Genetic signature of lung cancer

Abstract

The present invention relates to a method which allows for determining if the lung nodules of a patient are benign or cancerous. Furthermore the method also allows to determine if a patient is at high risk of developing lung cancer and if it should be subject to medical intervention.

Description

[0001] GENETIC SIGNATURE OF LUNG CANCER

[0002] TECHNICAL FIELD OF THE INVENTION

[0003] The present invention relates to the field of medicine, more in particular, to methods to diagnose and predict the evolution of lung cancer in a subject.

[0004] BACKGROUND OF THE INVENTION

[0005] Cancer caused an estimated 10 million deaths worldwide in 2020. This high mortality rate is largely a consequence of the lack of early detection tools for many tumors, which means that most are diagnosed in non-surgical stages, with a significantly worse prognosis. Although new therapies have been developed for some neoplasms in recent years, survival rates in advanced stages remain low. In the case of non-small cell lung cancer (NSCLC), 75-85% of cases are diagnosed in metastatic stages IIIB-IV, which have a 5-year survival rate of 17-20%. In contrast, patients diagnosed in early stages (I- IIIA) can undergo surgery with significantly better survival rates of 80-90% at 5 years.

[0006] Diagnosis of early-stage lung cancer is rare and often discovered incidentally when studying other diseases, because the symptoms in the early stages are very nonspecific. These cases may be asymptomatic, leading to a delay in diagnosis. Imaging technologies often detect pulmonary nodules of unknown significance, and as a result, some patients with benign nodules undergo unnecessary surgery, while others with small tumors are kept under observation, depriving them of potentially beneficial early treatment.

[0007] Both primary tumors and metastases, if they exist, release materials into the blood and other fluids (pleural, ascitic, cerebrospinal fluids, etc.), which can be used as "liquid biopsy". The information generated by “liquid” biopsies complements tissue biopsies, which are currently irreplaceable sources of information that cannot be obtained by any other means, such as tumor type and histology. Still, blood and other "liquid biopsies" have some advantages. Thus, it can be obtained repeatedly without risk and used to monitor the course of the disease, including early detection of response and relapse or the appearance of resistance to a particular therapy.

[0008] A gene expression signature would be particularly useful to guide the clinical decision in patients with pulmonary nodules suspected of malignancy, directing towards surgery or observation.

[0009] SUMMARY OF THE INVENTION The inventors have determined a gene signature which allows to differentiate between benign and cancerous lung nodules present in a patient with an improved accuracy.

[0010] Therefore, a first aspect of the present invention relates to a method for determining whether lung nodules in a patient are benign or cancerous comprising:

[0011] (i) determining the expression level of the MS4A1 gene in a sample from the subject, and

[0012] (ii) comparing the level obtained in (i) to a reference value, wherein an expression level of the MS4A1 gene that is decreased with respect to the reference value is indicative that the lung nodules are cancerous nodules.

[0013] Another aspect of the present invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy comprising:

[0014] (i) determining the expression level of the MS4A1 gene in a sample from the subject, and

[0015] (ii) comparing the level obtained in (i) to a reference value, wherein an expression level of the MS4A1 gene that is decreased with respect to the reference value is indicative that the subject is selected for a lung cancer therapy

[0016] A further aspect of the present invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0017] (i) determining the expression level of the MS4A1 gene in a sample from the subject, and

[0018] (ii) comparing the level obtained in (i) to a reference value,

[0019] (iii) treating the subject with the therapy if the subject is identified as having decreased expression level of the MS4A1 gene with respect to the reference value.

[0020] A further aspect of the present invention relates to a method for determining whether lung nodules in a patient are benign or cancerous comprising:

[0021] (i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,

[0022] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and

[0023] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii).

[0024] Yet another aspect of the present invention relates to a method for selecting a subject having one or more lung nodules for a lung cancer therapy comprising:

[0025] (i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,

[0026] (ii) determining whether the dataset obtained in step (i) is indicative of the patient being selected for lung cancer therapy by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as being selected for lung cancer therapy or not selected for lung cancer therapy and

[0027] (iii) selecting the patient for lung cancer therapy if the expression dataset obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were selected for lung cancer therapy in step (ii).

[0028] Another aspect of the present invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0029] (i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,

[0030] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules;

[0031] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii) and

[0032] (iv) treating the subject with the therapy if the patient is identified as having cancerous lung nodules.

[0033] Yet another aspect of the present invention relates to a method for determining whether lung nodules in a patient are benign or are cancerous comprising:

[0034] (i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient, and

[0035] (ii) comparing the level obtained in (i) to a reference value, wherein an increased level of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values is indicative that the lung nodules are cancerous nodules.

[0036] A further aspect of the invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy comprising:

[0037] (i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient, and

[0038] (ii) comparing the level obtained in (i) to a reference value, wherein an increased level of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values is indicative that the subject is selected for lung cancer therapy.

[0039] Yet another aspect of the present invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0040] (i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient, and

[0041] (ii) comparing the level obtained in (i) to a reference value,

[0042] (iii) treating the subject with the therapy if the patient is identified as having increased levels of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values. Another aspect of the present invention relates to a method for determining whether lung nodules in a patient are benign or are cancerous comprising:

[0043] (i) obtaining a dataset by determining the level of the hsa_circ_0000994, hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_circ_0004062, hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,

[0044] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and

[0045] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii),

[0046] In yet another aspect, the invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy comprising:

[0047] (i) obtaining a dataset by determining the level of the hsa_circ_0000994, hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458 hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256 hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809 hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,

[0048] (ii) determining whether the dataset obtained in step (i) is indicative of the patient being selected for lung cancer therapy by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as being selected for lung cancer therapy or not selected for lung cancer therapy and

[0049] (iii) selecting the patient for lung cancer therapy if the expression dataset obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were selected for lung cancer therapy in step (ii),

[0050] In yet another aspect, the invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0051] (i) obtaining a dataset by determining the level of the hsa_circ_0000994, hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_circ_0004062, hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,

[0052] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and

[0053] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii) and

[0054] (iv) treating the subject with the therapy if the patient is identified as having cancerous lung nodules.

[0055] Another aspect of the present invention relates to a kit comprising specific reagents for carrying out the method according to the invention.

[0056] A further aspect of the present invention relates to a computer system comprising one or more programs, wherein the one or more programs are include instructions for performing the method according to the invention.

[0057] A final aspect of the present invention relates to a computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to perform the method according to the invention.

[0058] BRIEF DESCRIPTION OF THE FIGURES Figure 1 : Schematic representation of the workflow used for the determination of the lung cancer signatures.

[0059] Figure 2: Feature selection pipeline.

[0060] Figure 3: Classification model pipeline.

[0061] Figure 4: Volcano Plot. Differential expression of mRNA. Volcano plot representing the Log2(FC) and nominal -LoglO(p-values) of all transcripts included in the 10360 panel for Cancer vs. healthy and healthy donor. Total of 750 features

[0062] Figure 5: Volcano Plot. Differential expression of circRNA. Volcano plot representing the Log2(FC) and nominal -LoglO(p-values) of all transcripts included in a nCounter custom-made panel for Cancer vs. healthy and healthy donor. Total of 78 features Figure 6: Expression levels of mRNA for DDB2 and MS4A1.

[0063] Figure 7: Expression levels of mRNA for hsa_circ_0000994 and hsa_circ_0012673.

[0064] DETAILED DESCRIPTION OF THE INVENTION

[0065] The inventors have identified transcriptomic signatures based on the levels of different mRNA and circRNA in biofluids that allow the determination in a patient showing one or more lung nodules whether the nodule is benign or cancerous.

[0066] Method for determining cancerous lung nodules

[0067] In a first aspect of the present invention relates to a method (hereinafter “Diagnostic method of the invention) for determining whether lung nodules in a patient are benign or are cancerous comprising:

[0068] (i) determining the expression level of MS4A1 in a sample from the subject, and

[0069] (ii) comparing the level obtained in (i) to a reference value, wherein an expression level of MS4A1 that is decreased with respect to the reference value is indicative that the lung nodules are cancerous nodules.

[0070] The diagnostic method of the invention can also be carried out by:

[0071] (i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673circular RNAs in a sample from the patient, and

[0072] (ii) comparing the level obtained in (i) to a reference value, wherein an increased level of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values is indicative that the lung nodules are cancerous nodules. The term “lung nodule” as used herein is interchangeable with “pulmonary lesion”. A lung nodule refers to abnormal growths, small lumps, swelling or collection of tissue in the lung of a patient. Around 95% of lung nodules are benign and are due to previous infections or caring of the lung tissue. Despite this, in 5% of the cases the lung nodules are cancerous. The skilled person will be aware of common techniques which are used to detect lung nodules such as X-Ray and / or chest computerized tomography scans (CT scans). While lung nodules detected by X-Ray are usually larger than 0.5 cm, CT scans can detect lung nodules which are smaller than 6 mm. In a particular embodiment of the method I of the invention the lung nodules are detected by chest X-ray and / or chest CT scan. A pulmonary nodule is any nodule less than or equal to three centimeters in diameter. In some embodiments, a pulmonary nodule has a diameter of about 0.8 cm to 2 cm.

[0073] The present diagnostic method of the invention permits the detection of lung nodules which are cancerous nodules. The term “cancerous nodules” as used herein refers to those nodules affected by or showing abnormalities characteristic of lung cancer.

[0074] The term “lung cancer”, as used herein, refers to a neoplasm, e.g., a malignant neoplasm, of the lung of the patient, wherein the neoplasm is of epithelial origin (i.e., carcinoma of the lung). Lung carcinomas are categorized by the size and appearance of the malignant cells and the term “lung cancer” includes both non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). NSCLC is any type of epithelial lung cancer other than SCLC. The most common types of NSCLC are squamous cell carcinoma, large cell carcinoma, and adenocarcinoma, but there are several other types that occur less frequently, and all types can occur in unusual histological variants. In some embodiments, the diagnostic method of the invention allows the determination of the tumor TNM stage (e.g., stage IA, stage IB, stage HA, and / or the like), tumor grade (e.g., well-differentiated, moderate, poorly-differentiated, undifferentiated, and / or the like), a tumor subtype (e.g., adenocarcinoma, squamous cell carcinoma, small cell carcinoma, and / or the like), a tumor histology (e.g., invasive carcinoma, metastasis, carcinoma in situ, and / or the like), and / or the like.

[0075] In a particular embodiment of the diagnostic method of the invention, the method allows the identification of the nodules as nodules from non-small cell lung cancer (NSCLC).

[0076] The term “patient” or “subject” refers to all animals classified as mammals and includes but is not limited to domestic and farm animals, primates and humans, for example, human beings, non-human primates, cows, horses, pigs, sheep, goats, dogs, cats, or rodents. Preferably, the subject is a human man or woman of any age or race. More preferably, the subject is a human adult, i.e having 18 years or more.

[0077] In some embodiments, the patient has been previously identified as carrying lung nodules by imaging techniques. As used herein, “imaging” may refer to an image (e.g., an X-ray, a CT scan, and / or the like) obtained by one or more radiography techniques, such as X-ray imaging, CT imaging (e.g., non-contrast CT imaging, contrast CT imaging, and / or the like), magnetic resonance imaging (MRI), positron-emission tomography (PET) imaging, and / or the like. In some implementations, the imaging described herein may be chest imaging that depicts an area below a neck of a patient and above a bowel of the patient. In such a case, the chest imaging may depict a cardiovascular system and / or a pulmonary system of the patient. For example, the chest imaging may depict lungs of the patient.

[0078] Step (i) of the diagnostic method

[0079] The first step of the diagnostic method of the invention relates to the determination of the expression levels of the MS4A1 gene or to the determination of the levels of the hsa_circ_0000994 and / or the hsa_circ_0012673 circular RNAs in a sample from the patient.

[0080] The expression “determination of the expression levels of the MS4A1 gene” refers to the determination of any value which provides an indication on the overall expression the gene, which includes either the mRNA levels or the levels of the protein encoded by the MS4A1 gene, when such expression is measured quantitatively, based on methods as disclosed herein and / or routinely known to the skilled person. The term "expression level”, as used herein, refers to a measurable quantity of a gene product produced by the gene in a sample of the subject, wherein the gene product can be a transcriptional product or a translational product. As understood by the person skilled in the art, the gene expression level can be quantified by measuring the levels of the transcriptional product of said gene (messenger RNA when the gene is a protein-coding gene) or of the protein encoded by said gene. In a particular embodiment of the method I of the invention the determination of the expression level of the gene is carried out by determining the levels of RNA transcripts of the gene(s).

