EP4377699A1 - Procede de caracterisation d'une tumeur - Google Patents
Procede de caracterisation d'une tumeurInfo
- Publication number
- EP4377699A1 EP4377699A1 EP22760870.0A EP22760870A EP4377699A1 EP 4377699 A1 EP4377699 A1 EP 4377699A1 EP 22760870 A EP22760870 A EP 22760870A EP 4377699 A1 EP4377699 A1 EP 4377699A1
- Authority
- EP
- European Patent Office
- Prior art keywords
- nucleosides
- tumor
- grade
- biological sample
- individual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
- G01N33/575—Immunoassay; Biospecific binding assay; Materials therefor for cancer
- G01N33/57557—Immunoassay; Biospecific binding assay; Materials therefor for cancer of other specific parts of the body, e.g. brain
-
- C—CHEMISTRY; METALLURGY
- C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
- C12N—MICROORGANISMS OR ENZYMES; COMPOSITIONS THEREOF; PROPAGATING, PRESERVING, OR MAINTAINING MICROORGANISMS; MUTATION OR GENETIC ENGINEERING; CULTURE MEDIA
- C12N15/00—Mutation or genetic engineering; DNA or RNA concerning genetic engineering, vectors, e.g. plasmids, or their isolation, preparation or purification; Use of hosts therefor
- C12N15/09—Recombinant DNA-technology
- C12N15/10—Processes for the isolation, preparation or purification of DNA or RNA
- C12N15/1003—Extracting or separating nucleic acids from biological samples, e.g. pure separation or isolation methods; Conditions, buffers or apparatuses therefor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/62—Detectors specially adapted therefor
- G01N30/72—Mass spectrometers
- G01N30/7233—Mass spectrometers interfaced to liquid or supercritical fluid chromatograph
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/88—Integrated analysis systems specially adapted therefor, not covered by a single one of the groups G01N30/04 - G01N30/86
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/53—Immunoassay; Biospecific binding assay; Materials therefor
- G01N33/575—Immunoassay; Biospecific binding assay; Materials therefor for cancer
- G01N33/57535—Immunoassay; Biospecific binding assay; Materials therefor for cancer of the large intestine, e.g. colon, rectum or anus
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/48—Biological material, e.g. blood, urine; Haemocytometers
- G01N33/50—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing
- G01N33/68—Chemical analysis of biological material, e.g. blood, urine; Testing involving biospecific ligand binding methods; Immunological testing involving proteins, peptides or amino acids
- G01N33/6803—General methods of protein analysis not limited to specific proteins or families of proteins
- G01N33/6848—Methods of protein analysis involving mass spectrometry
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/09—Supervised learning
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N2030/022—Column chromatography characterised by the kind of separation mechanism
- G01N2030/027—Liquid chromatography
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N30/00—Investigating or analysing materials by separation into components using adsorption, absorption or similar phenomena or using ion-exchange, e.g. chromatography or field flow fractionation
- G01N30/02—Column chromatography
- G01N30/88—Integrated analysis systems specially adapted therefor, not covered by a single one of the groups G01N30/04 - G01N30/86
- G01N2030/8809—Integrated analysis systems specially adapted therefor, not covered by a single one of the groups G01N30/04 - G01N30/86 analysis specially adapted for the sample
- G01N2030/8813—Integrated analysis systems specially adapted therefor, not covered by a single one of the groups G01N30/04 - G01N30/86 analysis specially adapted for the sample biological materials
- G01N2030/8827—Integrated analysis systems specially adapted therefor, not covered by a single one of the groups G01N30/04 - G01N30/86 analysis specially adapted for the sample biological materials involving nucleic acids
Definitions
- the present invention relates to an in vitro method for characterizing a tumor, based on the quantitative analysis of modified and unmodified nucleosides isolated from a biological sample. More particularly, the subject of the invention is a method for predicting the grade of a glial tumour. According to another particular aspect, the subject of the invention is a method for detecting the presence of a tumour.
- the present invention therefore lies in the fields of oncology and molecular biology more particularly applied to medical diagnosis.
- characterization of a tumor is an essential prerequisite for choosing the most appropriate treatment for the patient.
- characterization of a tumor is meant the characterization of the status or degree of evolution of a given tumor, it may be in particular for example the evaluation of the degree of evolution of a tumor of a known tissue, the attribution to a tumor of a previously defined grade, or any other characterization such as in particular the determination of the initial or metastatic nature of a tumor.
- Gliomas or glial tumors, are the most common tumors of the central nervous system, they are characterized by significant variability in age of onset, grading, histological features and ability to progress and eventually metastasize.
- Gliomas are classified according to their morphology and degree of malignancy.
- the consensus classification of the World Health Organization (WHO) assigns a degree of malignancy from I to IV to gliomas, with glioblastomas, or grade IV tumors, being the most aggressive and deadliest form.
- WHO World Health Organization
- gliomas and glioblastomas are related to the current lack of effective diagnostic strategies.
- the selection of a personalized treatment requires an accurate classification of tumors.
- the main diagnostic methods used clinically for the detection of gliomas are based on neurological tests and neuroimaging methods, performed when the disease is already at an advanced stage.
- Diagnosis of the tumor requires analysis of the patient's tissues from a biopsy or surgical resection. From this sample, several molecular analyzes are carried out: candidate gene expression test, DNA copy number counting, methylation profiling, phospho-protein pathway profiling and genetic sequencing.
- biopsy-based diagnoses have limitations regarding tumor grading and patient stratification.
- the grades of the glioma are difficult to distinguish, and more particularly grades II and III.
- the establishment of the grade requires a delicate anatomo-pathological analysis, often carried out independently by two specialists.
- Grade II denotes a benign tumor while grade III represents a transition to glioblastoma multiforme, which is the most aggressive state.
- Janzer's publication (“Neuropathologic and molecular pathology of gliomas.” R-C Janzer, Rev. Med. Switzerland, 5, 1501-4, 2009) describes the classification of gliomas according to the WHO, based on histological and immuno-histochemical criteria , as well as on genetic profiles highlighting the alteration of cell DNA: determination of hypermethylation of the MGMT gene promoter (for glioblastomas) and detection of loss of chromosomes lp and 19q (for tumors oligodendrogliales).
- the inventors have now developed a method for characterizing a tumor which exploits the quantitative data of the epitranscriptome.
- the epitranscriptome encompasses all of the chemical modifications carried by the bases of ribonucleic acids (RNA), a set which is also referred to by the term “epigenetics of RNA”.
- a method according to the invention comprises supplying a biological sample from a subject suffering from a tumor and obtaining quantities of modified and unmodified nucleosides from said sample, said quantities being grouped together in a vector (in the mathematical sense of the term).
