATATTT ATTTGCGCGAGCGTTTGGTCGCGTTTATTTTGTTTTTCGAGTTTAGGATTTAGGTG AGGGTTCGTTGCGTTCGTAAGTGCGCGTTGGAGCGGAGGCGTTGCGGATTTTAGGTTTGGATTGCGG (SEQ ID NO: 10).> Primers:Forward: TYGAATAGTTGTAGTTGTTGGAGTT (SEQ ID NO: 1).Reverse: CCRCAATCCAAACCTAAAATCC (SEQ ID NO: 2).> Fragment of 180 bp with 19 CpGs. Fragment of interest: chr1:228457730-228457883Gene: HIST3H2A > Genomic Amplicon Sequence: GGCCCACGGGGAACTGCAGCCCCGCGCGCGACGAGCGCGACTTAGCCTTGGCGC GCGCCTTGCCACCCTGCTTACCACGACCGGACATTTCCGAGTCAAGGAAAAAAGACAACGGCAACCGAAAAGCGAGACTAAAAACAAGAGGGCAGTGAAG (SEQ IDNO: 11). > Bisulfite Amplicon Sequence with primer: GGTTTACGGGGAATTGTAGTTTCGCGCGCGACGAGCGCGATTTAGTTTTGGCGCGC GTTTTGTTATTTTGTTTATTACGATCGGATATTTTCGAGTTAAGGAAAAAAGATAACGGTAATCGAAAAGCGAGATTAAAAATAAGAGGGTAGTGAAG (SEQ ID NO: 12).> Primers:Forward: GGTTTAYGGGGAATTGTAGTTT (degenerated primer) (SEQ ID NO: 3).Reverse: CTTCACTACCCTCTTATTTTTAATCT (SEQ ID NO: 4).> Fragment of 154 bp with 16 CpGs. Fragment of interest: chr13:44576100-44576218Gene: TSC22D1 > Genomic Amplicon Sequence: GCCTTTGGCTGAGGAGGAGGAGAAGGAGGAATCGCGCCAGGCGGAGCGTCAGG TCCCGTTTTCCTCTCCGGCGTCTCCAATACAAAGATTACGGTGCAGAAGGAAATTGCACTCGTCTC (SEQ ID NO: 13).> Bisulfite Amplicon Sequence with primer:GTTTTTGGTTGAGGAGGAGGAGAAGGAGGAATCGCGTTAGGCGGAGCGTTAGGTT TCGTTTTTTTTTTCGGCGTTTTTAATATAAAGATTACGGTGTAGAAGGAAATTGTATTCGTTTT (SEQ ID NO: 14).> Primers:Forward: GTTTTTGGTTGAGGAGGAG (SEQ ID NO: 5).Reverse: AAAACRAATACAATTTCCTTCTACAC (degenerated primer) (SEQ ID NO: 6).> Fragment of 119 bp with 8 CpGs. Fragment of interest: chr17:30971079-30971222Gene: RNF135 > Genomic Amplicon Sequence: GGGCCTGGGCCTGGGCTCCGCCGTTCCCGTGTGGCTGGCCGAGGACGACCTCGG CTGCATCATCTGCCAGGGGCTGCTGGACTGGCCCGCCACGCTGCCCTGCGGCCACAGCTTCTGCCGCCACTGCCTGGAGGCCCTGTGGGG (SEQ ID NO: 15). > Bisulfite Amplicon Sequence with primer: GGGTTTGGGTTTGGGTTTCGTCGTTTTCGTGTGGTTGGTCGAGGACGATTTCGGTT GTATTATTTGTTAGGGGTTGTTGGATTGGTTCGTTACGTTGTTTTGCGGTTATAGTTTTTGTCGTTATTGTTTGGAGGTTTTGTGGGG (SEQ ID NO: 16).> Primers:Forward: GGGTTTGGGTTTGGGTTT (SEQ ID NO: 7).Reverse: CCCCACAAAACCTCCAAACAA (SEQ ID NO: 8).> Fragment of 144 bp with 10 CpGs. Example 2. RESULTSExample 2.1. Screening process to identify methylation markersAs a first step, a specific protocol to identify the differential methylation status of CpGs in theDNA of HCC subjects including cfDNA (Figure 1 and Figure 2) was implemented.Particularly, to in silico identify putative blood-based HCC DNA methylation markers,available tissue methylation data on HCC were used and control samples from 3 different studies (GSE56588, GSE54503 and TCGA-LIHC), all carried out using the Illumina's Infinium HumanMethylation450 (HM450) BeadChip array (that interrogates more than 485,000 CpG sites covering 96% of known CpG islands (CGIs).A global hypermethylation enrichment was found in CpGs located within CGI in HCC samplescompared to controls (p-value=4.31E-28), being able to segregate HCC and controls into twodistinct clusters. Focusing on CGIs, a list of 3862 hypermethylated CpGs was extractedbetween HCC and control samples (median methylation difference >0.2 and a FDR <0.05; Student's two-sided T-test and Benjamini-Hochberg False Discovery Rate for p-value correction). As the final goal was to identify blood-based DNA methylation candidates, and leukocyte-derived DNA is the most abundant contaminant of cell-free DNA (cfDNA) in the blood, the levels of methylation of the 3862 selected CpGs was determined in blood leukocyte samples of healthy individuals (656 samples, GSE40279). After eliminating all CpGs with high DNA methylation levels (β-value>0.2) in at least 1 of the blood leukocyte samples analyzed, a final list of 1061 CpGs was retained as the putative candidates for HCC detection using blood as liquid biopsy.Among this extensive candidate list, it was selected: (1) Those CpGs