[0081] The level of a given RNA transcript can be determined by methods well known in the art. For example the nucleic acid contained in the sample is first extracted according to standard methods, for example using lytic enzymes or chemical solutions or extracted by nucleic-acid-binding resins following the manufacturer's instructions. The extracted RNA is then detected by hybridization (e.g., Northern blot analysis, y oligonucleotide microarrays after converting the mRNA into a labeled cDNA or by methods based on the binding of specific probes labelled with colour-coded molecular barcodes to the biomarker of interest such as nCounter.) and / or amplification (e.g., RT-PCR). Quantitative or semi-quantitative RT-PCR is preferred. Real-time quantitative or semiquantitative RT-PCR is particularly advantageous. Preferably, primer pairs are designed in order to overlap an intron, so as to distinguish cDNA amplification from putative genomic contamination. Suitable primers may be easily designed by the skilled person. Other methods of amplification include ligase chain reaction (LCR), transcription- mediated amplification (TMA), strand displacement amplification (SDA) and nucleic acid sequence based amplification (NASBA).

[0082] Alternatively, the first step of the diagnostic method of the invention comprises the determination of the levels of the hsa circ 0000994 and / or hsa circ 0012673 in the sample from the patient.

[0083] In an embodiment, this method further comprises determining the levels of at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256 hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in the sample.

[0084] As used herein, the terms “circular RNA” or “circular polyribonucleotide” or “circRNA” or “oRNA” are used interchangeably and refers to a polyribonucleotide that forms a circular structure through covalent bonds. Preferably, the circular RNA or RNAs to be used in the methods of the present invention is single stranded, thereby enabling translation of the RNA.

[0085] Methods for the detection or nucleic acid molecules such as circRNA in a sample comprise PCR based methods which involve the target specific amplification of nucleic acids. However, any nucleic acid detection assay known to the skilled artisan may be used to detect one or more circRNA of the invention. In a preferred embodiment, the level of circRNA transcripts can be determined by using the rolling circle amplification. As used herein, the term “nucleic acid detection assay” refers to any method of determining the nucleotide composition of a nucleic acid of interest. Nucleic acid detection assay include but are not limited to, sequencing methods, probe hybridization methods, structure specific cleavage assays (e.g., the INVADER assay, (Hologic, Inc); enzyme mismatch cleavage methods; polymerase chain reaction (PCR); branched hybridization methods; rolling circle replication; NASBA; molecular beacon technology; E-sensor technology (Motorola, U.S. Pat. No. 6,248,229); cycling probe technology; Dade Behring signal amplification methods; ligase chain reaction; and sandwich hybridization methods.

[0086] The methods of the invention may preferably include a further step of removing ribosomal and / or poly A RNA from the biological sample before determining the level of a circRNA of the invention. Thus, before determining the level of the circRNA of the invention, the biological sample or the isolated RNA therefrom is treated to remove any ribosomal and / or poly-A tail RNA (mRNA). Methods to remove rRNA or polyA RNA are well known in the art and are exemplary described in the example section of the present application. By removing any non circRNA from the sample, the results are more specific and sensitive due to a reduced RNA background “noise”. This step may be performed in any aspect or embodiment of the invention which requires the detection or isolation of a circRNA of the invention.

[0087] In a particular embodiment of the diagnostic method of the invention of the invention, the determination of the levels of the RNA transcripts and / or circular RNA is carried out by measuring the levels of circulating RNA. The term “circulating RNA” is used interchangeably with “circulating free “circulating cell-free RNA” and refers to RNA (mRNA or circRNA) which is found free from cells, primarily in the blood, plasma or serum in circulation in the subject’s body. Circulating free RNA is normally found encapsulated within extracellular membrane vesicles or they form ribonucleoprotein complexes.

[0088] MS4A1 protein refers to a B-lymphocyte-specific membrane protein that plays a role in the regulation of cellular calcium influx necessary for the development, differentiation, and activation of B-lymphocytes and is defined in the Ensembl database with accession number ENSG00000173334.4, release 109 of February 2023. In some embodiments, the methods of the invention comprises the determination of the levels of one or more of the ENST00000345732.9, the ENST00000534668.6, the ENST00000389939.2, the ENST00000532073.5. the ENST00000674194.1 , the

[0089] ENST00000528313.1 , the ENST00000533306.6, the ENST00000532491.5, the

[0090] ENST00000534503.5, the ENST00000527101 .5, the ENST00000532418.1 or the ENST00000530482.1 transcripts (accession numbers for the transcripts as provided in the Ensembl database release 110 of July 2023). The first step of the diagnostic method of the invention based on the determination of circRNA levels comprises the determination of the expression level of the hsa_circ_0000994 and / or of the hsa_circ_0012673 circRNA in a sample from the patient.

[0091] The term “hsa_circ_0012673” refers to the sequence according to SEQ ID NO: 2. The term “hsa_circ_0012673” refers to the sequence according to SEQ ID NO: 3.

[0092] It will be understood that the determination of the levels of the MS4A1 transcript or of the levels of at least one of the hsa_circ_0012673 and, hsa_circ_0000994 can be carried out by measuring the levels of a fragment of each of these transcripts. The term “fragment thereof” as used herein refers to a short sequence of RNA which allows the unambiguous identification of the gene in question. In a particular embodiment the fragment has a sequence length of between 100 and the full length minus 1 nucleotide of the polynucleotides defined above as SEQ ID NO: 1 to 3.

[0093] In the present invention, the term “sample” or “biological sample” means biological material isolated from a subject. The biological sample can contain any biological material suitable for detecting the expression levels of the desire genes, that is, the sample should contain mRNA and / or proteins or the levels of circRNA, preferably mRNA and / or proteins or circRNA from the lung nodules, and can comprise cell and / or non-cell material of the subject. In a particular embodiment, the sample is a tumor sample. In a particular embodiment of the method I of invention the sample is selected form blood, plasma or serum. In a more particular embodiment of the method I of the invention the sample is plasma.

[0094] In order to improve the accuracy of the diagnostic method of the invention based on the determination of the expression levels of the MS4A gene, the expression levels of further genes can be determined in step (i) of the method. In a particular embodiment of diagnostic method of the invention, step (i) further comprises determining the expression level of DDB2 in the sample from the patient, wherein an expression level of DDB2 that is increased with respect to the reference value is indicative that the lung nodules are cancerous nodules.

[0095] The term “DDB2” refers to the gene which encodes for the protein DNA damagebinding protein 2 which is involved in DNA repair and protein ubiquitination as part of the LIV-DDB complex and as part of the DCX (DDB1-CUL4-X-box) complex. The expression levels DDB2 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the DDB2 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000256996.9, ENST00000378603.7, ENST00000614825.4,

[0096] ENST00000378600.7, ENST00000622090.4, ENST00000622878.4,

[0097] ENST00000617847.4, ENST00000378601.7, ENST00000616278.4,

[0098] ENST00000610805.4, ENST00000620515.1 , ENST00000612309.4,

[0099] ENST00000617022.4, ENST00000614394.1 , ENST00000615695.1 and

[0100] ENST00000614884.1 in the Ensembl database under accession number ENSG00000134574 dated (Release 110). In a preferred embodiment, the expression levels of the DDB2 gene are determined by measuring the expression of the polypeptide defined under accession number Q92466 of the Uniprot database, entry version 28-Jun- 2023..

[0101] In a further particular embodiment of diagnostic method of the invention step (i) further comprises determining the expression level of at least one gene selected from group consisting of: BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 , wherein the alteration of the expression level of said at least one gene with respect to the reference value is indicative that the lung nodules are cancerous nodules.

[0102] The term “SPIB” refers to the gene which encodes for the protein “Transcription factor Spi-B” which is a sequence specific transcriptional activator which binds to the PU- boxthat can act as a lymphoid-specific enhancer. The expression levels SPIB gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the SPIB gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000595883.6, ENST00000439922.6, ENST00000270632.7, ENST00000596074.5,

[0103] ENST00000597855.5, ENST00000594685.1 , ENST00000599923.5 and

[0104] ENST00000594188.1 in the Ensembl database under accession number ENSG00000269404 dated (Release 110). In a preferred embodiment, the expression levels of the DDB2 gene are determined by measuring the expression of the polypeptide defined under accession number Q01892 of the Uniprot database, entry version 28-Jun- 2023..

[0105] The term “TAPBPL” refers to the gene which encodes for the protein “Tapasin- related protein” which is a component of the antigen processing and presentation pathway, which binds to MHC class I coupled with beta2-microglobulin / B2M. The expression levels TAPBL gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the TAPBL gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000266556.8, ENST00000544021.5, ENST00000545700.5,

[0106] ENST00000539384.5, ENST00000543567.5, ENST00000544289.1 ,

[0107] ENST00000544826.5 and ENST00000542160.1 in the Ensembl database under accession number ENSG00000139192 dated (Release 110).. In a preferred embodiment, the expression levels of the TAPBLP gene are determined by measuring the expression of the polypeptide defined under accession number Q9BX59 of the Uniprot database, entry version 28-Jun-2023.

[0108] The term “TLR8” refers to the gene which encodes for the protein “Toll-like receptor 8”. The expression levels TLR8 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the TLR8 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000218032.7 and ENST00000311912.5 in the Ensembl database under accession number ENSG00000101916 dated (Release 110). In a preferred embodiment, the expression levels of the TLR8 gene are determined by measuring the expression of the polypeptide defined under accession number Q9NR97 of the Uniprot database, entry version 28-Jun-2023.

[0109] The term “CXCL8” refers to the gene which encodes for the protein “lnterleukin-8” which is a chemotactic factor that mediates inflammatory response by attracting neutrophils, basophils, and T-cells to clear pathogens and protect the host from infection; also plays an important role in neutrophil activation. . The expression levels CXCL8 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the CXCL8 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000307407.8, ENST00000401931.1 , ENST00000696131.1 , ENST00000696132.1 and

[0110] ENST00000483500.1 in the Ensembl database under accession number ENSG00000169429 dated (Release 110). In a preferred embodiment, the expression levels of the CXCL8 gene are determined by measuring the expression of the polypeptide defined under accession number P10145 of the Uniprot database, entry version 28-Jun-2023. The term “KLRK1” refers to the gene which encodes for the protein “NKG2-D type II integral membrane protein” which functions as an activating and costimulatory receptor involved in immunosurveillance upon binding to various cellular stress-inducible ligands displayed at the surface of autologous tumor cells and virus-infected cells. The expression levels KLRK1 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the KLRK1 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000240618.11 , ENST00000540818.5, ENST00000396451.4,

[0111] ENST00000544449.5 and ENST00000540267.5 in the Ensembl database under accession number ENSG00000213809 dated (Release 110). In a preferred embodiment, the expression levels of the KLRK1 gene are determined by measuring the expression of the polypeptide defined under accession number P26718 of the Uniprot database, entry version 28-Jun-2023.

[0112] The term “BLK” refers to the gene which encodes for the protein “Tyrosine-protein kinase Blk”. The expression levels BLK gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the BLK gene are determined by determining the expression levels of the transcript defined under the accession numbers ENST00000259089.9 and ENST00000646615.3 in the Ensembl database under accession number ENSG00000136573 dated (Release 112). In a preferred embodiment, the expression levels of the BLK gene are determined by measuring the expression of the polypeptide defined under accession number P51451 of the Uniprot database, entry version 02-0ct-2024.

[0113] The term “NFKBIA” refers to the gene which encodes for the protein “NF-kappa-B inhibitor alpha” which inhibits the activity of dimeric NF-kappa-B / REL complexes by trapping REL (RELA / p65 and NFKB1 / p50) dimers in the cytoplasm by masking their nuclear localization signals. The expression levels of NFKBIA gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the NFKBIA gene are determined by determining the expression levels of the transcript defined under the accession number ENST00000216797.10 in the Ensembl database under accession number ENSG00000100906 dated (Release 112). In a preferred embodiment, the expression levels of NFKBIA gene are determined by measuring the expression of the polypeptide defined under accession number P25963 of the Uniprot database, entry version 02-0ct-2024

[0114] The term “KLRD1” refers to the gene which encodes for the protein “Natural killer cells antigen CD94” an immune receptor involved in self-nonself discrimination. In complex with KLRC1 or KLRC2 on cytotoxic and regulatory lymphocyte subsets, recognizes non-classical major histocompatibility (MHC) class lb molecule HLA-E loaded with self-peptides derived from the signal sequence of classical MHC class la and non- classical MHC class lb molecules. The expression levels of KLRD1 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the KLRD1 gene are determined by determining the expression levels of the transcript defined under the accession numbers ENST00000336164.9, ENST00000350274.9 and

[0115] ENST00000381908.7 in the Ensembl database under accession number ENSG00000134539 dated (Release 112). In a preferred embodiment, the expression levels of KLRD1 gene are determined by measuring the expression of the polypeptide defined under accession number Q13241 of the Uniprot database, entry version 02-Oct- 2024.

