Classification of severe hypercholesterolaemias

Abstract

The invention relates to determining a hypercholesterolaemia profile for a patient based on (i) a first joint statistical distribution as a function of a quantile scale (S_SPG) of polygenic risk score levels and of a quantile scale (S_LDL) of cholesterol concentration levels in low-density lipoproteins, the first joint statistical distribution giving, as a function of these first and second scales, a distribution of individuals of a first sub-sample of severely hypercholesterolaemic individuals having at least one genetic mutation at the origin of a hypercholesterolaemia score for at least one gene among a set of genes and (ii) a second joint statistical distribution as a function of the first quantile scale (S_SPG) of polygenic risk score levels and of the second quantile scale (S_LDL) of cholesterol concentration levels in the low-density lipoproteins, the second joint statistical distribution giving, as a function of these scales, a distribution of individuals of a second sub-sample of severely hypercholesterolaemic individuals not having a genetic mutation in the set of genes.

Description

CLASSIFICATION OF SEVERE HYPERCHOLESTEROLEMIA TECHNICAL FIELD

[0001] Various exemplary embodiments generally relate to a method for determining a hypercholesterolemic profile for a patient suffering from severe hypercholesterolemia and an associated device. TECHNICAL BACKGROUND

[0002] Hereditary hypercholesterolemia, known as familial hypercholesterolemia (FH) of monogenic origin (autosomal dominant / recessive) is linked to the presence of rare mutations in patients (“mutated”) and currently constitutes the only demonstrated etiology of severe hypercholesterolemia.

[0003] There is, however, a notable proportion of forms of severe hypercholesterolemia which appear to be hereditary but which do not present a mutation ("non-mutated", non-monogenic forms).

[0004] This leads to the consideration of other etiologies, notably polygenic. To date, the prevalence of these different forms has still not been clearly assessed.

[0005] Strategies for identifying HF patients were initially based on clinico-biological criteria such as those of the Dutch Lipid Clinic Network (DLCN) first described in 1998.

[0006] Fig. 1A illustrates a classification currently used in Europe and worldwide based on the DLCN criteria for genetic screening of FH.

[0007] According to this method, the threshold value for genetic screening of severe hypercholesterolemia was proposed, in the absence of treatment, at 1.9 gL -1 of LDLc, and this value also defines the lower limit of severe hypercholesterolemia.

[0008] The concentration, here denoted LDLc, corresponds to the concentration of cholesterol in low-density lipoproteins (LDL cholesterol, LDLc), for example the measurement in a biological sample, preferably a blood sample. In this document, circulating concentration means the concentration measured in a blood sample.

[0009] This LDLc level corresponds to 90 ème percentile of the distribution of circulating LDLc concentrations in the general French adult population. Since the prevalence of monogenic FH (mutated individuals) is estimated at 1 / 300 (i.e. 0.33%), this means that a majority of so-called severe hypercholesterolemia (LDLc s 1.9 g.L 1 ) are not of monogenic origin.

[0010] It is desirable to improve the situation and provide a patient classification strategy taking into account different origins of severe hypercholesterolemia. SUMMARY

[0011] The scope of protection is defined by the independent claims. Embodiments, examples and features, if any, described in this specification that are not covered by the protection are to be construed as examples useful for understanding the various embodiments or examples that are covered by the protection.

[0012] According to a first aspect, an in vitro method for determining a hypercholesterolemic profile for a severely hypercholesterolemic patient is described. The method is computer-implemented and comprises: - obtaining a first joint statistical distribution giving a distribution of individuals according to the combination of a first scale in quantiles of polygenic risk score levels and a second scale in quantiles of cholesterol concentration levels in low-density lipoproteins, the first joint statistical distribution giving the distribution, according to the first and second scales, of the individuals of a first sub-sample of severely hypercholesterolemic individuals presenting at least one genetic mutation causing hypercholesterolemia for at least one gene among a set of genes; - obtaining a second joint statistical distribution giving a distribution of individuals according to the combination of the first scale in quantiles of levels of the polygenic risk score and the second scale in quantiles of levels of cholesterol concentration in low-density lipoproteins, the second joint statistical distribution giving the distribution, according to the first and second scales, of the individuals of a second sub-sample of severely hypercholesterolemic individuals not presenting a genetic mutation in the set of genes; - obtaining, by reading from a memory, data determined from at least one biological sample for the patient, the data comprising: o genetic information on the presence or absence of at least one genetic mutation causing hypercholesterolemia for at least one gene among the set of genes; o a polygenic risk score; o a cholesterol concentration in low-density lipoproteins in the absence of treatment; - a determination, for the patient, of a position in one of the joint statistical distributions based on the polygenic risk score and the cholesterol concentration in low-density lipoproteins, the position being determined in the first joint statistical distribution when the genetic information obtained for the patient indicates the presence of at least one mutation and in the second distribution joint statistics when the genetic information obtained for the patient indicates the absence of mutation; - a determination of the hypercholesterolemic profile of the patient as a function of the determined position; the first and second joint statistical distributions giving distribution values ​​of the individuals according to a third scale, the second scale of cholesterol concentration levels in low-density lipoproteins being obtained on the basis of a quantile distribution of the cholesterol concentrations in low-density lipoproteins obtained for the individuals of the total population consisting of the first and second sub-samples of individuals; the first scale of polygenic risk score levels being obtained on the basis of a quantile distribution of the polygenic risk scores obtained for the individuals of the first sub-sample of individuals.