- a method according to the invention comprises the subsequent computer analysis of said vector for the characterization of a tumor. Said characterization of a tumor makes it possible to predict clinical and medical information about the tumor from the sample subject to analysis.
- a method according to the invention comprises the computer analysis of said vector for the prediction of the grade of said tumor.
- epitopranscriptomic profile or quite simply “profile”, the vector which groups together the quantities of each nucleoside, modified or unmodified.
- the modified and unmodified nucleosides are derived from: i) total RNA extracted from cells of a biological sample from a patient, ii) extracellular RNA from a biological sample from a patient, and/or iii) an extract of metabolites from a biological sample isolated from a patient.
- the nucleosides from the total RNA extracted from cells of a biological sample from a patient and/or from the extracellular RNA from a biological sample from a patient are obtained by the fragmentation of the RNA into nucleotides then their dephosphorylation.
- the nucleosides derived from an extract of metabolites from a biological sample isolated from a patient are obtained by extracting the metabolites from a biological sample then dephosphorylation of said metabolites, according to appropriate methods well known to those skilled in the art.
- Said metabolites are in particular derived from the catabolism of RNA, the nucleosides present in monomeric form can also be designated by the expression of so-called “free” nucleosides.
- the modified and unmodified nucleosides are the nucleosides present in the total RNA extracted from cells from a biopsy of said tumour.
- nucleosides is meant the glycosamines consisting of a nucleic base linked to the anomeric carbon atom of a pentose residue by a glycosidic bond from the NI nitrogen atom of a pyrimidine or the N9 atom of a purine.
- pentose is ribose
- nucleosides designates in this case ribonucleosides.
- RNA nucleosides are also referred to as “epitranscriptomic marks” or “epitranscriptomic modifications”.
- RNA nucleosides Table 1 which can be used for the characterization of tumours
- modified nucleosides which can be used in a method according to the invention, in particular in the analysis of gliomas, are listed (Table 2).
- Table 1 Table 1
- an epitranscriptomic profile can include, according to the needs of the application, a greater number of modified nucleosides, to be determined from among the known nucleosides (Jonkhout et al, “The RNA landscape modification in human disease”, RNA, Dec;23 (12): 1754-1769, 2017).
- a complete list of all modified nucleosides, which may, depending on the needs of the analysis, be included in the transcriptomic profiles is publicly available.
- the epitranscriptomic profile of a sample characterizes said sample.
- Said epitranscriptomic profile can be obtained by any known technique of the state of the art, and in particular by mass spectrometry, in particular by mass spectrometry coupled with chromatography.
- the step of analyzing the epitranscriptomic profile for clinical prediction purposes is based on a supervised automatic learning method. Learning is performed on the profiles from a cohort, that is to say cell samples for which the clinical characteristic variable to be predicted is known beforehand. The “computer model” thus created by learning can then be used (in prediction mode) in order to predict the clinical variable for any new sample.
- the inventors have also developed a method for normalizing the raw quantitative data making it possible to obtain an epitranscriptomic profile containing comparable relative quantities.
- a method according to the invention makes it possible to predict the grade of a glioma from a biological sample of a patient suffering from a tumor , in particular from a sample comprising tumor cells. More particularly, a method according to the invention makes it possible to distinguish grades II and III of a glioma from a sample comprising tumor cells.
- a method according to the invention makes it possible to predict the survival of a patient from a sample biological material isolated from said patient, in particular from a tumor sample.
- the inventors have also developed a method for detecting the presence of a tumor in an individual, from a biological sample isolated from this individual, comprising the steps of: a) isolating the nucleosides of said biological sample, by extracting : i) total cellular RNA and its fragmentation into nucleosides, ii) extracellular RNA and its fragmentation into nucleosides, and/or iii) nucleosides from monomeric catabolites, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10 preferably at least 20, different nucleosides obtained during step a), and c) establishment, for said biological sample, of a nucleoside profile from the respective amounts of each of the nucleosides obtained during step b), said profile being characteristic of the presence of said tumour.
- the technical steps of a method for detecting the presence of a tumor in an individual have the same characteristics as the technical steps of a method for characterizing a tumor.
- a method according to the invention is therefore advantageously used to characterize a tumour.
- a method according to the invention is advantageously used to detect the presence of a tumour.
- the subject of the invention is an in vitro method for characterizing a tumor of an individual, from a biological sample isolated from this individual, comprising the steps of: a) isolation of the nucleosides of said sample biological, by extracting: i) total cellular RNA and its fragmentation into nucleosides, ii) extracellular RNA and its fragmentation into nucleosides, and/or iii) nucleosides from monomeric catabolites, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10 preferably at least 20, different nucleosides obtained during step a), and c) establishment, for said biological sample, of a profile of nucleosides from the respective quantities of each of the nucleosides obtained during step b), said profile being characteristic of said tumour.
- a method according to the invention is based on the simultaneous analysis of the quantity of different nucleosides derived from the total cellular RNA of a biological sample, and/or derived from the extracellular RNA and its fragmentation into nucleosides and/or derived from nucleosides obtained from monomeric catabolites present in said sample, a method according to the invention therefore comprises the simultaneous analysis of multiple variables, and not the quantitative detection of a single marker.
- profile or “nucleoside profile” is meant a vector of quantities of nucleosides.
- total cellular RNA is meant all of the cellular RNA extracted according to well-known and accessible methods.
- Total cellular RNA includes transfer RNA (tRNA), messenger RNA (mRNA), ribosomal RNA (rRNA), and other non-coding RNAs. Said total cellular RNA is therefore present here in a polymeric form.
- extracellular RNA is meant all of the extracellular RNA present in polymeric form, extracted according to well-known and accessible methods. This polymeric form of extracellular RNA is in particular also designated by the expression “circulating RNA”. Said extracellular RNA is derived from the in vivo enzymatic degradation of transport RNA (tRNA), messenger RNA (mRNA) and/or ribosomal RNA (rRNA) and other types of RNA, in particular RNAs not coding.
- tRNA transport RNA
- mRNA messenger RNA
- rRNA ribosomal RNA
- nucleosides derived from monomeric catabolites is meant the nucleosides obtained, according to well-known and accessible methods, from the catabolites present in a monomeric form in the sample. These monomeric catabolites are derived from the in vivo enzymatic degradation of transport RNA (tRNA), messenger RNA (mRNA) and/or ribosomal RNA (rRNA) and other types of RNA, in particular non- coding.
- tRNA transport RNA
- mRNA messenger RNA
- rRNA ribosomal RNA
- isolated and determination of a respective quantity of at least 3 different nucleosides is meant the isolation and determination of a quantity of each of the “at least 3” nucleosides taken individually.