[0116] The term “PSMB9” refers to the gene which encodes for the protein “Proteasome subunit beta type-9” which is involved in antigen processing to generate class I binding peptides. Replacement of PSMB6 by PSMB9 increases the capacity of the immunoproteasome to cleave model peptides after hydrophobic and basic residues. The expression of PSMB9 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the PSMB9 gene are determined by the expression levels of the transcript defined under the accession numbers ENST00000374859.3, ENST00000383114.8, ENST00000383234.8, ENST00000422729.6 and

[0117] ENST00000427870.6 in the Ensembl database under accession number ENSG00000240065 dated (Release 112). In a preferred embodiment, the expression levels of PSMB9 gene are determined by measuring the expression of the polypeptide defined under accession number P28065 of the Uniprot database, entry version 02-Oct- 2024.

[0118] The term “S100A12” refers to the gene which encodes for the protein “S100-A12” which is a calcium-, zinc- and copper-binding protein which plays a prominent role in the regulation of inflammatory processes and immune response. The expression of S100A12 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the S100A12 gene are determined by determining the expression levels of the transcript defined under the accession number ENST00000368737.5 in the Ensembl database under accession number ENSG00000163221 dated (Release 112). In a preferred embodiment, the expression levels of S100A12 gene are determined by measuring the expression of the polypeptide defined under accession number P80511 of the Uniprot database, entry version 02-0ct-2024.

[0119] The term “TAPBP” refers to the gene which encodes for the protein “Tapasin” which is involved in the association of MHC class I with transporter associated with antigen processing (TAP) and in the assembly of MHC class I with peptide (peptide loading). The expression of TAPBP gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the TAPBP gene are determined by determining the expression levels of the transcript defined under the accession numbers ENST00000374572.7, ENST00000383065.8, ENST00000383066.8,

[0120] ENST00000383197.8 and ENST00000383198.6 in the Ensembl database under accession number ENSG00000112493 dated (Release 112). In a preferred embodiment, the expression levels of TAPBP gene are determined by measuring the expression of the polypeptide defined under accession number 015533 of the Uniprot database, entry version 02-0ct-2024.

[0121] The term “ARG1” refers to the gene which encodes for the protein “Arginase-1” which is a key element of the urea cycle converting L-arginine to urea and L-ornithine, which is further metabolized into metabolites proline and polyamides that drive collagen synthesis and bioenergetic pathways critical for cell proliferation, respectively; the urea cycle takes place primarily in the liver and, to a lesser extent, in the kidneys. The expression of ARG1 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the ARG1 gene are determined by determining the expression levels of the transcript defined under the accession numbers ENST00000356962.2 and ENST00000368087.8 in the Ensembl database under accession number ENSG00000118520 dated (Release 112). In a preferred embodiment, the expression levels of ARG1 gene are determined by measuring the expression of the polypeptide defined under accession number P05089 of the Uniprot database, entry version 02-Oct- 2024. The term “CD69” refers to the gene which encodes for the protein “Early activation antigen CD69” which is an early activation market that is expressed in hematopoietic stem cells, T cells, and many other cell types in the immune system. The expression of CD69 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the CD69 gene are determined by determining the expression levels of the transcript defined under accession number ENST00000228434.7 in the Ensembl database under accession number ENSG00000110848 dated (Release 112). In a preferred embodiment, the expression levels of CD69 gene are determined by measuring the expression of the polypeptide defined under accession number Q07108 of the Uniprot database, entry version 02-0ct-2024.

[0122] The term “PRKCA” refers to the gene which encodes for the protein “Protein kinase C alpha type” which is involved in positive and negative regulation of cell proliferation, apoptosis, differentiation, migration and adhesion, tumorigenesis, cardiac hypertrophy, angiogenesis, platelet function and inflammation, by directly phosphorylating targets such as RAF1 , BCL2, CSPG4, TNNT2 / CTNT, or activating signaling cascade involving MAPK1 / 3 (ERK1 / 2) and RAP1GAP. The expression levels PRKCA gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the PRKCA gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000413366.8 in the Ensembl database under accession number ENSG00000154229 dated (Release 112). In a preferred embodiment, the expression levels of the PRKCA gene are determined by measuring the expression of the polypeptide defined under accession number P17252 of the Uniprot database, entry version 02-0ct-2024.

[0123] The term “RPL23” refers to the gene which encodes for the protein “Large ribosomal subunit protein uL14” which is a component of the large ribosomal subunit. The expression levels RPL23 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the RPL23 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000394332.5 and ENST00000479035.7 in the Ensembl database under accession number ENSG00000125691 dated (Release 112). In a preferred embodiment, the expression levels of the RPL23 gene are determined by measuring the expression of the polypeptide defined under accession number P62829 of the Uniprot database, entry version 02-0ct-2024.

[0124] The term “CD48” refers to the gene which encodes for the protein “CD48 antigen” which is a member of the CD2 subfamily of the immunoglobulin superfamily (IgSF) which includes SLAM (signaling lymphocyte activation molecules) proteins, such as CD84, CD150, CD229 and CD244. CD48 is found on the surface of lymphocytes and other immune cells, dendritic cells and endothelial cells, and participates in activation and differentiation pathways in these cells. The expression levels CD48 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the CD48 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000368045.3 and ENST00000368046.8 in the Ensembl database under accession number ENSG00000117091 dated (Release 112). In a preferred embodiment, the expression levels of the CD48 gene are determined by measuring the expression of the polypeptide defined under accession number P09326 of the Uniprot database, entry version 02-Oct- 2024.

[0125] The term “CD8B” refers to the gene which encodes for the protein “T-cell surface glycoprotein CD8 beta chain” which plays an essential role in the immune response and serves multiple functions in responses against both external and internal offenses. The expression levels CD8B gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the CD8B gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000331469.6, ENST00000349455.7, ENST00000390655.12,

[0126] ENST00000393759.6 and ENST00000393761.6 in the Ensembl database under accession number ENSG00000172116 dated (Release 112). In a preferred embodiment, the expression levels of the CD8B gene are determined by measuring the expression of the polypeptide defined under accession number P10966 of the Uniprot database, entry version 02-0ct-2024.

[0127] The term “LYZ” refers to the gene which encodes for the protein “Lysozyme C” which is capable of both hydrolysis and transglycosylation; it shows also a slight esterase activity. It acts rapidly on both peptide-substituted and unsubstituted peptidoglycan, and slowly on chitin oligosaccharides. The expression levels LYZ gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the LYZ gene are determined by determining the expression levels of one or more of the transcripts defined under the accession number ENST00000261267.7 in the Ensembl database under accession number ENSG00000090382 dated (Release 112). In a preferred embodiment, the expression levels of the LYZ gene are determined by measuring the expression of the polypeptide defined under accession number P61626 of the Uniprot database, entry version 02-0ct-2024.

[0128] The term “IFITM1” refers to the gene which encodes for the protein “Interferon- induced transmembrane protein 1” which inhibits the entry of viruses to the host cell cytoplasm, permitting endocytosis, but preventing subsequent viral fusion and release of viral contents into the cytosol. The expression levels IFITMIgene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the IFITM1 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000328221.5, ENST00000408968.4,

[0129] ENST00000528780.5 and ENST00000680588.1 in the Ensembl database under accession number ENSG00000185885 dated (Release 112). In a preferred embodiment, the expression levels of the IFITM1 gene are determined by measuring the expression of the polypeptide defined under accession number P13164 of the Uniprot database, entry version 02-0ct-2024.

[0130] The term “ITGB3” refers to the gene which encodes for the protein “Integrin betas’’ which is a receptor for cytotactin, fibronectin, laminin, matrix metalloproteinase-2, osteopontin, osteomodulin, prothrombin, thrombospondin, vitronectin and von Willebrand factor. The expression levels ITGB3 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the ITGB3 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000559488.7 and ENST00000696963.1 in the Ensembl database under accession number ENSG00000259207 dated (Release 112). In a preferred embodiment, the expression levels of the ITGB3 gene are determined by measuring the expression of the polypeptide defined under accession number P05106 of the Uniprot database, entry version 02-0ct-2024.

[0131] The term “SIRPA” refers to the gene which encodes for the protein “Tyrosineprotein phosphatase non-receptor type substrate 1”. The expression levels SIRPA gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the SIRPA gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000356025.7, ENST00000358771.5, ENST00000400068.7 and ENST00000622179.4 in the Ensembl database under accession number ENSG00000198053 dated (Release 112). In a preferred embodiment, the expression levels of the SIRPA gene are determined by measuring the expression of the polypeptide defined under accession number P78324 of the Uniprot database, entry version 02-0ct-2024.

[0132] The term “TCL1A” refers to the gene which encodes for the protein “T-cell leukemia / lymphoma protein 1A” which enhances the phosphorylation and activation of AKT 1 , AKT2 and AKT3. Promotes nuclear translocation of AKT 1 . The expression levels TCL1A gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the TCL1 A gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000402399.6, ENST00000554012.5, ENST00000555202.1 and ENST00000556450.5 in the Ensembl database under accession number ENSG00000100721 dated (Release 112). In a preferred embodiment, the expression levels of the TCL1A gene are determined by measuring the expression of the polypeptide defined under accession number P56279 of the Uniprot database, entry version 02-0ct-2024.

[0133] The term “S100A9” refers to the gene which encodes for the protein “Protein S100- A9” which is a calcium- and zinc-binding protein which plays a prominent role in the regulation of inflammatory processes and immune response. The expression levels S100A9 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the S100A9 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession number ENST00000368738.4 in the Ensembl database under accession number ENSG00000163220 dated (Release 112). In a preferred embodiment, the expression levels of the S100A9 gene are determined by measuring the expression of the polypeptide defined under accession number P06702 of the Uniprot database, entry version 02-0ct-2024.

[0134] In a further particular form of the diagnostic method of the invention based on the levels of circRNAs, the method further comprises determining the expression levels of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458 hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414 hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 in the sample from the patient. These circ-RNAs are as defined in the circBase database under accession numbers hsa_circ_0000994, hsa_circ_0012673, hsa_circ_0000896, hsa_circ_0000284, hsa_circ_0083329, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_circ_0004062, hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circ_0063526, hsa_circ_0072088, hsa_circ_0016142 and hsa_circ_0084641 respectively..

[0135] Step (ii) of the method of the invention

[0136] The second step of the invention relates to the comparison of the expression level of the gene or the expression level of the circRNA obtained in step (i) of the method and comparing it to a reference value.

[0137] The term “reference value”, as used herein, relates to a predetermined criteria used as a reference for evaluating the values or data obtained from the samples collected from a subject. The reference value or reference level can be an absolute value, a relative value, a value that has an upper or a lower limit, a range of values, an average value, a median value, a mean value, or a value as compared to a particular control or baseline value. A reference value can be based on a large number of samples, such as from population of subjects of the chronological age matched group, or based on a pool of samples. In a particular embodiment of the method I of the invention the reference value relates to the expression level of the gene or genes in a patient or a group of patients with benign lung nodules. In a particular embodiment the group of patients if of at least 2, at least 100, at least 200, at least 300, at least 400, at least 500, at least 600, at least 700, at least 800, at least 900, at least 1000 patients. Typically, the reference value is the value of the same variable which is measured (the expression levels of a gene or combination of genes or the levels of a circRNA) determined in a control sample.

[0138] Such a control sample may comprise any suitable sample, including but not limited to a sample from a control lung cancer patient (can be stored sample or previous sample measurement) with a known outcome; normal tissue or cells isolated from a subject, such as a normal patient or the lung cancer patient, cultured primary cells / tissues isolated from a subject such as a normal subject or the lung cancer patient, adjacent normal cells / tissues obtained from the same organ or body location of the lung cancer patient, a tissue or cell sample isolated from a normal subject, or a primary cells / tissues obtained from a depository. In another preferred embodiment, the control may comprise a reference gene expression level or a circRNA levels from any suitable source, including but not limited to housekeeping genes, an antigen level range from normal tissue (or other previously analyzed control sample), a previously determined level range within a test sample from a group of patients, or a set of patients with a certain outcome or receiving a certain treatment. It will be understood by those of skill in the art that such reference levels obtained from control samples can be used in combination as controls in the methods of the present invention.

[0139] When the expression level of the gene of interest is determined as the levels of mRNA or protein or of the circRNA, then the reference value should also be obtained from mRNA, protein or circRNA levels. When the expression level of the gene of interest is determined as the level of the protein encoded by said gene, then the reference value should also be obtained from protein levels.