not methylated (β-value<0.2) in all the 135 non-tumoral samples (including healthy, cirrhotic and normal tissue adjacent to the tumour) and with a methylation difference higher than 0.3, resulting in 27 CpGs, and (2) those CpGs not methylated (β-value<0.2) in all the healthy and cirrhotic subjects and with a methylation difference higher than 0.4, resulting in 12 CpGs. This panel of 39 CpGs revealed a robust performance (sensitivity 98.2%; specificity 94.8%;Figure 3) for control and HCC subject classification in silico.To establish an assay to measure the methylation status of this panel of CpGs, the primers forthe amplification were designed, after bisulfite treatment, of the regions containing each of the 39 CpGs of interest and tried to set up the conditions of each PCR. For that, it was taken into account that the cfDNA is highly fragmented, showing a predominant mononucleosome peak of ~167 bp, therefore PCR products length should not be larger than that size. It was also considered that the 39 CpGs were within CpG islands and that they were therefore surrounded by other CpGs for which we did not know the methylation status, they were avoided, or degenerate primers were designed. As a result, some fragments amplified efficiently and consistently, while others did not. Then, the empirical specificity of thosefragments that were successfully amplified was explored, using cfDNA obtained from a panelof sera of control donors. Therefore, the panel was reduced, excluding: 1. Those CpGs for which primers could not be designed2. Those CpGs for which several couple of primers tested did not amplify in cfDNA3. Those CpGs that were methylated in any of the tested controls, to avoid false positives4. Those CpGs that, even though were amplified correctly, were not methylated in any ofthe tested HCCs. Note: It is true that, the testing cohort had a reduced number of HCC , therefore those CpGs could have been methylated in other HCC subjects; however, we decided to sacrifice them in order to reduce the number of biomarkers to a more technically manageable number. 5. In order to reduce the panel to a minimum number of markers we also eliminated thosethat were empirically redundant (positive in the same group of HCC subjects).Thus, a final candidate list (Table 2) was obtained, called HepaMeth panel, with four CpGswith the highest individual empirical sensitivity and specificity. Table 2 When the performance of HepaMeth Panel was interrogated in the 135 controls and 670 HCC tissues used for the identification of HepaMeth Panel (Figure 1; GSE56588, GSE54503 andTCGA-LIHC), it was observed an in silico sensitivity of 92% and a specificity of 98.5%(Figure 4).Example 2.2. In silico validation of HepaMeth panel in other available tissue methylomedata The performance of HepaMeth Panel was then evaluated in three different and independent available methylomes with control and HCC tissues (GSE60753, GSE89852 and GSE157341). HepaMeth Panel showed a sensitivity of 87.5%, 100% and 94.4% and a specificity of 96.4%, 97.3% and 94.3%, respectively (Figure 5).Example 2.3. Validation of HepaMeth in control and HCC tissuesThe in silico findings were validated in 19 HCC tissue. After isolating the genomic DNA, itwas bisulfite converted and PCRs were carried out to amplify the regions of interest. PCR products were sequenced by Sanger. All tissue samples were positive (methylated) for at least three of the 4 markers, resulting in a sensitivity of 100% (Figure 6).Example 2.4. The methylation status of all the CpGs presented in each DNA fragmentamplified by PCR enhances the diagnostic performance of the 4 individual CpGs The validation process revealed that not only the individual CpG of interest selected in silico, but all the CpGs within each of the amplified PCR fragments