- the subject of the invention is therefore an in vitro method for characterizing a tumor of an individual, from a biological sample isolated from this individual, comprising the steps of: a) isolating the nucleosides of said biological sample by the extraction of the total cellular RNA and its fragmentation into nucleosides, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10 preferably at least least 20, different nucleosides obtained during step a), and c) establishment, for said biological sample, of a profile of nucleosides from the respective quantities of each of the nucleosides obtained during step b), said profile being characteristic of said tumour.
- the nucleosides can also be present in the biological sample in an extracellular polymeric form, in particular also designated by the expression “circulating RNA”. Nucleosides can also be present in a monomeric form (metabolites) in the biological sample. Said extracellular RNAs and monomeric nucleosides are derived from the in vivo enzymatic degradation of transport RNA (tRNA), messenger RNA (mRNA) and/or ribosomal RNA (rRNA) and other types of RNA, in particular non-coding RNAs.
- tRNA transport RNA
- mRNA messenger RNA
- rRNA ribosomal RNA
- a method according to the invention is based on the simultaneous analysis of the amount of different nucleosides from the extracellular RNA of a biological sample.
- the subject of the invention is an in vitro method for characterizing a tumor of an individual, from a biological sample isolated from this individual, comprising the steps of: a) isolation of the nucleosides said biological sample, by the extraction of the extracellular RNA and its fragmentation into nucleosides, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10 preferably at least 20, different nucleosides obtained during step a), and c) establishment, for said biological sample, of a profile of nucleosides from the respective quantities of each of the nucleosides obtained during step b), said profile being characteristic of said tumour.
- a method according to the invention is based on the simultaneous analysis of the quantity of different nucleosides resulting from the monomeric catabolites present in a biological sample.
- the subject of the invention is an in vitro method for characterizing a tumor of an individual, from a biological sample isolated from this individual, comprising the steps of: a) isolation of the nucleosides said biological sample, by extracting the monomeric catabolites present in said sample, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10 preferably at least 20, nucleosides different values obtained during step a), and c) establishment, for said biological sample, of a profile of nucleosides from the respective quantities of each of the nucleosides obtained during step b), said profile being characteristic of said tumor.
- said biological sample is in particular to be chosen from: a solid biological sample, particularly a biopsy, and more particularly a biopsy of said tumor, and a liquid biological sample, in particular a sample of a bodily fluid from said individual, more particularly a sample of blood, plasma, serum or urine.
- biopsy we mean the removal of a very small part of an organ or tissue.
- the biological sample is a biopsy
- the first embodiment of the method according to the invention in which the total cellular RNA is extracted then fragmented, is preferred.
- the biological sample is a liquid biological sample
- the second and the third embodiment of the method according to the invention in which respectively the extracellular RNA is extracted then fragmented, or in which the RNA in the form of isolated nucleosides is extract, are preferred.
- said biological sample is of sufficient volume, or comprises a sufficient number of cells, to allow a reliable quantitative determination of at least 3 nucleosides resulting from the fragmentation of an extract of total cellular RNA of said sample.
- the cellular RNA is total is extracted according to a method chosen from the methods accessible to those skilled in the art, in particular a method as described in the present example.
- a liquid sample such as blood or urine
- said sample is treated beforehand if necessary, in particular in order to eliminate any interfering compounds, to concentrate said sample and/or to determine a standard value of concentration of a reference element, such as creatinine in urine, this standard value serving to calibrate the concentration of the sample from which the nucleoside profile is established.
- the total cellular RNA, the extracellular RNA and the isolated nucleosides are obtained from a biological sample by any method known to a person skilled in the art, said method comprises in particular an extraction step, optionally a fragmentation step, and a dephosphorylation step.
- the subject of the invention is therefore an in vitro method for characterizing a tumor of an individual from a biopsy of said tumor, said method comprising the preparation, from said biopsy, of an extract of total cellular RNA and fragmentation of said RNA into nucleosides.
- At least 3 isolated nucleosides from the biological sample obtained by i) the preparation of an extract of total cellular RNA and its fragmentation into nucleosides, ii) by the preparation of an extract of the extracellular RNA and its fragmentation into nucleosides, and/or iii) by the extraction of the isolated nucleosides, are isolated and their quantity respectively determined, said at least 3 nucleosides are chosen from: unmodified nucleosides: adenosine (A), cytidine (C), guanosine (G), uridine
- Modified nucleosides result from the action of a large number of highly specific enzymes, nucleosides in particular undergo methylation and rearrangement of carbon-nitrogen bonds.
- Said modified nucleosides are all modified nucleosides known at the date of this application, these nucleosides are cited in particular in the publication by Jonkhout et al ("The RNA modification landscape in human disease", RNA, Dec; 23 (12): 1754-1769, 2017), and in Table 2 of this application.
- said at least 3 nucleosides are chosen from the groups consisting of: unmodified nucleosides: adenosine (A), cytidine (C), guanosine (G), uridine ( U),
- said at least 3 nucleosides are chosen from the groups consisting of: unmodified nucleosides: adenosine (A), cytidine (C), guanosine (G), uridine (U), and
- a method which is the subject of the invention comprises the isolation and the quantitative determination of at least 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 , 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28 or 29 different nucleosides resulting from the fragmentation of the total RNA of said biological sample.
- a method which is the subject of the invention comprises the isolation and the quantitative determination of at least 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28 or 29 different nucleosides resulting from the fragmentation of the extracellular RNA of said biological sample.
- a method which is the subject of the invention comprises the isolation and the quantitative determination of at least 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28 or 29 different nucleosides resulting from the extraction of nucleosides from said biological sample.
- a method which is the subject of the invention comprises the isolation and the quantitative determination of at least 3 different nucleosides resulting from the fragmentation of the total RNA of said biological sample and/or from the fragmentation of the extracellular RNA and/or the extraction of isolated nucleosides, said nucleosides being chosen from the following: adenosine (A), cytidine (C), guanosine (G), uridine (U), 2'-0- methyladenosine (Am ), 1-methyladenosine (mlA), N6,N6-dimethyladenosine (m66A), N6,N6,2'-0-trimethyladenosine (m66Am), N6-methyladenosine (m6A), N6,2'-0-dimethyladenosine (m6Am) , N4-acetylcytidine (ac4C), 2'-0-methylcytidine (Cm), 5- hydroxymethylcy
- the isolation and the determination of a respective quantity of at least 3 nucleosides are implemented by any means of analysis known to those skilled in the art. These means include in particular chromatography, in particular high performance reverse phase liquid chromatography (RP-HPLC) or capillary electrophoresis (CE).