[0140] The expression levels of the genes in question and / or of the reference value can be normalized in order to remove variations due to other factors such as the age, weight, sex, general physical condition of the patient and the like. The term “normalization” or “normalizer” as used herein refers to the expression of a differential value in terms of a standard value to adjust for effects which arise from technical variation due to sample handling, sample preparation and measurement rather than biological variation of protein concentration in a sample. For example, when measuring the expression of a differentially expressed protein, the absolute value for the expression of the protein can be expressed in terms of an absolute value for the expression of a standard protein that is substantially constant in expression. This prevents the technical variation of sample preparation and measurement from impeding the measurement of protein concentration levels in the sample.

[0141] As the expert is aware a common source of genes used for normalization of expression levels are housekeeping genes. The term “housekeeping gene” in the present context refers to any gene(s), which display consistent expression under normal and patho-physiological conditions, such that their expression levels can be used to normalize gene expression in a biological sample. In a particular embodiment of the method I of the invention the expression levels are normalized based on the expression level of at least one housekeeping gene. In a more particular embodiment of the diagnostic method of the invention based on the determination of the expression levels of one or more genes, the expression levels are normalized using the expression levels of one or more housekeeping genes selected from a group consisting of: OAZ1 , UBB, MRPL19 and any combination thereof. In a more particular embodiment of the diagnostic method of the invention based on the determination of the levels of circRNAs, the levels are normalized using the expression levels of one or more housekeeping genes selected from a group consisting of the UBB, the MRPL19, the hsa_circ_0001535 genes or a combination thereof.

[0142] The term “0AZ1” refers to the gene which encodes for the protein “Ornithine decarboxylase antizyme 1” which plays a role in the regulation of polyamine synthesis by binding to and inhibiting ornithine decarboxylase. The expression levels OAZ1 gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the 0AZ1 gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000602676.6, ENST00000582888.8, ENST00000588673.3, ENST00000590943.6,

[0143] ENST00000586054.2, ENST00000593012.1 , ENST00000581150.6,

[0144] ENST00000592787.2, ENST00000592727.3, ENST00000589361 .2 and

[0145] ENST00000589739.3 in the Ensembl database under accession number

[0146] ENSG00000104904 (Release 110).. In a preferred embodiment, the expression levels of the OAZ1 gene are determined by measuring the expression of the polypeptide defined under accession number P54368 of the Uniprot database, entry version 28-Jun- 2023

[0147] The term “UBB” refers to the gene which encodes for the protein “Polyubiquitin-B” which exists either covalently attached to another protein, or free (unanchored). The term “MRPL19” refers to the gene which encodes for the protein “39S ribosomal protein L19, mitochondrial” which help in protein synthesis within the mitochondrion. The expression levels MRPL19gene can be determined by the levels of the corresponding mRNA or by the levels of the polypeptide encoded by said gene. In a preferred embodiment, the expression levels of the MRPL19gene are determined by determining the expression levels of one or more of the transcripts defined under the accession numbers ENST00000393909.7, ENST00000409374.5, ENST00000476622.1 ,

[0148] ENST00000358788.10, ENST00000453233.1 , ENST00000492255.1 and

[0149] ENST00000493686.1 in the Ensembl database under accession number

[0150] ENSG00000115364 (Release 110).. In a preferred embodiment, the expression levels of the MRPL19 gene are determined by measuring the expression of the polypeptide defined under accession number of the P49406 Uniprot database, entry version 28-Jun- 2023.

[0151] The term “hsa_circ_0001535” refers to a circRNA which has been studied for its role in various cancer. In colorectal cancer (CRC), hsa_circ_0001535 regulates the progression of colorectal cancer. It acts as a competing endogenous RNA (ceRNA) to inhibit miR-433-3p, which in turn up-regulates the expression of RBPJ, a gene involved in cancer progression. Knockdown of hsa_circ_0001535 has been shown to inhibit cell proliferation, migration, invasion, and angiogenesis, while promoting apoptosis in CRC cells. In ovarian cancer, hsa_circ_0001535 inhibits the proliferation and migration of cancer cells by sponging miR-593-3p and up-regulating PTEN expression.

[0152] Once the expression levels of the genes or of the circRNA in relation to reference values for said genes or circRNAs have been determined and optionally normalized with the expression levels of the adequate housekeeping genes, it is necessary to identify if there are alterations in the expression of said genes (increase or decrease of the expression). The increase or decrease of the expression levels can be described both in increase / decrease percentage as well as fold increase / decrease. The expression level of a gene, either measured as the levels of the mRNA encoded by the genes or measured as the level of the polypeptide encoded by the gene is considered increased in a sample of the patient under study when the levels increase with respect to the reference values by at least 5%, by at least 10%, by at least 15%, by at least 20%, by at least 25%, by at least 30%, by at least 35%, by at least 40%, by at least 45%, by at least 50%, by at least 55%, by at least 60%, by at least 65%, by at least 70%, by at least 75%, by at least 80%, by at least 85%, by at least 90%, by at least 95%, by at least 100%, by at least 110%, by at least 120%, by at least 130%, by at least 140%, by at least 150%, or more. Likewise, the expression of a gene is considered increased in a sample of the patient under study when the levels increase with respect to the reference values by at least 1.1 fold, by at least 1.2 fold, by at least 1.3 fold, by at least 1.4 fold, by at least 1.5 fold, by at least 1.6 fold, by at least 1.7 fold, by at least 1.8 fold, by at least 1.9 fold, by at least 2 fold, by at least 3 fold, by at least 4 fold, by at least 5 fold, by at least 6 fold, by at least 7 fold, by at least 8 fold, by at least 9 fold, by at least 10 fold or more. Similarly, the expression of a gene is considered decreased when its levels decrease with respect to the reference sample by at least 5%, by at least 10%, by at least 15%, by at least 20%, by at least 25%, by at least 30%, by at least 35%, by at least 40%, by at least 45%, by at least 50%, by at least 55%, by at least 60%, by at least 65%, by at least 70%, by at least 75%, by at least 80%, by at least 85%, by at least 90%, by at least 95%, by at least 100% (i.e., absent). Likewise, the expression of a gene is considered decreased in a sample of the patient under study when the levels decrease with respect to the reference values by at least 1.1 fold, by at least 1.2 fold, by at least 1.3 fold, by at least 1.4 fold, by at least 1 .5 fold, by at least 1 .6 fold, by at least 1 .7 fold, by at least 1 .8 fold, by at least 1.9 fold, by at least 2 fold, by at least 3 fold, by at least 4 fold, by at least 5 fold, by at least 6 fold, by at least 7 fold, by at least 8 fold, by at least 9 fold, by at least 10 fold or more.

[0153] In a particular embodiment of diagnostic method of the invention the level of MS4A1 is decreased by at least 1.1 fold, by at least 1.2 fold, by at least 1.3 fold, by at least 1 .4 fold, by at least 1 .5 fold, by at least 1.6 fold, by at least 1 .7 fold, by at least 1.8 fold, by at least 1.9 fold, by at least 2 fold with respect to the reference value, preferably at least 1 .5 fold with respect to the reference value.

[0154] In a particular embodiment of diagnostic method of the invention the level of hsa_circ_0000994 is decreased by at least 1.1 fold, by at least 1.2 fold, by at least 1.3 fold, by at least 1.4 fold, by at least 1.5 fold, by at least 1.6 fold, by at least 1.7 fold, by at least 1 .8 fold, by at least 1 .9 fold, by at least 2 fold with respect to the reference value, preferably at least 1 .5 fold with respect to the reference value.

[0155] In a particular embodiment of diagnostic method of the invention the level of hsa_circ_0012673 is decreased by at least 1.1 fold, by at least 1.2 fold, by at least 1.3 fold, by at least 1.4 fold, by at least 1.5 fold, by at least 1.6 fold, by at least 1.7 fold, by at least 1 .8 fold, by at least 1 .9 fold, by at least 2 fold with respect to the reference value, preferably at least 1 .5 fold with respect to the reference value.

[0156] In another particular embodiment of method I of the invention the level of DDB2 is decreased by at least 1.1 fold, by at least 1 .2 fold, by at least 1.3 fold, by at least 1.4 fold, by at least 1 .5 fold, by at least 1.6 fold, by at least 1.7 fold, by at least 1.8 fold, by at least 1.9 fold, by at least 2 fold with respect to the reference value, preferably at least 1 .5 fold with respect to the reference value.

[0157] Once the expression levels of the gene or genes or the levels of the circRNA as defined in the present invention are determined, the lung nodules in a patient are characterized as being benign or cancerous.

[0158] The expression “is indicative that the lung nodules are cancerous nodules” as it will be understood the expert in the field will understand, although preferred to be, need not be correct for 100% of the patients to be diagnosed or evaluated. The term, however, requires that a statistically significant portion of subjects can be identified as having an increased probability of having a given outcome. Whether a subject is statistically significant can be determined without further ado by the person skilled in the art using various well known statistic evaluation tools, e.g., determination of confidence intervals, p-value determination, cross-validated classification rates and the like etc. Details are found in Dowdy and Wearden, Statistics for Research, John Wiley & Sons, New York 1983. Preferred confidence intervals are at least 50%, at least 60%, at least 70%, at least 80%, at least 90% or at least 95%. The p-values are, preferably, 0.05, 0.01 , 0,005 or lower.

[0159] In an additional embodiment, the determination of whether the lung nodules in the patient are benign or cancerous is further assisted by an analysis of a biopsy sample from one or more nodules obtained from the patient.

[0160] In some embodiments, when lung cancer is ruled out, or a nodule is prognosed to be benign, the subject is monitored periodically. In some embodiments, the periodic monitoring is a pulmonary function test (PET), pulmonary imaging, a biopsy or any combination thereof. In some embodiments, said pulmonary imaging is an x-ray, a chest computed tomography (CT) scan, or a positron emission tomography (PET) scan. In some embodiments, said pulmonary nodule has a diameter of less than or equal to 3 cm. In some embodiments, said pulmonary nodule has a diameter of about 0.8 cm up to 3.0 cm.

[0161] In some embodiments, when the lung nodules in the patient are characterized as being cancerous, the subject is monitored periodically. In some embodiments, the subject has low to moderate cancer risk.

[0162] Method for selecting a subject for a lung cancer therapy

[0163] Another aspect of the present invention relates to a method for selecting a patient having one or more lung nodules for lung cancer therapy (hereinafter “method for personalized therapy of the invention), the method comprising:

[0164] (i) determining the expression level of the MS4A1 gene in a sample from the subject, and

[0165] (ii) comparing the level obtained in (i) to a reference value, wherein an expression level of MS4A1 that is decreased with respect to the reference value is indicative that the subject is selected for a lung cancer therapy.

[0166] A further aspect of the present invention related to a method for selecting a subject having one or more lung nodules for lung cancer therapy comprising:

[0167] (i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient, and (ii) comparing the level obtained in (i) to a reference value, wherein an increased level of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values is indicative that the subject is selected for lung cancer therapy.

[0168] From here onwards both methods are referred to as “therapy assignation method” of the invention.

[0169] All previous embodiment and definitions are equally and validly applicable to the present aspect.

[0170] In one embodiment, the therapy assignation method of the invention based on the determination of the expression levels of the MS4A1 gene further comprises determining the expression level of the DDB2 gene wherein an expression level of the DDB2 gene that is increased with respect to the reference value is indicative that the subject is selected for lung cancer therapy.

[0171] In another embodiment, the therapy assignation method of the invention based on the determination of the expression levels of the MS4A1 gene and of the DDB2 gene further comprise the determination of the expression levels of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in the sample from the subject, wherein if the expression levels of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes are decreased with respect to the reference value, then it is indicative that the subject is selected for lung cancer therapy, then the subject is selected for lung cancer therapy.

[0172] In some embodiments, the therapy assignation method of the invention, the lung cancer therapy is selected from surgical resection of the lung nodules, chemotherapy, radiation therapy, targeted therapy or a combination thereof.

[0173] In some embodiments, the therapy assignation method of the invention, the sample in which the determination of the expression levels of one or more genes or the determination of the levels of the different circRNA is selected from blood, plasma or serum. In a yet more preferred embodiment, the sample is plasma.

[0174] In some embodiments of the therapy assignation method of the invention, the determination of the expression levels of the one or more genes is carried out by determining the levels of RNA transcripts of the gene(s). In yet another embodiment, the determination of the levels of the RNA transcripts is measured in the circulating free RNA.

[0175] In some embodiments of the therapy assignation method of the invention, the expression levels of the one or more genes is normalized using the expression levels of one or more housekeeping genes. In some embodiments, wherein the method for personalized therapy of the invention is based on the determination of the expression levels of one or genes including the MS4A1 gene, then the housekeeping gene is selected from a group consisting of the OAZ1 , UBB, MRPL19 genes. In some embodiments, wherein the method for personalized therapy of the invention is based on the determination of the levels of one or circRNAs including the hsa_circ_0000994 and / or hsa_circ_0012673 then the housekeeping gene is selected from a group consisting of the UBB, the MRPL19, the hsa_circ_0001535 gene or a combination thereof.