were informative. Therefore, considering the methylation status of all the CpGs presented in each PCR significantly (Fisher's exact test 0.035) enhance the performance of the panel (Figure 7). Example 2.5. Evaluation of the diagnostic capacity of HepaMeth panel in plasma cfDNA compared to the gold-standard alpha-fetoprotein *Note: The serological detection of AFP remains the most commonly used parameter for HCC diagnosis, often in combination with imaging approaches. AFP levels superior to 20 ng / mL show good sensitivity but low specificity, whereas at higher cut-offs of 200 ng / mL the sensitivity drops to 22% with high specificity. In order to have an objective value of the diagnostic performance of HepaMeth we have chosen a value of 15 ng / mL of AFP as positive for comparison. This is the cutoff of AFP considered at CUN.Using plasma samples from a cohort of 78 HCC subjects, AFP was higher than 15 ng / mL in46.2% of subjects whereas HepaMeth was positive in 79.5%. Statistical analyses revealed that HepaMeth significantly outperformed AFP in this cohort of HCC subjects (Figure 8).Example 2.6. Evaluation of HepaMeth performance in plasma cfDNA for the earlydiagnostic of HCC in comparison with the gold-standard alpha-fetoproteinAmong the 78 HCC subjects evaluated below, 30 subjects were classified according to the BCLC index as BCLC 0 / A or early HCCs. HepaMeth Panel was positive for 22 of these 30 HCC BCLC 0 / A subjects (73%), whereas AFP was higher than 15 ng / mL in 5 of the 30 subjects (16.7%). Thus, HepaMeth significantly (p<0,0001) outperformed AFP (>15 ng / mL) for the early diagnosis of HCC in this cohort of HCC subjects (Figure 9). Example 2.7. Evaluation of the prognostic capacity of HepaMeth panel in plasma cfDNAThe potential of HepaMeth Panel as a prognosis marker was also explored using the samecohort of HCC subjects used in sections 10 (n=78). We observed that subjects with a positive HepaMeth Panel have significantly worse survival than subjects with negative HepaMeth Panel (Figure 10). Example 2.8. Evaluation of the capacity of HepaMeth panel to predict the response to inmunotherapy using plasma cfDNAFor this purpose, the plasma samples obtained at baseline of a cohort of 42 HCC subjectsincluded in a clinical trial conducted in our hospital (CUN) were used, involving the treatmentof subjects with immunotherapy (Nivolumab). It was observed that the positivity at baselinefor at least two of the HepaMeth markers significantly predicts poorer response toimmunotherapy treatment (Figure 11).Example 3. Sensitivity of the assayed biomarkers for the diagnosis of liver cancerSensitivity of each marker and their combinations were assayed, and the results are shown inTable 3. Specifically, Table 3 shows the sensitivity of individual biomarkers, or combinationsof two, three or four biomarkers (n=117 HCCs). Table 3 Example 4. Prognostic value of the assayed biomarkers in liver cancerThe prognostic value of each biomarker was assayed. The prognostic value of MIXL1 isstatistically significant (p = 0.004) (Figure 12), the prognostic value of HIST3H2A isstatistically significant (p = 0.04) (Figure 13), the prognostic value of TSC22D1 is notstatistically significant (p = 0.09) (Figure 14) and the prognostic value of RNF135 isstatistically significant (p = 0.04) (Figure 15).Example 5. Capacity of the assayed biomarkers to predict response of subjects sufferingfrom liver cancer to immunotherapy (anti-PD1 or anti-PDL1 immunotherapy)The capacity of the biomarkers to predict response to immunotherapy was assayed. The results are shown in Table 4. Table 4 Of note, HIST3H2A and RNF135 biomarkers are statistically significant to predict response ofsubjects suffering from liver cancer to immunotherapy (anti-PD1 or anti-PDL1 immunotherapy).