- RP-HPLC high performance reverse phase liquid chromatography
- CE capillary electrophoresis
- spectrometry means in particular mass spectrometry. More specifically, these means include liquid chromatography-tandem mass spectrometry (LC-MS/MS), an analytical technique that combines the separating power of liquid chromatography with the mass analysis capability highly sensitive and selective triple quadrupole mass spectrometry. The strength of this technique lies in the separation power of liquid chromatography for a wide range of compounds, combined with the ability of mass spectrometry to quantify compounds with a high degree of sensitivity and selectivity, depending unique mass/charge (m/z) transitions of each compound of interest.
- LC-MS/MS liquid chromatography-tandem mass spectrometry
- the mixture of nucleosides obtained by fragmentation is analyzed using high performance liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) of the triple quadrupole type in the Multiple Reaction Monitoring (MRM) mode.
- MRM mode is a highly sensitive and specific technique that allows the quantification of molecules by mass spectrometry. This scanning mode is dependent on tandem mass spectrometry and more particularly on triple quadrupoles or hybrid trap mass spectrometry systems.
- the MRM scan mode is based on the selection of ions of specific mass and charge of a molecule, ions called precursor ions or parent ions, as well as on the corresponding fragment ions after fragmentation in the collision cell.
- the first quadrupole will allow the precise selection of the specific precursor ions of the molecules of interest which will then be fragmented in the second quadrupole.
- the resulting fragment ions are then selected in the third quadrupole.
- the two ions mass/charge then correspond to a highly specific transition of the molecule of interest.
- the subject of the invention is an in vitro method for characterizing a tumor of an individual, from a biopsy of this individual, comprising the steps of: a) preparation, from of said biological sample, of an extract of total cellular RNA and fragmentation of the polymeric RNA into nucleosides, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10, preferably at least 20, different nucleosides from step a), c) establishment, for said biological sample, of a profile of nucleosides from the respective amounts of each of the nucleosides obtained during step b), said profile being characteristic of said tumour.
- the subject of the invention is an in vitro method for characterizing a tumor of an individual, said tumor being a tumor located in one of the following organs: rectum, colon, breast, pancreas, kidney, lung, or hematologic tumor, including leukemia.
- the subject of the invention is an in vitro method for characterizing a glial tumor of an individual, from a biological sample isolated from this individual, comprising the steps of: a) preparation , from said biological sample, of an extract of total cellular RNA and fragmentation of said RNA into nucleosides, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10, preferably at least 20, different nucleosides from step a), c) establishment, for said biological sample, of a profile of nucleosides from the respective amounts of each of the nucleosides obtained during step b), said profile being characteristic of said tumour, and d) prediction of a grade of said glial tumor by a first previously trained classification model, from the profile established during step c).
- glial tumor or "glioma” refer to various brain tumors that develop from normal glial cells in the brain.
- the grade of a glial tumor represents the most important determinant for the survival of an individual carrying such a tumor.
- Non-tumor brain tissue is characterized by many cells with normal characteristics and few mitotic characteristics, without endothelial proliferation.
- Grade II tumors also called “astroblastomas” include a greater number of cells comprising polymorphic nuclei in mitosis.
- Grade III tumors are also called “anaplastic astroblastomas”.
- Grade IV tumors correspond to glioblastoma multiforme.
- classification model means a previously trained automatic learning algorithm, in particular during supervised learning, as well as a set of learning data allowing the training of the aforementioned algorithm, and a set of evaluation data.
- said first classification model may comprise: an automatic learning algorithm, more particularly a supervised learning neural network, or a multi-class probabilistic classification algorithm, trained beforehand with a set of learning data .
- the training data are specific to the question asked and, on the other hand, specific to the type of cancer targeted.
- the training phase of the learning algorithm uses the learning data to produce a classification model, which is itself specific to the question asked and specific to the type of cancer targeted.
- the learning phase infers the parameters of the model based on this data and the question. For example, for a grade determination question, the classification model returns one response among four possible responses if four grades are distinguished.
- the classification model answers with "tumor" or "healthy", i.e. a choice among two possible answers.
- the classification model produced by the learning phase is a program implemented on a computer in order to obtain a prediction on the question considered from the data of an epitranscriptomic profile from a sample.
- This program is downloadable and installable, thus allowing installation on a system other than the one on which it was produced.
- the training data set can comprise a multitude of data pairs, each of the data pairs comprising a first data item representing a nucleoside profile and a second data item representing the tumor grade for this profile.
- the training data set may include a training set and a test set, also referred to as "(evaluation set)[CFi]", of the model.
- the model can thus be tested on the training set and the test set can be used to determine whether the model has been learned successfully or not.
- the training game and the test game may be different.
- the test set may correspond to part of the training set.
- the training data set can be formed beforehand from data obtained in the laboratory by analysis of samples obtained from individuals suffering from cancer and whose tumor grade has been determined beforehand.
- the classification model has reached a satisfactory level of learning on all the profiles of the test set if the classification reaches, for example, 85% accuracy; in other words, it is estimated that the classification model has reached a satisfactory level of learning on all the profiles of the test set if the classification reaches for example at most 15% error.
- the classification model may consist of a computer program.
- a classification model implemented in a method according to the invention consists of a computer program which potentially executes a technical function consisting of steps of the classification method.
- the execution of said program by a computer produces a digital object, which is a technical object.
- Said computer program can be written in any computer language such as for example in C, C++, JAVA, Python, etc.
- the classification model may comprise a support vector machine, a random forest, a linear discriminant analysis; these methods are called in English respectively: "Support Vector Machines”, “Random Forests” and “Linear Discriminant Analysis” (LDA).
- LDA Linear Discriminant Analysis
- said learning algorithm is chosen in particular from:
- the prediction of a grade of a glial tumor can comprise: the prediction of a grade II glial tumor, the prediction of a grade III glial tumor or prediction of a grade IV glial tumour.
- the invention more particularly relates to an in vitro method for predicting a grade of a glial tumor of an individual, from a biological sample of said individual, and in particular a biopsy of said glial tumor, in which the predicting a grade of said glial tumor by a previously trained classification model, comprising: predicting a grade II glial tumor, predicting a grade III glial tumor and predicting a grade glial tumor IV.
- a method for predicting a grade of a glial tumor of an individual comprises distinguishing between a grade II glial tumor and a grade III or IV glial tumor; the distinction between a grade III glial tumor and a grade II or IV glial tumour; the distinction between a grade IV glial tumor and a grade II or III glial tumour.