[0176] In some embodiments, wherein the method for personalized therapy of the invention is based on the determination of the levels of one or circRNAs including the hsa_circ_0000994 and / or hsa_circ_0012673 and one or more of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847 hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458 hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414 hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 ,

[0177] In some embodiments, the reference value that is used for the determination of whether the expression levels of one or more genes or the levels of one or more circRNA are the expression levels of said one or more genes or the levels of said one or more circRNA present in a patient with benign lung nodules. It will be understood that the reference value is typically determined in the same type of sample in which the expression level / level is determined. In an embodiment, wherein the expression levels of the one or more genes or the levels of the one or more circRNA are determined in a biofluid from the patient, then the reference value will be obtained also from a biofluid, preferably in the same type of biofluid.

[0178] Once the patient has been characterized using the methods of the present invention, the patient will be selected for a therapy for lung cancer if the lung nodules are characterized as being cancerous.

[0179] (As used herein, the terms “therapy”, "treat", "treatment", "treatment", or "amelioration" are used interchangeably. The term refers to therapeutic treatment, the purpose of which is to reverse, reduce, suppress, delay or stop the progression or severity of the condition associated with the disease or disorder. The term "treatment" includes reducing or alleviating at least one adverse effect or condition of a condition, such as cancer, a disease or disorder. Treatment is usually "effective" when one or more symptoms or clinical markers are reduced. Alternatively, treatment is "effective" if disease progression is delayed or halted. That is, "treatment" includes not only the improvement of symptoms or markers, but also the interruption of at least a condition that indicates the progression or worsening of symptoms that would be expected in the absence of treatment. The beneficial or desirable clinical outcome, whether detectable or not, is a reduction in one or more symptoms, a reduction in the extent of the disease, a stable (i.e, not aggravated) condition of the disease, a disease These include, but are not limited to, delayed or slowed progression, amelioration or alleviation of the disease state, and remission (partial or total). The term "treatment" of a disease also includes providing relief from symptoms or side effects of the disease (including symptomatic treatment). In the present context the term therapy refers to lung cancer therapy. In a particular embodiment of the present invention the lung cancer therapy is selected from surgery, chemotherapy, radiation therapy, targeted therapy or a combination thereof. In another preferred embodiment the lung cancer therapy is surgical resection.

[0180] The term “surgery”, as the expert in the field is aware, refers to invasive measures for treating the cancer, in particular, the excision, resection or removal of the tumor tissue.

[0181] The term "chemotherapy" refers to the use of drugs to treat cancer. The drugs are generally administered through oral or intravenous route. Sometimes, chemotherapy is used together with radiation treatment. A "chemotherapeutic drug" is used to connote a compound or composition that is administered in the treatment of cancer. These agents or drugs are categorized by their mode of activity within a cell, for example, whether and at what stage they affect the cell cycle. Alternatively, an agent may be characterized based on its ability to directly cross-link DNA, to intercalate into DNA, or to induce chromosomal and mitotic aberrations by affecting nucleic acid synthesis. Examples of chemotherapy drugs, without limitation, which can be used in the treatment of lung cancer are Abraxane (Paclitaxel Albumin-stabilized Nanoparticle Formulation), Adagrasib, Afatinib Dimaleate, Afinitor (Everolimus), Afinitor Disperz (Everolimus), Alecensa (Alectinib), Alectinib, Alimta (Pemetrexed Disodium), Alunbrig (Brigatinib), Alymsys (Bevacizumab), Amivantamab-vmjw, Atezolizumab, Avastin (Bevacizumab), Bevacizumab, Brigatinib, Capmatinib Hydrochloride, Cemiplimab-rwlc, Ceritinib, Crizotinib, Cyramza (Ramucirumab), Dabrafenib Mesylate, Dacomitinib, Docetaxel, Doxorubicin Hydrochloride, Durvalumab, Enhertu (Fam-Trastuzumab Deruxtecan-nxki), Entrectinib, Erlotinib Hydrochloride, Everolimus, Exkivity (Mobocertinib Succinate), Fam- Trastuzumab Deruxtecan-nxki, Gavreto (Pralsetinib), Gefitinib, Gilotrif (Afatinib Dimaleate), Gemcitabine Hydrochloride, Gemzar (Gemcitabine Hydrochloride), Imfinzi (Durvalumab), Imjudo (Tremelimumab-actl), Infugem (Gemcitabine Hydrochloride), Ipilimumab, Iressa (Gefitinib), Keytruda (Pembrolizumab), Krazati (Adagrasib), Libtayo (Cemiplimab-rwlc), Lorbrena (Lorlatinib), Lorlatinib, Lumakras (Sotorasib), Mekinist (Trametinib Dimethyl Sulfoxide), Methotrexate Sodium, Mobocertinib Succinate, Mvasi (Bevacizumab), Necitumumab, Nivolumab, Opdivo (Nivolumab), Osimertinib Mesylate, Paclitaxel, Paclitaxel Albumin-stabilized Nanoparticle Formulation, Pembrolizumab, Pemetrexed Disodium, Portrazza (Necitumumab), Pralsetinib, Ramucirumab, Retevmo (Selpercatinib), Rozlytrek (Entrectinib), Rybrevant (Amivantamab-vmjw), Selpercatinib, Sotorasib, Tabrecta (Capmatinib Hydrochloride), Tafinlar (Dabrafenib Mesylate), Tagrisso (Osimertinib Mesylate), Tarceva (Erlotinib Hydrochloride), Taxotere (Docetaxel), Tecentriq (Atezolizumab), Tepmetko (Tepotinib Hydrochloride), Tepotinib Hydrochloride, Trametinib Dimethyl Sulfoxide, Tremelimumab-actl, Trexall (Methotrexate Sodium), Vizimpro (Dacomitinib), Vinorelbine Tartrate, Xalkori (Crizotinib), Yervoy (Ipilimumab), Zirabev (Bevacizumab), Zykadia (Ceritinib). Chemotherapy may be administered with using a single drug or a combination of two drugs which are complementary in their action mode of treatment.

[0182] Chemotherapy may be administered before, “neoadjuvant” or after, “adjuvant”, another treatment such as surgical resection of the tumor. The terms “neoadjuvant chemotherapy” and “adjuvant chemotherapy”, at used herein refers to a treatment of cancer with standard chemotherapeutic agents before and after, respectively, surgery where all detectable disease has been removed, to either reduce the size of the tumor and / or reduce the risk of small amounts of cancer remaining.

[0183] The term “radiation therapy” may be used interchangeably with the term “radiotherapy”, is a type of cancer treatment that uses beams of intense energy to kill cancer cells. Radiation therapy most often uses X-rays, but gamma rays, electron beams, or protons also can be used. The term “radiation therapy” most often refers to external beam radiation therapy. During this type of radiation, the high-energy beams come from a machine outside of the patient's body that aims the beams at a precise point on the body.

[0184] The term "targeted therapy", as used herein, relates to application to a patient of a chemical substance known to block growth of cancer cells by interfering with specific molecules known to be necessary for tumorigenesis or cancer or cancer cell growth. Examples known to the skilled artisan are small molecules like, e.g. PARP-inhibitors (e.g. Iniparib), or monoclonal antibodies like, e.g., Trastuzumab. An example of targeted therapy is immunotherapy. The term "immunotherapy" as used herein relates to the treatment of cancer by modulation of the immune response of a subject, e.g. as cell based immunotherapy. Said modulation may be inducing, enhancing, or suppressing said immune response. The term "cell based immunotherapy" relates to a cancer therapy comprising application of immune cells, e.g. T-cells, preferably tumor-specific NK cells, to a subject.

[0185] Method for selecting a subject for lung cancer therapy and treating the subject

[0186] Another aspect of the present invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0187] (i) determining the expression level of MS4A1 in a sample from the subject, and

[0188] (ii) comparing the level obtained in (i) to a reference value,

[0189] (iii) treating the subject with the therapy if the subject is identified as having decreased expression level of MS4A1 gene with respect to the reference value.

[0190] A further aspect of the present invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0191] (a) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient, and

[0192] (ii) comparing the level obtained in (i) to a reference value,

[0193] (iii) treating the subject with the therapy if the patient is identified as having increased levels of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values.

[0194] These two methods will be referred below collectively as “method for personalized therapy of the invention”. All previous embodiment and definitions are equally and validly applicable to the present aspect.

[0195] In one embodiment, the method for personalized therapy of the invention based on the determination of the expression levels of the MS4A1 gene further comprises determining the expression level of the DDB2 gene wherein an expression level of the DDB2 gene that is increased with respect to the reference value is indicative that the subject is selected for lung cancer therapy.

[0196] In another embodiment, the method for personalized therapy of the invention based on the determination of the expression levels of the MS4A1 gene and of the DDB2 gene further comprise the determination of the expression levels of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA,

[0197] RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in the sample from the subject, wherein if the expression levels of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes are decreased with respect to the reference value, then it is indicative that the subject is selected for lung cancer therapy, then the subject is selected for lung cancer therapy.

[0198] In one embodiment, the method for personalized therapy of the invention based on the determination of the expression levels of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs further comprises determining the expression level of at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa circRNA 404643 and hsa circRNA 407081.

[0199] In some embodiments of the methods for personalized therapy of the invention, the lung cancer therapy is selected from surgical resection of the lung nodules, chemotherapy, radiation therapy, targeted therapy or a combination thereof. These terms have been described above in the context of the therapy assignation method and are equally applicable to the present methods.

[0200] In some embodiments, the methods for personalized therapy of the invention, the sample in which the determination of the expression levels of one or more genes or the determination of the levels of the different circRNA is selected from blood, plasma or serum. In a yet more preferred embodiment, the sample is plasma.

[0201] In some embodiments of the methods for personalized therapy of the invention, the determination of the expression levels of the one or more genes is carried out by determining the levels of RNA transcripts of the gene(s). In yet another embodiment, the determination of the levels of the RNA transcripts is measured in the circulating free RNA.

[0202] In some embodiments of the methods for personalized therapy of the invention, the expression levels of the one or more genes is normalized using the expression levels of one or more housekeeping genes. In some embodiments, wherein the method for personalized therapy of the invention is based on the determination of the expression levels of one or genes including the MS4A1 gene, then the housekeeping gene is selected from a group consisting of the OAZ1 , UBB, MRPL19 genes.

[0203] In some embodiments, wherein the method for personalized therapy of the invention is based on the determination of the levels of one or circRNAs including the hsa_circ_0000994 and / or hsa_circ_0012673 then the housekeeping gene is selected from a group consisting of the UBB, the MRPL19, the hsa_circ_0001535 gene or a combination thereof.

[0204] In some embodiments, wherein the method for personalized therapy of the invention is based on the determination of the levels of one or circRNAs including the hsa_circ_0000994 and / or hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847 hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458 hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414 hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 , then the housekeeping gene is selected from a group consisting of the UBB, the MRPL19, the hsa_circ_0001535 gene or a combination thereof.

[0205] In some embodiments, the reference value that is used for the determination of whether the expression levels of one or more genes or the levels of one or more circRNA are the expression levels of said one or more genes or the levels of said one or more circRNA present in a patient with benign lung nodules. It will be understood that the reference value is typically determined in the same type of sample in which the expression level / level is determined. In an embodiment, wherein the expression levels of the one or more genes or the levels of the one or more circRNA are determined in a biofluid from the patient, then the reference value will be obtained also from a biofluid, preferably in the same type of biofluid.

[0206] Once the patient has been characterized using the methods of the present invention, the patient will be selected for a therapy for lung cancer if the lung nodules are characterized as being cancerous. All previous embodiment and definitions are equally and validly applicable to the present aspect.

[0207] Machine learning methods according to the invention

[0208] In a further aspect, the invention relates to a method for determining whether lung nodules in a patient are benign or cancerous comprising:

[0209] (i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,

[0210] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and

[0211] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii),

[0212] In another aspect, the invention relates to a method for selecting a subject having one or more lung nodules for a lung cancer therapy comprising:

[0213] (i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,

[0214] (ii) determining whether the dataset obtained in step (i) is indicative of the patient being selected for lung cancer therapy by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as being selected for lung cancer therapy or not selected for lung cancer therapy and

[0215] (iii) selecting the patient for lung cancer therapy if the expression dataset obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were selected for lung cancer therapy in step (ii).

[0216] In another aspect, the invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0217] (i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,

[0218] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules;

[0219] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii) and

[0220] (iv) treating the subject with the therapy if the patient is identified as having cancerous lung nodules.