Claims

CLAIMS 1. In vitro method for screening, diagnosis and / or prognosis of liver cancer, whichcomprises determining the methylation status of the gene HIST3H2A in a biologicalsample obtained from the subject, wherein a higher level of methylation of the geneHIST3H2A, as compared with a reference level of methylation of the gene HIST3H2A measured in healthy control subjects, is an indication that the subject is at risk of suffering from liver cancer, is suffering from liver cancer and / or has a poor prognosis.

2. In vitro method, according to claim 1, which comprises determining themethylation status of the gene HIST3H2A in combination with at least a gene selectedfrom: MIXL1, TSC22D1 and / or RNF135, wherein a higher level of methylation of anyof these genes, as compared with a reference level of methylation measured in healthy control subjects, is an indication that the subject is at risk of suffering from liver cancer, is suffering from liver cancer and / or has a poor prognosis.

3. In vitro method, according to any of the previous claims, which comprisesdetermining the methylation status of the following combination of genes: MIXL1, HIST3H2A, TSC22D1 and RNF135, wherein a higher level of methylation of any ofthese genes, as compared with a reference level of methylation measured in healthy control subjects, is an indication that the subject is at risk of suffering from liver cancer, is suffering from liver cancer and / or has a poor prognosis.

4. In vitro method, according to any of the previous claims, wherein themethylation status of the gene is determined in at least a CpG site of the gene; preferablya CpG site of the promoter region.

5. In vitro method, according to any of the previous claims, wherein themethylation status of the gene MIXL1 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr1:226223965- 226224144, wherein the methylation status of the gene HIST3H2A is determined in at least a CpG site of the promoter region located between the chromosomal positions chr1:228457730-228457883, wherein the methylation status of the gene TSC22D1 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr13:44576100-44576218, and / or wherein the methylation status of the gene RNF135 is determined in at least a CpG site of the promoter regionlocated between the chromosomal positions chr17: 30971079-30971222 according tothe human reference genome hg38.

6. In vitro use of the methylation status of the gene HIST3H2A, or of a kitcomprising reagents for determining the methylation status of the gene HIST3H2A, forscreening, diagnosis and / or prognosis of liver cancer.

7. In vitro use, according to claim 6, of the methylation status of the geneHIST3H2A in combination with the methylation status of at least a gene selected from:MIXL1, TSC22D1 and / or RNF135, or of a kit comprising reagents for determining themethylation status of the gene HIST3H2A in combination with the methylation statusof at least a gene selected from: MIXL1, TSC22D1 and / or RNF135, for screening,diagnosis and / or prognosis of liver cancer.

8. In vitro use, according to any of the claims 6 or 7, of the methylation status ofthe following combination of genes: MIXL1, HIST3H2A, TSC22D1 and RNF135, or ofa kit comprising reagents for determining the methylation status of the followingcombination of genes: MIXL1, HIST3H2A, TSC22D1 and RNF135, for screening,diagnosis and / or prognosis of liver cancer.

9. In vitro use, according to any of the claims 6 to 8, wherein the methylation statusof the gene is determined in at least a CpG site; preferably a CpG site located in thepromoter region.

10. In vitro use, according to any of the claims 6 to 9, wherein the methylation statusof the gene MIXL1 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr1:226223965-226224144, wherein the methylation status of the gene HIST3H2A is determined in at least a CpG site of the promoter region located between the chromosomal positions chr1:228457730- 228457883, wherein the methylation status of the gene TSC22D1 is determined in at least a CpG site of the promoter region located between the chromosomal positions chr13:44576100-44576218, and / or wherein the methylation status of the gene RNF135 is determined in at least a CpG site of the promoter region located between thechromosomal positions chr17: 30971079-30971222 according to the human referencegenome hg38.

11. Kit, suitable for screening, diagnosis and / or prognosis of liver cancer, whichcomprises: a. A pair of primers for the amplification of the fragment chr1:228457730-228457883 of gene HIST3H2A, wherein at least one primer, preferably both, is selected from the forward primer GGTTTAYGGGGAATTGTAGTTT (SEQ ID NO: 3) and the reverse primer CTTCACTACCCTCTTATTTTTAATCT (SEQ ID NO: 4).

12. Kit, suitable for screening, diagnosis and / or prognosis of liver cancer, accordingto claim 11, which further comprises at least one pair of primers selected from: b. A pair of primers for the amplification of the fragment chr1:226223965-226224144 of gene MIXL1, wherein at least one primer, preferably both, is selected from the forward primer TYGAATAGTTGTAGTTGTTGGAGTT (SEQ ID NO: 1) and the reverse primer CCRCAATCCAAACCTAAAATCC (SEQ ID NO: 2); c. A pair of primers for the amplification of the fragment chr13:44576100-44576218 of gene TSC22D1, wherein at least one primer, preferably both, isselected from the forward primer GTTTTTGGTTGAGGAGGAG (SEQ IDNO: 5) and the reverse primer AAAACRAATACAATTTCCTTCTACAC(SEQ ID NO: 6); and / or d. A pair of primers for the amplification of the fragment chr17:30971079-30971222 of gene RNF135, wherein at least one primer, preferably both, is selected from the forward primer GGGTTTGGGTTTGGGTTT (SEQ ID NO:7) and reverse primer CCCCACAAAACCTCCAAACAA (SEQ ID NO: 8).