- the invention even more particularly relates to an in vitro method for characterizing a glial tumor of an individual, from a biopsy of said tumor, comprising: a) the preparation, from said biopsy, of an extract of total cellular RNA and fragmentation of said RNA into nucleosides, b) isolation and quantitative determination of at least 3 nucleosides resulting from said fragmentation, chosen from: adenosine (A), cytidine (C), guanosine (G ), uridine (U), 2'-0-methyladenosine (Am), 1-methyladenosine (mA), N6,N6- dimethyladenosine (m66A), N6,N6,2'-0-trimethyladenosine (m66Am), N6- methyladenosine (m6A), N6,2'-0-dimethyladenosine (m6Am), N4-acetylcytidine (ac4C), 2'-0-methylcytidine (Cm), 5-
- step c) establishing, for said tumor, a profile from the respective quantitative values of the nucleosides obtained during step b), said profile being characteristic of said tumor, and d) predicting a grade of said glial tumor by a previously trained classification model, from the profile established during step c), in which the prediction of a grade of a glial tumor is chosen from: the prediction of a grade II glial tumor , the prediction of a grade III glial tumor and the prediction of a grade IV glial tumor.
- the subject of the invention is an in vitro method for characterizing a glial tumor of an individual, said method comprising the steps of: a) preparing, from said biological sample, an extract of total cellular RNA and fragmentation of said RNA into nucleosides, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10, preferably at least 20, different nucleosides from the step a), c) establishment, for said biological sample, of a nucleoside profile from the respective quantities of each of the nucleosides obtained during step b), said profile being characteristic of said tumor, and d) prediction of a state of survival of said individual, by a second classification model trained beforehand, from the profile established during step c).
- said second classification model may comprise: an automatic learning algorithm, more particularly a supervised learning neural network, or a probabilistic classification algorithm, previously trained with a second set of learning data.
- Said second training data set may comprise a multitude of data pairs, each of the data pairs comprising a first data item representing a nucleoside profile and a second data item representing the survival status for this profile.
- This training dataset may include a training set and a model evaluation set.
- the model can thus be tested on the training game and the game evaluation can be used to determine if the training of the model is satisfactory or not.
- the training game and the evaluation game can be different.
- the evaluation game may correspond to part of the training game.
- the learning data set can be constituted beforehand from data obtained in the laboratory by analysis of samples obtained from individuals suffering from cancer and whose survival status has been determined beforehand.
- the classification model has reached a satisfactory level of learning on all the profiles of the evaluation set if the classification reaches 85% accuracy; in other words, it is estimated that the classification model has reached a satisfactory level of learning on all the profiles of the evaluation set if the classification reaches at most 15% error.
- the second classification model can consist of a computer program.
- the computer program can be written in any computer language such as for example in C, C++, JAVA, Python, etc.
- the second classification model may comprise a support vector machine, a random forest, a linear discriminant analysis; these methods are called in English respectively: "Support Vector Machines”, “Random Forests” and “Linear Discriminant Analysis”.
- the subject of the invention is a classification model, previously trained on a learning data set, to predict a grade of a glial tumor of an individual suffering from a tumor, from a profile of nucleosides obtained by implementing a method according to the invention.
- Said classification model for predicting the grade of a glial tumor comprises a machine learning algorithm previously trained and evaluated, in particular during supervised learning, with a set of training data relating to the prediction of a grade of a glial tumor, said learning game comprising a training game and an evaluation game, both relating to the prediction of a grade of a glial tumor.
- the invention also relates to a method for constructing a classification model for predicting the grade of a glial tumor, comprising at least:
- the subject of the invention is a second classification model, previously trained on a learning data set, to predict a survival status of an individual suffering from a tumor, from a profile of nucleosides obtained by implementing a method according to the invention.
- Said classification model for predicting a survival status of an individual suffering from a tumor comprises an automatic learning algorithm previously trained and evaluated, in particular during supervised learning, with a set of learning data relating to prediction of a survival status of an individual, said learning game comprises a training game and a test game, both relating to the prediction of a survival status of an individual suffering from a tumor.
- the invention also relates to a method for constructing a classification model to predict an individual's survival status, comprising at least:
- a learning data set relating to the prediction of an individual's survival status comprising a training set and a test set
- the present invention relates to the use of a classification model according to the invention for the prediction of a grade of a glial tumour.
- the subject of the present invention is the use of a classification model according to the invention for the stratification of a patient suffering from a glial tumour, in combination with at least one other biological marker characteristic of said patient.
- the present invention relates to the use of a classification model according to the invention for the prediction of a state of survival of an individual.
- the subject of the invention is an in vitro method for detecting the presence of a tumor in an individual, from a biological sample isolated from this individual, comprising the steps of: a) isolation of nucleosides from said biological sample, by extracting: i) total cellular RNA and its fragmentation into nucleosides, ii) extracellular RNA and its fragmentation into nucleosides, and/or iii) nucleosides derived from monomeric catabolites , and preferably nucleosides derived from monomeric catabolites, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10 preferably at least 20, different nucleosides obtained during the step a), and c) establishment, for said biological sample, of a nucleoside profile from the respective quantities of each of the nucleosides
- the subject of the invention is an in vitro method for detecting the presence of a tumor in an individual, from a blood sample isolated from this individual, comprising the steps of: a ) isolation of the nucleosides of said biological sample, by extracting the nucleosides from the monomeric catabolites, b) isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10, preferably at least 20, different nucleosides from step a), c) establishment, for said biological sample, of a nucleoside profile from the respective amounts of each of the nucleosides obtained during step b), said profile being characteristic of said tumor, and d) prediction of the presence of said tumor by a previously trained classification model, from the profile established during step c).
- the subject of the invention is an in vitro method for detecting the presence of a colorectal tumor in an individual, from a blood sample isolated from this individual, comprising the steps of: To. isolation of the nucleosides of said biological sample, by the extraction of the nucleosides resulting from the monomeric catabolites, b. isolation and determination of a respective quantity of at least 3, preferably at least 5, preferably at least 10 preferably at least 20, different nucleosides from step a), vs. establishment, for said biological sample, of a profile of nucleosides from the respective quantities of each of the nucleosides obtained during step b), said profile being characteristic of said tumour, and d. prediction of the presence of said colorectal tumor by a previously trained classification model, from the profile established during step c).
- the invention also relates to the use of a method according to the invention for detecting the presence of a tumor, said tumor being a tumor located in one of the following organs: rectum, colon, breast, pancreas , kidney, lung, or hematologic tumor, including leukemia.
- a subject of the invention is also the use of a method according to the invention for detecting the presence of a tumor of the digestive tract, in particular a colorectal tumor.
- the subject of the invention is a classification model, previously trained on a learning data set, to detect the presence of a tumor in an individual, from a nucleoside profile obtained by implementation of a method according to the invention.