[0221] In another aspect, the invention relates to a method for determining whether lung nodules in a patient are benign or are cancerous comprising:

[0222] (i) obtaining a dataset by determining the level of the hsa_circ_0000994, the hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,

[0223] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and

[0224] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii),

[0225] In another aspect, the invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy comprising:

[0226] (i) obtaining a dataset by determining the level of the hsa_circ_0000994, the hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_circ_0004062, hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,

[0227] (ii) determining whether the dataset obtained in step (i) is indicative of the patient being selected for lung cancer therapy by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as being selected for lung cancer therapy or not selected for lung cancer therapy and

[0228] (iii) selecting the patient for lung cancer therapy if the expression dataset obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were selected for lung cancer therapy in step (ii),

[0229] In another aspect, the invention relates to a method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:

[0230] (i) obtaining a dataset by determining the level of the hsa_circ_0000994, hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_circ_0004062, hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,

[0231] (ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and

[0232] (iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii) and

[0233] (iv) treating the subject with the therapy if the patient is identified as having cancerous lung nodules.

[0234] The machine learning methods according to the invention involve a first step wherein a dataset is obtained by determining the level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23,

[0235] TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and

[0236] KLRK1 genes in a sample from the patient.

[0237] Alternatively, in other machine learning methods according to the invention, the first step involves a first step wherein a dataset is obtained by determining the level of the hsa_circ_0000994, the hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847 hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458 hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414 hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient.

[0238] The term dataset, as used herein, refers to data characterizing the expression level of one or more genes or circRNA. Methods suitable for determining the expression of genes or of circRNA have been described above in the context of the previous methods of the invention and are equally applicable to the present methods.

[0239] The terms “"RNA transcript”, circRNA”, “sample” have been descrined above and are equally applicable to the machine learning-based methods according to the present invention. Once the expression data has been determined and the dataset obtaining in step (i), the next step is determining whether said dataset is indicative of the subject having cancerous lung nodules or benign lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with a representative dataset obtained from subjects previously identified as having cancerous lung nodules or being lunb nodules

[0240] The term "machine learning" as used herein refers to algorithms that give a computer the ability to learn without being explicitly programmed, including algorithms that learn from and make predictions about data. Machine learning algorithms employed by the embodiments disclosed herein may include, but are not limited to, random forest ("RF"), least absolute shrinkage and selection operator ("LASSO") logistic regression, regularized logistic regression, XGBoost, decision tree learning, artificial neural networks ("ANN"), deep neural networks ("DNN"), support vector machines (SVM), rule-based machine learning, and / or others. In a particular embodiment of method I of the invention the classification method is selected from logistic regression, random forest, gradient boosting (GB), adaptive boosting (AB), extreme Gradient Boosting (XGB) k-nearest neighbors (kNN), artificial neural network (ANN), support vector machine (SVM), ANOVA, Averaged one-dependence estimators, Convolutional neural network, Extreme learning machine, Feedforward neural network, Logic learning machine, Long short-term memory, Recurrent neural network, Self-organizing map, Bayesian networks, Boosting, Case-based reasoning, Conditional random field, Decision tree algorithms, C4.5 algorithm, C5.0 algorithm, Chi-squared automatic interaction detection, Classification and regression tree, Conditional decision tree, Decision stump, Decision tree, ID3 algorithm, Iterative dichotomiser 3, Random forest, SLIQ, Ensembles of classifiers, Bootstrap aggregating, Boosting, Gaussian process regression, Gene expression programming, Group method of data handling, Inductive logic programming, Information fuzzy networks, Instance-based learning, K-nearest neighbor, Lazy learning, Learning vector quantization, Elastic-net, Lasso, Linear discriminant analysis, Linear regression, Logistic regression, Multinomial logistic regression, Naive bayes classifier, Ordinary least squares, Passive aggressive algorithms, Perceptron, Polynomial regression, Ridge regression I classification, Logistic model tree, Minimum message length, Analogical modelling, Nearest neighbour algorithm, Ordinal classification, Probably approximately correct learning, Quadratic classifiers, Random forests, Ripple down rules, Symbolic machine learning, Active learning, Co-training, Graph-based methods, Generative models, Low-density separation, Transduction, Association rule learning, Apriori algorithm, Eclat algorithm, FP-growth algorithm, Auto-encoders, Cluster analysis, BIRCH, Conceptual clustering, DBSCAN, Expectation-maximization, Fuzzy clustering, Hierarchical clustering, K-means clustering, K-medians, Mean-shift, OPTICS algorithm, Single-linkage clustering, Dimensionality reduction, Canonical correlation analysis, Dynamic mode decomposition, Factor analysis, Feature extraction, Feature selection, Independent component analysis, Linear discriminant analysis, Multidimensional scaling, Non-negative matrix factorization, Partial least squares regression, Principal component analysis, Principal component regression, Projection pursuit, Sammon mapping, T-distributed stochastic neighbour embedding, Expectation-maximization algorithm, Generative topographic map, Information bottleneck method, Manifold learning, Vector quantization, Reinforcement learning, Deterministic policy gradient, Learning automata, Proximal policy optimization, Q-learning, Soft actor-critic, State- action-reward-state-action, Temporal difference learning, Trust region policy Optimization, Bayesian belief network, Bayesian knowledge base, Deep belief networks, Deep boltzmann machines, Deep neural networks, Discrepancy modelling, Gaussian naive bayes, Generative adversarial network, Hierarchical temporal memory, Knowledge-enhanced machine learning, Markov models, Multinomial naive bayes, Neural style transfer, Physics-informed machine learning, Sparse identification of nonlinear dynamics, Transformer, and combinations thereof.

[0241] The expression “machine learning classifier” as used herein refers to a machine learning algorithm which main goal is to identifying to which set of categories a dataset belongs, in the present case the categories are at risk of suffering from preeclampsia or not at risk of suffering from preeclampsia, i.e., a normal pregnancy.

[0242] Machine learning classifiers can be trained to discriminate between the expression data of subjects previously identified as suffering from preeclampsia and the expression data of control subjects or as having a healthy pregnancy by inputting into the classifier the expression data from these two groups. Trained machine learning classifiers can then be used to classify a dataset as indicative of the subject being at risk of suffering preeclampsia sample or a normal pregnancy sample by classifying expression data from the sample. Trained classifier may optionally be tested using expression data from subjects that are known to be suffering or having suffered from preeclampsia and from subjects that have normal pregnancies to determine the sensitivity, specificity, and / or accuracy of the trained machine learning classifier. In a particular embodiment of method I of the invention the trained machine learning classifiers preferably allows a determination of if a subject is at risk of suffering from preeclampsia with an accuracy, a specificity, and / or a sensitivity of at least 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98% or 99%.

[0243] In a particular embodiment of the method I of the invention the trained machine learning classifiers preferably allows a determination of if a subject is at risk of suffering from preeclampsia with a positive predictive value, a negative predictive value and / or an area under the curve of at least 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98% or 99%.

[0244] In some embodiments, the number of variables (or features) in the dataset can be reduced to improve classification by machine learning classifiers. Suitable feature selection methods include, but are not limited to, recursive genetic algorithm (GA), recursive feature elimination (RFE), ANOVA feature selection, Bayesian Adaptive Sampling (BAS), and simple sub-sampling. Additionally, SVMs such as LISVM and SVMRW, which are described below, can simultaneously perform classification as well as feature selection.

[0245] In some embodiments, the classifier is selected from the group consisting of Random Forest (RF), Decision Tree C5.0 (C50), Neuronal Network (NN), gradient boosting machine (GBM), K-Nearest Neighbour (KNN) Naive Bayes (NB) and Polinomial support vector machine (SVMPOLY).

[0246] For clarity, algorithms such as linear regression or logistic regression can be used as part of a machine learning process. However, it will be understood that using linear regression or another algorithm as part of a machine learning process is distinct from performing a statistical analysis such as regression with a spreadsheet program. Whereas statistical modeling relies on finding static relationships between variables, i.e. , finding a solution to outcome determinate problem, a machine learning process may continually update model parameters and adjust a classifier as new data becomes available.

[0247] The expression “machine learning classifier” as used herein refers to a machine learning algorithm which main goal is to identifying to which set of categories a dataset belongs, in the present case the categories are presence of cancerous lung nodules or presence of benign lung nodules.

[0248] Kit of the invention

[0249] In another aspect of the present invention relates to a kit comprising specific reagents for carrying out the methods according to the invention, including the diagnostic methods based on the determination of the expression levels of one or more genes and on the determination of the levels of circRNA, the therapy assignation methods based in the determination of the expression levels of certain genes and on the determination of the levels of circRNA as well as the methods for personalized therapy based on the determination of the expression levels of one or more genes and on the determination of the levels of circRNA

[0250] All previous embodiment and definitions are equally and validly applicable to the present aspect.

[0251] In the context of the present invention, “kit” is understood as a product containing the different reagents necessary for carrying out the methods of the invention packed so as to allow their transport and storage. Additionally, the kits of the invention can contain instructions for the simultaneous, sequential or separate use of the different components which are in the kit. Said instructions can be in the form of printed material or in the form of an electronic support capable of storing instructions such that they can be read by a subject, such as electronic storage media (magnetic disks, tapes and the like), optical media (CD-ROM, DVD) and the like.

[0252] In a particular embodiment the kit of the invention the specific reagents are adequate for determining the expression levels of the genes MS4A1 and DDB2, BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and / or KLRK1.

[0253] In a particular embodiment the kit of the invention the specific reagents are adequate for determining the levels of the circular RNAs hsa_circ_0000994 and / or hsa_circ_0012673 and at least one of circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_circ_0004062, hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081.

[0254] In a further particular embodiment the kit of the invention further comprises reagents adequate for determining the expression levels of one or more housekeeping genes, preferably one or more of the housekeeping genes OAZ1 , UBB, MRPL19 and hsa_circ_0001535.

[0255] The expression “reagent which allows determining the expression level of a gene” means a compound or set of compounds that allows determining the expression level of a gene both by means of the determination of the level of mRNA or by means of the determination of the level of protein. Thus, reagents of the first type include probes capable of specifically hybridizing with the mRNAs encoded by said genes. Reagents of the second type include compounds that bind specifically with the proteins encoded by the marker genes and preferably include antibodies, although they can be specific aptamers.

[0256] In further embodiments of the kit of the invention, the reagents adequate for the determination of the expression levels of one or more genes or circRNAs comprise at least 55% at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98% or at least 99% of the total amount of reagents forming the kit.

[0257] In a particular embodiment of the kit of the invention, the reagents of the kit are nucleic acids which are capable of specifically detecting the RNA level of the genes or circRNAs mentioned above and / or the level of proteins encoded by one or more of the genes mentioned above. Nucleic acids capable of specifically hybridizing with the genes mentioned above can be one or more pairs of primer oligonucleotides for the specific amplification of fragments of the mRNAs (or of their corresponding cDNAs) of said genes or circRNAs.

[0258] In a preferred embodiment, the first component of the kit of the invention comprises a probe which can specifically hybridize to the genes mentioned above.

[0259] The term “specifically hybridizing”, as used herein, refers to conditions which allow hybridizing of two polynucleotides under high stringent conditions or moderately stringent conditions.

[0260] "Stringency" of hybridization reactions is readily determinable by one of ordinary skill in the art, and generally is an empirical calculation dependent upon probe length, washing temperature, and salt concentration. In general, longer probes require higher temperatures for proper annealing, while shorter probes need lower temperatures. Hybridization generally depends on the ability of denatured DNA to reanneal when complementary strands are present in an environment below their melting temperature. The higher the degree of desired homology between the probe and hybridizable sequence, the higher the relative temperature which can be used. As a result, it follows that higher relative temperatures would tend to make the reaction conditions more stringent, while lower temperatures less so.

[0261] In the event that the expression levels of several of the genes identified in the present invention are to be simultaneously determined, it is useful to include probes for all the genes the expression of which is to be determined in a microarray hybridization. The microarrays comprise a plurality of nucleic acids that are spatially distributed and stably associated to a support (for example, a biochip). The nucleic acids have a sequence complementary to particular subsequences of genes the expression of which is to be detected, therefore are capable of hybridizing with said nucleic acids. In the methods of the invention, a microarray comprising an array of nucleic acids is put into contact with a preparation of nucleic acids isolated from the patient object of the study. The incubation of the microarray with the preparation of nucleic acids is carried out in conditions suitable for the hybridization. Subsequently, after the elimination of the nucleic acids which have not been retained in the support, the hybridization pattern is detected, which provides information on the genetic profile of the sample analyzed. Although the microarrays are capable of providing both qualitative and quantitative information of the nucleic acids present in a sample, the invention requires the use of arrays and methodologies capable of providing quantitative information.