13. Kit, suitable for screening, diagnosis and / or prognosis of liver cancer, accordingto claims 11 or 12 which comprises: a. A pair of primers for the amplification of the fragment chr1:228457730-228457883 of gene HIST3H2A, wherein at least one primer, preferably both, is selected from the forward primer GGTTTAYGGGGAATTGTAGTTT (SEQ ID NO: 3) and the reverse primer CTTCACTACCCTCTTATTTTTAATCT (SEQ ID NO: 4),b. A pair of primers for the amplification of the fragment chr1:226223965-226224144 of gene MIXL1, wherein at least one primer, preferably both, is selected from the forward primer TYGAATAGTTGTAGTTGTTGGAGTT (SEQ ID NO: 1) and the reverse primer CCRCAATCCAAACCTAAAATCC (SEQ ID NO: 2), c. A pair of primers for the amplification of the fragment chr13:44576100-44576218 of gene TSC22D1, wherein at least one primer, preferably both, is selected from the forward primer GTTTTTGGTTGAGGAGGAG (SEQ IDNO: 5) and the reverse primer AAAACRAATACAATTTCCTTCTACAC(SEQ ID NO: 6); and d. A pair of primers for the amplification of the fragment chr17:30971079-30971222 of gene RNF135, wherein at least one primer, preferably both, is selected from the forward primer GGGTTTGGGTTTGGGTTT (SEQ ID NO: 7) and reverse primer CCCCACAAAACCTCCAAACAA (SEQ ID NO: 8).

14. In vitro method for predicting the response of subjects suffering from livercancer to a treatment with anti-PD1 or anti-PDL1 immunotherapy which comprisesdetermining the methylation status of the gene HIST3H2A in a biological sampleobtained from the subject, wherein a higher level of methylation, as compared with a reference level of methylation of the gene measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.

15. In vitro method, according to claim 14, for predicting the response of subjectssuffering from liver cancer to a treatment with anti-PD1 or anti-PDL1 immunotherapywhich comprises determining the methylation status of the gene HIST3H2A incombination with at least a gene selected from: MIXL1, TSC22D1 and / or RNF135 in abiological sample obtained from the subject, wherein a higher level of methylation, as compared with a reference level of methylation of the gene measured in healthy controlsubjects, is an indication that the subject will have a worse response to the treatment.

16. In vitro method, according to any of the claims 14 or 15, for predicting theresponse of subjects suffering from liver cancer to a treatment with anti-PD1 or anti-PDL1 immunotherapy which comprises determining the methylation status of the genesMIXL1, HIST3H2A, TSC22D1 and RNF135 in a biological sample obtained from thesubject, wherein a higher level of methylation of at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects,is an indication that the subject will have a worse response to the treatment.

17. In vitro method for selecting subjects suffering from liver cancer for a treatmentwith anti-PD1 or anti-PDL1 immunotherapy which comprises determining themethylation status of the genes HIST3H2A in a biological sample obtained from thesubject, wherein a higher level of methylation of the gene, as compared with a reference level of methylation measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.

18. In vitro method, according to claim 17, for selecting subjects suffering fromliver cancer for a treatment with anti-PD1 or anti-PDL1 immunotherapy whichcomprises determining the methylation status of the genes HIST3H2A in combinationwith at least a gene selected from: MIXL1, TSC22D1 and / or RNF135, in a biologicalsample obtained from the subject, wherein a higher level of methylation of the genes, as compared with a reference level of methylation measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.

19. In vitro method, according to any of the claims 17 or 18, for selecting subjectssuffering from liver cancer for a treatment with anti-PD1 or anti-PDL1 immunotherapywhich comprises determining the methylation status of the genes MIXL1, HIST3H2A,TSC22D1 and RNF135 in a biological sample obtained from the subject, wherein ahigher level of methylation of at least two of the genes, as compared with a reference level of methylation of the genes measured in healthy control subjects, is an indication that the subject will have a worse response to the treatment.

20. Anti PD1 and / or PD-L1 antibody for use in a method for the treatment of livercancer, wherein the method comprises selecting subjects suffering from liver cancerwho are not classified as having a worse response to a treatment with anti PD1 and / oranti PD-L1 immunotherapy by implementing the method of claim 17.

21. Anti PD1 and / or PD-L1 antibody for use, according to claim 20, in a methodfor the treatment of liver cancer, wherein the method comprises selecting subjects suffering from liver cancer who are not classified as having a worse response to a treatment with anti PD1 and / or anti PD-L1 immunotherapy by implementing the method of claim 18.

22. Anti PD1 and / or PD-L1 antibody for use, according to any of the claims 20 or21, in a method for the treatment of liver cancer, wherein the method comprisesselecting subjects suffering from liver cancer who are not classified as having a worse response to a treatment with anti PD1 and / or anti PD-L1 immunotherapy by implementing the method of claim 19.

Images