- This classification model comprises an automatic learning algorithm previously trained and evaluated, in particular during supervised learning, with a set of learning data relating to the detection of the presence of a tumor in an individual, said set learning includes a training game and a test game, both relating to the detection of the presence of a tumor in an individual.
- the subject of the invention is a classification model, trained beforehand on a learning data set, to detect the presence of a colorectal tumor in an individual, from a nucleoside profile obtained by placing implementation of a method according to the invention.
- Said classification model comprises a machine learning algorithm previously trained and evaluated, in particular during supervised learning, with a set of learning data relating to the detection of the presence of a colorectal tumor in an individual.
- the invention also relates to a method for constructing said classification model for detecting the presence of a tumor, comprising at least:
- a machine learning algorithm for a classification task - the selection of a machine learning algorithm for a classification task, - the supply of a learning data set relating to the detection of the presence of a tumor in an individual, comprising a training set and a test set,
- the invention also relates to a method for constructing a classification model for detecting the presence of a colorectal tumor, comprising at least: the selection of an automatic learning algorithm for a classification task, providing a set of training data relating to the detection of the presence of a colorectal tumor in an individual, comprising a training set and a test set, and a step of training the prediction of the presence of a colorectal tumor in an individual, by said algorithm, using said training data set.
- the present invention relates to the use of a classification model according to the invention for the detection of a tumour, in particular a colorectal tumour.
- the subject of the present invention is the use of a classification model according to the invention for the detection of a tumor, in particular a colorectal tumor, in combination with at least one other biological marker characteristic of said patient.
- the present invention finally relates to a diagnostic method comprising the implementation of a method according to the invention for characterizing a tumour.
- the present invention also relates to a diagnostic method comprising the implementation of a method according to the invention for predicting a grade of a glial tumour.
- the present invention also relates to a diagnostic method comprising the implementation of a method according to the invention for predicting the state of survival of a patient.
- Said diagnostic method may further comprise a histological analysis of the tissues.
- the present invention finally relates to a diagnostic method comprising the implementation of a method according to the invention for detecting a tumour.
- the present invention also relates to a diagnostic method comprising the implementation of a method according to the invention for detecting a colorectal tumour.
- Said diagnostic method can comprise a histological analysis of the tissues.
- FIG. 1 represents the overall scheme of the experiment, where LC-MS/MS denotes liquid chromatography associated with mass spectrometry and the raw data (data) are the epitranscriptomic profiles obtained by LC-MS/MS.
- FIG. 2 represents the overall diagram of the bioinformatics process
- the raw data are the epitranscriptomic profiles obtained by LC-MS/MS
- the normalized data are the epitranscriptomic profiles after normalization
- MS designates mass spectrometry (combined with liquid chromatography).
- FIGS. 3A, 3B and 3C represent, in the form of a boxplot, six graphs representing respectively the relative quantity (in percentage) of six nucleosides modified according to the grade of glial tumor.
- said grade is designated on the abscissa by: "Normal”, “Grade-II”, “Grade-III” or “Grade IV” respectively indicating a sample of non-tumorous glial tissue or a sample of grade glial tumor II, III or IV.
- Figure 3A shows two examples of nucleosides whose quantity decreases with increasing glial tumor grade: (from left to right) oxo8G and mlG.
- Figure 3B shows two examples of nucleosides whose quantity increases with increasing glial tumor grade: (from left to right) m6Am and Gm.
- Figure 3C shows two examples of nucleosides whose quantity varies slightly with increasing glial tumor grade: (from left to right) mA and m7G. The scales are different depending on the graphs.
- Figure 4 represents the percentage of explained variance of the first components of the Principal Component Analysis (PCA) of the epitranscriptomic profiles of the cohort. Along the abscissa, the components are numbered from 0 to 9. Along the ordinate, the percentages of variance explained by these components.
- PCA Principal Component Analysis
- PCA Principal Component Analysis
- Each of the axes represents, respectively, principal component 0 (39.24%), principal component 1 (23.27%) and principal component 2 (8.58%).
- the "star” symbols represent the guard "normal”, “triangle” grade II, "square” grade III and “cross” grade IV, respectively.
- EXAMPLE 1 Analysis of transcriptomic data from glial cell samples
- This section presents the cohort used, the sample preparation, the method for obtaining epitranscriptomic profiles and the computer analysis program. This section then presents the results of the exploratory analysis of the profiles of the cohort, the prediction of tumor grades and the prediction of survival.
- the processing of the biological sample begins with the extraction of the RNA by phase separation in order to obtain an RNA sample of at least 100 ng.
- the treatment continues with the enzymatic hydrolysis of the polymeric RNA and the dephosphorylation of the nucleosides.
- RNA The enzymatic digestion of RNA is carried out as follows: a quantity of 400 ng of RNA is diluted in a total volume of 20 ⁇ L of milliQ water, to which 3 ⁇ l of ammonium acetate (0.1 M pH 5.3) and 0.001 enzymatic unit (U) of Nuclease PI (Sigma, N8630). Incubation at 42° C. is carried out for 2 hours. Then, 3 ⁇ l of 1 M ammonium acetate and 0.001 U of alkaline phosphatase (Sigma, P4252) are added. The mixture is then incubated at 37° C. for 2 hours.
- nucleoside solution is diluted twice and filtered with 0.22 ⁇ m filters (Millex®-GV, Millipore, SLGVR04NL). Finally, 5 ⁇ L of each sample is injected and all samples are analyzed in triplicate by LC-MSMS.
- LC Liquid chromatography
- nucleosides are separated by Nexera LC-40 systems (Shimadzu) using a SynergiTM Fusion-RP C18 column (4 ⁇ m particle size, 250 mm x 2 mm, 80 ⁇ ) (Phenomenex, 00G-4424-B0).
- the mobile phase consists of 5 mM ammonium acetate adjusted to pH 5.3 with acetic acid (solvent A) and pure acetonitrile (solvent B).
- solvent A acetic acid
- solvent B pure acetonitrile
- the 30 minute gradient elution starts with 100% phase A followed by a linear gradient to 8% solvent B at 13 minutes.
- Solvent B is further increased to 40% in 10 minutes. After 2 minutes, solvent B is reduced to 0% at 25.5 minutes.
- the initial conditions are regenerated by rinsing with 100% solvent A for an additional 4.5 minutes.
- the flow rate is 0.4 ml/min and the column temperature is 35°
- Mass spectrometry in “Multiple Reaction Monitoring” (MRM) mode is carried out as follows: detection is carried out by Shimadzu TripleQuad 8060 in positive ion mode. The mass spectrometry operates in dynamic MRM mode with a retention time window of 3 min and a maximum cycle time set at 258 ms. Peak areas are determined using Skyline 4.1 software (Pino LK et al, “The Skyline ecosystem: Informatics for quantitative mass spectrometry proteomics.” Mass Spectrom Rev. 2020 May;39(3): 229-244. 2020 ).