[0262] The invention contemplates a variety of arrays with regard to the type of probes and with regard to the type of support used. The probes included in the arrays that are capable of hybridizing with the nucleic acids can be nucleic acids or analogs thereof which maintain the hybridization capacity such as for example, nucleic acids in which the phosphodiester bond has been substituted with a phosphorothioate, methylimine, methylphosphonate, phosphoramidate, guanidine bond and the like, nucleic acids in which the ribose of the nucleotides is substituted with another hexose, peptide nucleic acids (PNA). The length of the probes can of 5 to 50 nucleotides and, preferably, of 7, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 100 nucleotides and vary in the range of 10 to 1000 nucleotides, preferably in the range of 15 to 150 nucleotides, more preferably in the range of 15 to 100 nucleotides and can be single-stranded or doublestranded nucleic acids. The array can contain all the specific probes of a certain mRNA of a certain length or can contain probes selected from different regions of an mRNA.

[0263] The microarrays of the invention contain not only specific probes for the polynucleotides indicating a determined pathophysiological situation, but also containing a series of control probes, which can be of three types: normalization controls, expression level controls and hybridization controls.

[0264] In the event that the expression levels of the genes according to the present invention is determined by measuring the levels of the polypeptide or polypeptides encoded by said gene or genes, the kits according to the present invention comprise reagents which are capable of specifically binding to said polypeptides. For this purpose, the arrays of antibodies are useful. The antibodies of the array include any immunological agent capable of binding to a ligand with high affinity, including IgG, IgM, IgA, IgD and IgE, as well as molecules similar to antibodies which have an antigen binding site, such as Fab', Fab, F(ab')2, single domain antibodies or DABS, Fv, scFv and the like.

[0265] Computer system

[0266] In a particular embodiment of any of the methods of the invention, the method is computer-implemented.

[0267] The term “computer implemented” as used herein refers to fact that the steps of the methods according to the invention are performed by a computer or a computer system. The term “computer implemented” is meant to also include the term “partially computer-implemented”, which refers to a method in which only particular steps, e.g., calculating steps, are computer-implemented, whereas other steps of the method are not.

[0268] A further aspect of the present invention relates to a computer system comprising one or more programs, wherein the one or more programs are include instructions for performing any of the methods of the invention.

[0269] Another aspect of the present invention relates to a computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to perform any of the methods of the invention.

[0270] As will be appreciated by one of skill in the art, the computer readable instructions cause the computing system to carry out the method according to the invention. The device may take the form of an entire hardware embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may include a computer program product on a computer-usable storage medium having computer-usable program code means embodied in the medium. Any suitable computer readable medium may be utilized including hard disks, CD-ROMs, optical storage devices, or magnetic storage devices. The computer-usable or computer-readable medium may be or include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, device, or propagation medium. More specific examples (a non-exhaustive list) of the computer readable medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CD-ROM), a CD ROM, a DVD (digital video disk), or other electronic storage medium, even when accessed through wireless connection. Note that the computer-usable or computer- readable medium could even be paper or another. Suitable medium upon which the program is printed, as the program can be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.

[0271] Computer program code for carrying out operations of the present invention may be written in an object-oriented programming language such as R, Python, Matlab, Java, and Javascript, C #, Smalltalk or C++. However, the computer program code for carrying out operations of the present invention may also be written in conventional procedural programming languages, such as the “C” or FORTRAN programming language or even assembly language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer. In the latter scenario, the remote computer may be connected to the user's computer through a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0272] All previous embodiment and definitions are equally and validly applicable to the present aspect.

[0273] ***

[0274] The invention will be described by way of the following examples which are to be considered as merely illustrative and not limitative of the scope of the invention.

[0275] EXAMPLES

[0276] 1. MATERIALS AND METHODS

[0277] 1.1. Patient samples

[0278] On one hand, a total of three hundred six (32 healthy donors, 118 patients with non-tumor nodules and 156 patients with NSCLC confirmed from tumor nodules) were enrolled to develop de mRNA signature. On the other hand, two hundred and ninety six samples (24 healthy donors, 120 patients with non-tumor nodules and 152 patients with NSCLC confirmed from tumor nodules) were enrolled to develop de circRNA signature.

[0279] The healthy donors did not show any clinical symptoms of cancer, did not suffer from any systemic disease and were not taking medications that could interfere with the study results. The demographic data of patients and control subjects included in the study are shown in Table 1.

[0280] Table 1 . Demographic data of patients and control subjects included in the study.

[0281] Inclusion criteria were patients with pulmonary nodules diagnosed by imaging techniques such as: computed tomography (CT) with or without contrast and / or positron emission tomography (PET). These nodules had to be at least 5 mm in size.

[0282] For the control group were include patients with non-malignant pulmonary pathologies such as pneumonia, tuberculosis, etc., whether or not pulmonary nodules are present by imaging techniques described above.

[0283] The exclusion criteria were patients with a history of cancer and / or under 18 years of age.

[0284] 1.2. Preanalytic step

[0285] For cfRNA isolation, peripheral whole-blood samples were collected in 10-mL sterile BD Vacutainer® K2E tubes (BD, Plymouth, UK), containing 18 mg of EDTA as anticoagulant. Collection tubes were centrifuged for 10 min at 500 xg within 24 hours. Then, using a Pasteur pipette, the supernatant was recovered, transferred to a new sterile tube and recentrifuged using the same conditions.

[0286] 1.3. RNA isolation from clinical samples

[0287] The cfRNA was immediately isolated from 1 .2 mL of the supernatant of the second centrifugation using the QIAsymphony® DSP Virus / Pathogen Midi Kit in a QIAsymphony robot (Qiagen), following the manufacturer’s instructions. The maximum 24 h interval between blood collection and RNA purification was decided based on a stability study performed in-house, which was in line with previous publications about RNA stability in K2-EDTA tubes. The automatic extraction procedure is based on magnetic particles and involves bind, wash and elution steps. Final elution volume was 50 pL in all cases. RNA concentration was estimated by Qubit™ and 5 pL were retrotranscribed to cDNA. We used 2.5 ng / uL of cfRNA as the maximum input for the retrotranscription (RT) reaction. The RT reaction was performed with the nCounter® Low RNA Input Amplification Kit (NanoString Technologies); which contains commercial retrotranscription primer and enzyme mixes. All RNA samples were retrotranscribed and subsequently submitted to nCounter analysis, independently of the quantity or quality of the purified RNA.

[0288] 1.4. nCounter analysis

[0289] The nCounter is a hybridization-based platform (Nanostring Technologies, Seattle, WA) based in a fluorescent barcode that enables direct detection of hundreds (<800) of different target molecules in a single assay. The technology can be used for gene expression profiling, detection of fusion and alternative splicing transcripts or protein analysis, can be easily incorporated into the diagnostic routine and is cost-effective compared to alternative techniques. Regarding gene expression and detection of altered transcripts, the panels can be commercial or custom-made.

[0290] The cDNAs retrotranscribed from cfRNA were pre-amplified with 14 cycles using the corresponding enzyme mix of the nCounter® Low RNA Input Amplification Kit (Nanostring Technologies). Finally, the pre-amplified cDNA was denatured at 95°C for 5 minutes and hybridized at 67°C for 18-24 hours with a mixture of commercial biotinylated capture tags (Nanostring Technologies) and a custom nCounter panel of 48 fluorescently labeled reporter probes, following the Elements Chemistry. The pre- amplification, denaturation and hybridization steps were performed in a Verity thermal cycler (Applied Biosystems).

[0291] Finally, capture, clean up and digital data acquisition of the hybridized mixture were performed in the nCounter Prep Station™ and Digital Analyzer™ (NanoString Technologies). Raw nCounter counts were exported to Excel 2016 (Microsoft) using nSolver software v4.0 (NanoString Technologies). Figure 1 shows an schematic representation of the procedure from sample collection to data analysis.

[0292] 1.5. Software

[0293] Most of the research completed and statistical analysis, in this study was carried out using R language with a collection of libraries of machine learning algorithms for generate prediction models. The principal libraries used were “nanoStringNorm” who provided all the needed tools to pre-process the expression data generate by nCounter technology, and “caret” for make the feature selection ( genes included in the panel) and build models.

[0294] The pipeline Figure 2, return many sets of Feature Selection, based on different normalizations, cut-off of DE and subsets of data. After the analysis of the results, were selected genes as housekeeping genes, and small number of features. The pipeline, Figure 3, use one normalization flow, and the selected features to build models with 6 algorithms.

[0295] 1.6. Normalization and Outlier Samples

[0296] The normalization process is crucial to work with gene expression data, helping us to reduce data noise, pipetting errors, small deviation in the reactive mixing, sample position in the machine.

[0297] In this study the normalization includes the following correction steps by: positive control, negative controls also called background correction, and RNA-levels.

[0298] Different combination of normalization parameters provides different values and that implies small differences on samples suggest to delete, outlier samples and the differential expressed (DE) genes.

[0299] 1.7. Housekeeping genes

[0300] The housekeeping genes, will be genes with low coefficient of variation, expressed in the most of the samples.

[0301] 1.8. Determine Differential Expressed genes A typical Volcano plot is the graphical view of DE genes. Could be use p-value < 0.05 and fold change > 1 or 1.5 as cut-off to determine the DE genes. Those genes could be varying a little bit across different normalizations.

[0302] 1.9. Gene Selection Method

[0303] Feature selection algorithms function under the premise that not all the attributes of the instances are necessary for accurate classification; thus these algorithms seek to identify a subset of the attributes which still accurately represents the characteristics of the instances. There are two common methods to accomplish this: filtering and wrapping. Filtering is typically applied as a preprocessing technique that assigns a score to each attribute and only keeps the attributes which exceed a score threshold. Wrappers select an attribute subset based on the prediction accuracy of a particular classification model.

[0304] Initial set provide a challenge because have more features than samples. So, before to generate prediction models we must reduce the features to a non-redundant set, where the selected features will the most orthogonals between them. Six initial set were tested: All genes, DE by p-value, DE by fold change of 1 and 1.5 and the combination of DE by p-value + fold change. All the highly correlated genes were remove using as cut-off an spearman rank correlation > 0.75. After that the RFE algorithm was used to extract at most 48 features to obtain the maximum accuracy. The feature selection was run for all the normalizations, and selected the genes who appear selected more times.

[0305] 1.10. Prediction Model and Internal Validation

[0306] Classification algorithms seek to assign a class label to each instance in a dataset. To achieve this, the dataset is divided into two non-overlapping subsets: a training set and a test set; alternatively, multiple datasets may be used, one for training and the rest for testing. Classification of the data is completed through a two phase process. In the training phase the algorithm associates the attributes of an instance in the training set with the instance’s known class label; this association defines a prediction model. In the testing phase the model is applied to the instances in the test set in order to evaluate the effectiveness of the model at correctly predicting class labels. This process of training and testing is known as supervised learning. The process of training and testing can be made several times to get an internal validation of the model as an example as LOOCV (leave-one-out cross-validation)

[0307] LOOCV provides an unbiased estimate of a model's performance since it uses all available data for both training and testing in each iteration. It is particularly useful when dealing with small datasets where maximizing the use of available data for training and testing is crucial. LOOCV tends to have lower variance compared to other cross- validation methods because it uses almost the entire dataset for training in each iteration.

[0308] Prediction models were built with seven classification algorithms: Random Forest (RF), Decision Tree C5.0 (C50), Neuronal Network (NN), gradient boosting machine (GBM), K-Nearest Neighbour (KNN) Naive Bayes (NB) and Polinomial support vector machine (SVMPOLY) using different combinations of the selected features. To adjust the model is made an internal validation with LOOCV, and using the receive curve, Receiver-operating characteristic curve (ROC) area - under the roc curve (AUC) to compare the models.

[0309] 1.11. Reproducibility of the algorithm classification

[0310] We had determined the reproducibility of the generated signatures. For this purpose, we have analyzed a total of 35 samples with different times, reagents and personnel.

[0311] For the mRNA-based gene signature, the same result was obtained in all the samples analyzed, so the concordance and reproducibility was 100%.

[0312] For the circRNA-based gene signature, the same result was also obtained in all the samples analyzed, so the concordance and reproducibility was 100%.

[0313] 2. RESULTS

[0314] A commercial panel called “IO360” was applied to get the mRNA gene expression levels. That panel has 750 mRNAs as endogenous genes, 6 positive control, 8 negative controls and 20 mRNA housekeeping genes. On the other hand, a custom panel was applied to get the circRNA gene expression levels. That panel has 78 circRNAs as endogenous genes, 6 positive control, 6 negative controls and 13 mRNA housekeeping genes.