- the mass spectrometer was calibrated to accurately identify and quantify 25 modified nucleosides (Table 2) and 4 unmodified nucleosides (A, U, G, T) (Table 1).
- the mass spectrometry device used is a Shimadzu TripleQuad 8060 in multiple reaction monitoring mode. Each sample was injected three times, providing three technical replicates. For each nucleoside, the homogeneity of the retention time given by the mass spectrometer is checked. Measurements showing a discrepancy of more than 6% were discarded. This results in a data table containing the quantity measurements of each nucleoside, in each replicate, for all samples. This table is then analyzed using our computer programs.
- the necessary characteristics of the programs are as follows: a) they take as input mass spectrometry data in a file in tabular format (CSV format); b) the quantifications of area and retention time from the spectrometer must be given in the form of real values with a precision of at least 10-1; c) they implement a multi-class supervised machine learning algorithm among those mentioned above; d) they implement the learning phase, the evaluation phase, and the prediction mode; e) they use the classification model in prediction mode to classify the epitranscriptome profile of a patient sample in order to predict the tumor grade.
- CSV format tabular format
- Tables 3, 4 and 5 indicate, for each of the nucleosides analyzed, the normalized data value for each of grades II, III and IV of glioma, and for healthy (“normal”) tissue.
- Fig. 1 The joint analysis of epitranscriptomic profiles and clinical variables of interest (Fig. 1) is carried out, in particular concerning the grade in the case of gliomas. This process can be adapted to any type of clinical variable. In this example, we are trying to distinguish the grades of cancer, which can be difficult to establish by pathological examination.
- Pre-processing the cohort profiles resulted in a table of 77 rows, with one row per sample, and 29 columns, with one column per measurement. For each of the samples, the mention of the grades of the tumors or the mention “normal” for the healthy samples, was added. An exploratory statistical analysis of this table was carried out to assess the relevance of the signal contained in the profiles on the grade information.
- nucleosides are studied in the samples of the same grade, and we compare these variations between the grades.
- the experimental results suggest a grouping of nucleosides into four groups: i) those whose quantity increases with grade, i.e. between tissue non-tumor cerebral brain (designated for simplification as “normal” on the ordinate of the graphs) and the grades, II, III and IV, in particular the nucleosides oxo8G, mlG, queuosine and Ac4C (as shown for example in Fig. 3A); ii) those whose quantity decreases with grade (as shown for example in Fig. 3B), iii) those which vary weakly with grades (as shown for example in Fig. 3C) and iv) the remaining nucleosides, which do not satisfy not the conditions of membership of the first three groups.
- a Principal Component Analysis (PCA) of these data was carried out, in order to carry out a reduction in dimension, not to be confused with a selection of the "characteristics", in other words of the nucleosides, to see if the variations in quantity could be combined into a small number of components.
- PCA is an exploratory multivariate analysis method that reduces the dimensions of data while capturing their variability.
- Components are new variables that combine data from initial observations to best capture their variability while reducing the number of variables to analyze.
- the components result from the projection of the initial data on other axes of the multidimensional space.
- the components are ordered in descending order of percentage of variance explained. This percentage associated with each component indicates its importance in describing the initial data.
- Fig. 4 presents the graph of the percentage of variance explained for the first 10 components.
- PCA is a classic data analysis technique.
- Machine learning method enabling accurate grade prediction of tumors and healthy samples
- the learning method must belong to the classification category.
- SVM Support Vector Machine
- the prediction accuracy of the SVM algorithm equipped with a linear kernel on the profiles of the subset tested is 0.90, out of a maximum of 1, which is remarkable.
- the level of prediction accuracy is maintained when the learning is reiterated then the tests with new random partitionings of the dataset, which shows the robustness of the learning tool developed.
- results of the evaluation allow us to compare our method of normalization (denoted by SUM, for sum) with the formulas used in the literature.
- the classic standardization which consists in dividing the measurement of a modified nucleoside, for example mlA, by that of the corresponding unmodified nucleoside, here the measurement of A.
- the precision according to the use of different formulas is between 0.8 and 0.9, and is therefore always less than or equal to (but never greater than) the precision of the normalization formula SUM.
- the quality of grade prediction is not particularly related to the optimization of a learning method on a given cohort, since two very different learning methods obtain similar results.
- the quality of the prediction is therefore linked to the power of the signal contained in the transcriptomic profiles.
- RNAs Differences in the relative amounts of certain epigenetic modifications of RNAs have been demonstrated according to different samples, whether healthy or tumorous. These differences make it possible in particular to separate the different tumor grades.
- a supervised machine learning algorithm applied to the nucleoside quantity vectors effectively distinguishes glioma grades, and in particular distinguishes grades II and III, with remarkable accuracy given the relatively small cohort size. Moreover, this method also makes it possible, from the same data, to estimate patient survival using a supervised automatic learning method.
- Circulating RNA is extracted from plasma using a kit (miRNeasy Serum/Plasma). The RNAs are digested with PI nuclease and treated with alkaline phosphatase in order to obtain a mixture of nucleosides. Circulating free nucleosides are extracted from the same plasma samples using a methanol extraction procedure. They do not require enzymatic treatment before passing through mass spectrometry.
- LC Liquid chromatography
- mass spectrometry and bioinformatics analyzes are carried out as indicated in example 1.
- each sample is analyzed three times independently, thus making it possible to obtain three technical replicas for each.
- the raw data is processed in order to be normalized as in example 1.
- these steps produce an epitranscriptomic profile per sample.
- a machine learning method has been developed and tested to determine the presence or absence of a tumor from epitranscriptomic profiles alone, ie without using any information other than the quantities of nucleosides.
- SVM Support Vector Machine
- the algorithm was first trained and then tested in order to assess its ability to predict the presence or not of a tumor in the sample.
- the machine learning method gives a prediction with 100% accuracy and 100% sensitivity.
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Molecular Biology (AREA)
- Immunology (AREA)
- Biomedical Technology (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Hematology (AREA)
- Urology & Nephrology (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- Biotechnology (AREA)
- Microbiology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Cell Biology (AREA)
- Biophysics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Genetics & Genomics (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- Wood Science & Technology (AREA)
- Zoology (AREA)
- Organic Chemistry (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- Epidemiology (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
Abstract
Description
Claims
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| FR2108279A FR3125882B1 (fr) | 2021-07-29 | 2021-07-29 | Procede de caracterisation d’une tumeur |
| PCT/EP2022/070551 WO2023006588A1 (fr) | 2021-07-29 | 2022-07-21 | Procede de caracterisation d'une tumeur |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| EP4377699A1 true EP4377699A1 (fr) | 2024-06-05 |
Family
ID=80446963
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| EP22760870.0A Pending EP4377699A1 (fr) | 2021-07-29 | 2022-07-21 | Procede de caracterisation d'une tumeur |
Country Status (10)
| Country | Link |
|---|---|
| US (1) | US20240363254A1 (fr) |
| EP (1) | EP4377699A1 (fr) |
| JP (1) | JP2024529994A (fr) |
| KR (1) | KR20240035609A (fr) |
| CN (1) | CN118019981A (fr) |
| AU (1) | AU2022320853A1 (fr) |
| CA (1) | CA3226423A1 (fr) |
| FR (1) | FR3125882B1 (fr) |
| MX (1) | MX2024001349A (fr) |
| WO (1) | WO2023006588A1 (fr) |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7799519B2 (en) | 2005-07-07 | 2010-09-21 | Vanderbilt University | Diagnosing and grading gliomas using a proteomics approach |
| PL214886B1 (pl) * | 2008-10-03 | 2013-09-30 | Akad Medyczna | Sposób diagnozowania choroby nowotworowej układu moczowo-płciowego |
| PL409275A1 (pl) * | 2014-08-26 | 2016-02-29 | Gdański Uniwersytet Medyczny | Sposób wykrywania nowotworów układu moczowo-płciowego, zastosowanie oraz zestaw do wykrywania choroby nowotworowej układu moczowo-płciowego |
-
2021
- 2021-07-29 FR FR2108279A patent/FR3125882B1/fr active Active
-
2022
- 2022-07-21 KR KR1020247006194A patent/KR20240035609A/ko active Pending
- 2022-07-21 CN CN202280065175.3A patent/CN118019981A/zh active Pending
- 2022-07-21 US US18/292,155 patent/US20240363254A1/en active Pending
- 2022-07-21 WO PCT/EP2022/070551 patent/WO2023006588A1/fr not_active Ceased
- 2022-07-21 EP EP22760870.0A patent/EP4377699A1/fr active Pending
- 2022-07-21 MX MX2024001349A patent/MX2024001349A/es unknown
- 2022-07-21 JP JP2024506155A patent/JP2024529994A/ja active Pending
- 2022-07-21 AU AU2022320853A patent/AU2022320853A1/en active Pending
- 2022-07-21 CA CA3226423A patent/CA3226423A1/fr active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| KR20240035609A (ko) | 2024-03-15 |
| FR3125882B1 (fr) | 2025-07-18 |
| MX2024001349A (es) | 2024-04-19 |
| FR3125882A1 (fr) | 2023-02-03 |
| US20240363254A1 (en) | 2024-10-31 |
| CA3226423A1 (fr) | 2023-02-02 |
| JP2024529994A (ja) | 2024-08-14 |
| CN118019981A (zh) | 2024-05-10 |
| WO2023006588A1 (fr) | 2023-02-02 |
| AU2022320853A1 (en) | 2024-02-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| He et al. | Several miRNAs derived from serum extracellular vesicles are potential biomarkers for early diagnosis and progression of Parkinson’s disease | |
| JP6659809B2 (ja) | 口腔癌検出用唾液バイオマーカー | |
| CN112005306A (zh) | 选择、管理和分析高维数据的方法和系统 | |
| Herrero‐Lorenzo et al. | Small RNAs in plasma extracellular vesicles define biomarkers of premanifest changes in Huntington's disease | |
| AU2024250885A1 (en) | Systems and methods for early-stage cancer detection and subtyping | |
| Yan et al. | Deep neural network based tissue deconvolution of circulating tumor cell RNA | |
| CN113862351B (zh) | 体液样本中鉴定胞外rna生物标志物的试剂盒及方法 | |
| EP4377699A1 (fr) | Procede de caracterisation d'une tumeur | |
| WO2024137664A1 (fr) | Méthodes de détection de glioblastome dans des vésicules extracellulaires | |
| JP2008532495A (ja) | 神経精神障害に関するバイオマーカーの検出 | |
| WO2016027029A2 (fr) | Procédé pour déterminer le pronostic de survie d'un patient atteint d'un cancer du pancréas | |
| AU2024210217A1 (en) | Methods and systems for detecting and assessing liver conditions | |
| Mullins et al. | Proof of concept: Detection of cell free RNA from EDTA plasma in patients with lung cancer and non-cancer patients | |
| WO2023017543A1 (fr) | Dosage de biomarqueurs à base de miarn salivaires pour le cancer buccal | |
| Zhu et al. | Identification of key genes and regulatory mechanisms in adult degenerative scoliosis | |
| HK40108352A (zh) | 用於表徵肿瘤的方法 | |
| Kim et al. | Low-level brain somatic mutations in exonic regions are collectively implicated in autism with germline mutations in autism risk genes | |
| Beaudin | The nosology of hereditary cerebellar ataxias: development of a classification for recessive ataxias and phenotypical description of Spinocerebellar ataxia 34 | |
| Xiang et al. | Transcriptomic Analysis of Circular RNA as Potential Biomarkers in Diagnosing Atopic Dermatitis | |
| EP4590858A1 (fr) | Procede de caracterisation de la degradation d'un organe | |
| FR3150527A1 (fr) | Procede pour determiner le pronostic de survie d’un patient atteint d’un cancer du pancreas | |
| FR3158742A1 (fr) | Biomarqueurs et outil pronostique pour les patients atteint de lymphomes b diffus a grandes cellules | |
| Fu et al. | Transcriptome Analysis Reveals Auto-Inflammatory Genes of Chronic Nonbacterial Osteomyelitis (CNO) Based on The Machine Learning | |
| WO2026094720A1 (fr) | Procédé de génération de modèle de détermination de propriété, modèle de détermination de propriété, procédé de détermination de propriété, dispositif de détermination de propriété, programme et support d'enregistrement | |
| CN121506252A (zh) | 一种无创预测乳腺癌分子分型的方法及其应用 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: UNKNOWN |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: THE INTERNATIONAL PUBLICATION HAS BEEN MADE |
|
| PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
| STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE |
|
| 17P | Request for examination filed |
Effective date: 20240222 |
|
| AK | Designated contracting states |
Kind code of ref document: A1 Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR |
|
| DAV | Request for validation of the european patent (deleted) | ||
| DAX | Request for extension of the european patent (deleted) | ||
| RIN1 | Information on inventor provided before grant (corrected) |
Inventor name: BAUCHET, LUC Inventor name: RELIER, SEBASTIEN Inventor name: DAVID, ALEXANDRE Inventor name: HIRTZ, CHRISTOPHE Inventor name: RIVALS, ERIC |