[0315] 2.1. Selection of Housekeeping genes

[0316] The housekeeping genes were selected according to the following criteria:

[0317] The mean should be higher than 4.3 (mean of the positive control F included in the panel)

[0318] The variation coefficient should be lower than 60

[0319] Less number of missing samples

[0320] - And the majority of these genes should be in common between the mRNA and circ-RNA panel Table 2: Results of the housekeeping genes selected

[0321] SD: standard deviation; CV: variation coefficient

[0322] 2.2. Normalization selected Both gene signatures, were development using various sets of normalization parameters. The best conditions selected were:

[0323] Normalization by the selected housekeeping genes

[0324] Normalization by sum of positive controls

[0325] Normalization by the background correction (mean plus 2*SD) 2.3. Differential expressed genes

[0326] DE genes for mRNA and circRNA are selected by p-value < 0.05 plus a cut-off

[0327] | Fold Change| > 1 and 1.5 for circRNA and mRNA respectively.

[0328] Table 3: Results of the selected differential expressed genes If the |Fold_Change| > 1 were used as cut-off for mRNA, a total of 106 genes were detected as Differential Expressed Genes (Figure 4 and figure 5). 2.4. Feature selection - panel

[0329] The pipeline is applied varying the parameters in the steps to obtain a collection of feature selection who appear more times, as the strongest genes. Core genes for mRNA are: MS4A1 and DDB2, beside circRNA are: hsa_circ_0012673 and hsa_circ_0000994. A total of 25 mRNA and 19 circRNA genes were selected.

[0330] Feature selection for the mRNA panel: MS4A1, DDB2, ARG1 , BLK, CD48, CD69, CD8B, CXCL8, IFITM1 , ITGB3, KLRD1 , KLRK1 , LYZ, NFKBIA, PRKCA, PSMB9, RPL23, S100A12, S100A9, SIRPA, SPIB, TAPBP, TAPBPL, TCL1A and TLR8.

[0331] Feature selection for the circRNA panel: hsa_circ_0012673, hsa_circ_0000994, circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_circ_0004062, hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_ 103809, hsa_circRNA_404643 and hsa_circRNA_407081.

[0332] MRNA. Model Selections. Summary

[0333] In the table below is represented the number of times that each genes appears dependently the type of normalization for mRNA. In bold are represented the genes that are overexpressed.

[0334] Table 4: Number of times in which each genes is represented in the feature selection for mRNA

[0335] Normalization based on circRNA. Model Selections. Summary

[0336] The table shows the number of times that each of the genes appears dependently the type of normalization for circRNA. In bold are represented the genes that are overexpressed.

[0337] Table 5: Number of time of each genes is represented in the feature selection for circRNA

[0338] Normalization based on 2.5. Classification Model - Signature

[0339] The classification models were built with the 7 algorithms. The best results were obtained with Random Forest for both the mRNA and the circRNA signature. Table 6: Seven classification models with the results of ROC, sensitivity and specificity. 2.6. Role of the core genes in mRNA and circRNA signatures

[0340] In the mRNA gene signature, DDB2 and MS4A1 have a special role. The DDB2 gene has lower expression in healthy (donors or benign nodules) than in cancer patients. In contrast, the MS4A1 gene has higher expression in healthy subjects than in cancer patients (Figure 6).

[0341] In the circRNA signature, hsa_circ_0000994 and hsa_circ_0012673 have special role. hsa_circ_0000994 shows lower expression in healthy (donors or benign nodules) than cancer patients. hsa_circ_0012673 has higher expression in healthy than cancer patients (Figure 7).

Claims

CLAIMS1. A method for determining whether lung nodules in a patient are benign or cancerous comprising:(i) determining the expression level of the MS4A1 gene in a sample from the patient, and(ii) comparing the level obtained in (i) to a reference value, wherein an expression level of the MS4A1 gene that is decreased with respect to the reference value is indicative that the lung nodules are cancerous nodules.

2. A method for selecting a subject having one or more lung nodules for a lung cancer therapy comprising:(i) determining the expression level of the MS4A1 gene in a sample from the subject, and(ii) comparing the level obtained in (i) to a reference value, wherein an expression level of the MS4A1 gene that is decreased with respect to the reference value is indicative that the subject is selected for a lung cancer therapy.

3. A method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:(i) determining the expression level of the MS4A1 gene in a sample from the subject, and(ii) comparing the level obtained in (i) to a reference value,(iii) treating the subject with the therapy if the subject is identified as having decreased expression level of the MS4A1 gene with respect to the reference value.

4. The method according to any of claims 1 to 3 wherein step (i) further comprises determining the expression level of the DDB2 gene in the sample from the subject wherein, an expression level of the DDB2 gene that is increased with respect to the reference value is indicative that the lung nodules are cancerous nodules, an expression level of the DDB2 gene that is increased with respect to the reference value is indicative that the subject is selected for lung cancer therapy orthe subject is treated with the therapy if the subject is identified as having increased expression level of the DDB2 gene with respect to the reference value5. The method according to claim 4 wherein step (i) further comprises determining the expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1 A, S100A9, CXCL8 and KLRK1 genes in the sample from the subject, wherein- if the expression levels of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes are decreased with respect to the reference value, then it is indicative that the lung nodules are cancerous nodules,- if the expression levels of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes are decreased with respect to the reference value, then it is indicative that the subject is selected for lung cancer therapy. the subject is treated with the therapy if the subject is identified as having decreased levels of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes.

6. The method according to any of claims 1 to 5 wherein step (i) comprises determining the expression level of the MS4A1 , DDB2, BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1 A, S100A9, CXCL8 and KLRK1 genes in the sample from the subject, wherein an increased expression level of the DDB2 with respect to the reference value and decreased expression levels of the MS4A1 , DDB2, BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes with respect to their respective reference values are indicative that the lung nodules are cancerous nodules,an increased expression level of the DDB2 with respect to the reference value and decreased expression levels of the MS4A1 , DDB2, BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes with respect to their respective reference values are indicative that the subject is selected for lung cancer therapy. the subject is treated with the therapy if the subject is identified as having increased expression level of the DDB2 with respect to the reference value and decreased expression levels of the MS4A1 , DDB2, BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes.

7. A method for determining whether lung nodules in a patient are benign or cancerous comprising:(i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,(ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and(iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii),8. A method for selecting a subject having one or more lung nodules for a lung cancer therapy comprising:(i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69,PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,(ii) determining whether the dataset obtained in step (i) is indicative of the patient being selected for lung cancer therapy by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as being selected for lung cancer therapy or not selected for lung cancer therapy and(iii) selecting the patient for lung cancer therapy if the expression dataset obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were selected for lung cancer therapy in step (ii).

9. A method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:(i) obtaining a dataset by determining the expression level of the MS4A1 gene, the expression level of the DDB2 gene and expression level of one or more of the BLK, TAPBPL, NFKBIA, KLRD1 , PSMB9, S100A12, TAPBP, ARG1 , CD69, PRKCA, RPL23, TLR8, CD48, CD8B, LYZ, IFITM1 , ITGB3, SIRPA, SPIB, TCL1A, S100A9, CXCL8 and KLRK1 genes in a sample from the patient,(ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules;(iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii) and(iv) treating the subject with the therapy if the patient is identified as having cancerous lung nodules.

10. The methods of any of claims 7 to 9 wherein the classifier is selected from the following algorithms: Random Forest (RF), Decision Tree C5.0 (C50), Neuronal Network (NN), gradient boosting machine (GBM), K-Nearest Neighbour (KNN) Naive Bayes (NB) and Polinomial support vector machine (SVMPOLY).

11. The method according to any of claims 1 to 9 wherein the expression levels are normalized based on the expression level of at least one housekeeping gene.

12. The method according to claim 10 wherein the housekeeping gene is selected from a group consisting of the OAZ1 , UBB, MRPL19 genes or a combination thereof.

13. A method for determining whether lung nodules in a patient are benign or are cancerous comprising:(i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient and(ii) comparing the level obtained in (i) to a reference value, wherein an increased level of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values is indicative that the lung nodules are cancerous nodules.

14. A method for selecting a subject having one or more lung nodules for lung cancer therapy comprising:(i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient, and(ii) comparing the level obtained in (i) to a reference value, wherein an increased level of the hsa_circ_0000994 circular RNA and / or a decreased level of the hsa_circ_0012673 circular RNA with respect to their respective reference values is indicative that the subject is selected for lung cancer therapy.

15. A method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:(i) determining the level of the hsa_circ_0000994 and / or hsa_circ_0012673 circular RNAs in a sample from the patient, and(ii) comparing the level obtained in (i) to a reference value and(iii) treating the subject with the therapy if the patient is identified as having increased levels of the hsa_circ_0000994 circular RNA and / or a decreasedlevel of the hsa_circ_0012673 circular RNA with respect to their respective reference values.

16. A method for determining whether lung nodules in a patient are benign or are cancerous comprising:(i) obtaining a dataset by determining the level of the hsa_circ_0000994, hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458 hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256 hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809 hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,(ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and(iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii),17. A method for selecting a subject having one or more lung nodules for lung cancer therapy comprising:(i) obtaining a dataset by determining the level of the hsa_circ_0000994, hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458 hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256 hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809 hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,(ii) determining whether the dataset obtained in step (i) is indicative of the patient being selected for lung cancer therapy by using a trained machine learningclassifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as being selected for lung cancer therapy or not selected for lung cancer therapy and (iii) selecting the patient for lung cancer therapy if the expression dataset obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were selected for lung cancer therapy in step (ii),18. A method for selecting a subject having one or more lung nodules for lung cancer therapy and treating the subject with said therapy comprising:(i) obtaining a dataset by determining the level of the hsa_circ_0000994, hsa_circ_0012673 and at least one of the circFARSA, circRNA_HIPK3, circ_C1orf116, hsa_circ_0000566, hsa_circ_0000847, hsa_circ_0002360, hsa_circ_0003148, hsa_ci rc_0004062 , hsa_circ_0004458, hsa_circ_0006276, hsa_circ_0007037, hsa_circ_0043256, hsa_circ_0086414, hsa_circRNA_103237, hsa_circRNA_103809, hsa_circRNA_404643 and hsa_circRNA_407081 circular RNAs in a sample from the patient,(ii) determining whether the dataset obtained in step (i) is indicative of the subject having cancerous lung nodules by using a trained machine learning classifier which correlates the dataset obtained in step (i) with representative datasets obtained from subjects which were previously identified as suffering from benign or from cancerous lung nodules; and(iii) identifying the lung nodules in the patient as cancerous if the expression data obtained in step (i) is correlated by the machine learning classifier with the representative expression data from subjects which were identified as suffering from cancerous lung nodules in step (ii) and(iv) treating the subject with the therapy if the patient is identified as having cancerous lung nodules.

19. The methods of any of claims 16 to 18 wherein the classifier is selected from the following algorithms: Random Forest (RF), Decision Tree C5.0 (C50), Neuronal Network (NN), gradient boosting machine (GBM), K-Nearest Neighbour (KNN) Naive Bayes (NB) and Polinomial support vector machine (SVMPOLY).

20. The method according to any of claims 13 to 19 wherein the expression levels is normalized based on the expression level of at least one housekeeping gene.

21. The method according to claim 20 wherein the housekeeping gene is selected from a group consisting of the UBB, the MRPL19, the hsa_circ_0001535 gene or a combination thereof.

22. The method according to any of claims 1 to 21 wherein(i) The determining of whether the lung nodules in the patient are benign or cancerous,(ii) The selection of the subject having one or more lung nodules for a lung cancer therapy or(iii) The selecting of the subject for lung cancer therapy and the treatment of the subject: is further assisted by an analysis of a biopsy sample from one or more nodules obtained from the patient.

23. The method according to claim 2, 3, 8, 9, 14,15, 17 or 18 wherein the lung cancer therapy is selected from surgical resection of the lung nodules, chemotherapy, radiation therapy, targeted therapy or a combination thereof.

24. The method according to any of claims 1 to 23 wherein the lung cancer is non-small cell lung cancer (NSCLC).

25. The method according to any of claims 1 24 wherein the sample is selected from blood, plasma or serum.

26. The method according to claim 25 wherein the sample is plasma.

27. The method according to any of claims 1 to 26 wherein the determination of the expression level of the one or more genes is carried out by determining the levels of RNA transcripts of the gene(s).

28. The method according to claim 27 wherein the determination of the levels of the RNA transcripts is measured in the circulating free RNA.

29. The method according to any of claims 1 to 28 wherein the reference value relates to the expression level of the gene in a patient with benign lung nodules.

30. A kit comprising specific reagents for carrying out the method according to any of claims 1 to 29.

31. A computer system comprising one or more programs, wherein the one or more programs are include instructions for performing the method of any one of claims 1 to 2932. A computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by one or more processors of an electronic device, cause the electronic device to perform the method of any one of claims 1- 29.

Images