[0013] According to one or more embodiments, the determination of the hypercholesterolemic profile of the patient comprises a determination of an index corresponding to the given value, according to the third scale, at the position determined for the patient, either by the first joint statistical distribution when the genetic information obtained for the patient indicates the presence of at least one mutation, or by the second joint statistical distribution when the genetic information obtained for the patient indicates the absence of mutation.

[0014] According to one or more embodiments, determining the hypercholesterolemic profile of the patient comprises determining a class to which the patient belongs based on the obtained position, the class being - either a first class when the genetic information indicates the presence of a genetic mutation; - or a second class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint statistical distribution presenting a value greater than or equal to a first threshold in the third scale; - or a third class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint statistical distribution indicate a value greater than or equal to a second threshold in the third scale and strictly less than the first threshold in the third scale; - or a fourth class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint distribution indicate a value strictly lower than the second threshold in the third scale.

[0015] According to one or more embodiments, the first class corresponds to a hypercholesterolemic profile with confirmed so-called monogenic heredity, in which the second class corresponds to a hypercholesterolemic profile with so-called polygenic heredity, in which the third class corresponds to a hypercholesterolemic profile with so-called undetermined heredity, in which the fourth class corresponds to a hypercholesterolemic profile with so-called monogenic heredity of unidentified cause.

[0016] According to one or more embodiments, the first joint statistical distribution is a cumulative statistical distribution obtained by generalized additive model type regression from an initial statistical distribution giving a distribution of the individuals of the first sub-sample of individuals according to the combination of the first and second quantile scales obtained.

[0017] According to one or more embodiments, the second joint statistical distribution is a cumulative statistical distribution obtained by generalized additive model type regression from an initial statistical distribution giving a distribution of the individuals of the second sub-sample of individuals according to the combination of the first and second quantile scales obtained.

[0018] According to one or more embodiments, the second scale is a quantile scale whose quantiles (deciles or percentiles) are calculated on the basis of a statistical distribution of cholesterol concentrations in low-density lipoproteins obtained from biological samples for individuals in the total population consisting of the first and second sub-samples of individuals.

[0019] According to one or more embodiments, the first scale is a quantile scale whose quantiles (deciles or percentiles) are calculated on the basis of the statistical distribution of polygenic risk scores obtained from biological samples for the individuals of the first subsample of individuals.

[0020] According to one or more embodiments, the first and second joint statistical distributions provide distribution values ​​of individuals according to a third scale in quantiles (for example, in deciles or percentiles).

[0021] According to one or more embodiments, the second threshold corresponds to 2 ème (or 3 ème ) decile and the first threshold corresponds to 6 ème (or 7 ème ) decile when the third scale is in deciles. According to one or more embodiments, the second threshold corresponds to H ème (or 21 ième ) percentile and the first threshold corresponds to the 5th ème (or 61 ème ) percentile when the third scale is in percentiles. The threshold values ​​are likely to vary according to the sub-samples of severely hypercholesterolemic individuals considered for obtaining the first and second distributions.

[0022] According to one or more embodiments, the set of genes comprises at least: LDLR, APOB, PCSK9, APOE.

[0023] According to one or more embodiments, the determination of the hypercholesterolemic profile of the patient comprises: a prediction of a level of risk of cardiovascular accident for the patient on the basis of the position obtained in a third joint statistical distribution, a function of the first and second scales, the third joint statistical distribution giving, as a function of these first and second scales, a level of risk of cardiovascular accident.

[0024] According to one or more embodiments, the risk level is calculated in the third statistical distribution as a median value of the CAC score, Coronary Artery Calcium, of a statistical distribution of the CAC scores obtained for the subset of individuals having the same level of polygenic risk score as the patient and the same level of cholesterol concentration in low-density lipoproteins as the patient, this subset of individuals being either a subset of the first subsample of individuals when the genetic information obtained for the patient indicates the presence of at least one mutation, or a subset of the second subsample of individuals when the genetic information obtained for the patient indicates the absence of a mutation.

[0025] According to another aspect, a device comprises means for implementing a method according to the first aspect. The device may be a computing device.

[0026] The device may comprise software and / or hardware means for performing one or more or all of the steps of the method according to the first aspect. These means may comprise at least one processor and at least one memory storing program instructions which, when executed by at least one processor, cause the device to perform one or more or all of the steps of the method according to the first aspect. These means may comprise electronic circuits (e.g., data processing circuits) for performing one or more or all of the steps of the method according to the first aspect or of another method described herein. The electronic circuits may be generic circuits programmed for the execution of such a method or dedicated circuits, specifically configured for the execution of such a method.

[0027] According to another aspect, a device comprises at least one processor and at least one memory storing instructions which, when executed by at least one processor, cause the device to execute one or more or all of the steps of a method according to the first aspect or another method described herein.

[0028] In another aspect, a computer program comprises instructions which, when performed by a device, cause the device to perform one or more or all of the steps of a method according to the first aspect or of another method described herein.

[0029] In another aspect, a non-transitory computer-readable medium includes program instructions stored therein to enable a device to perform one or more or all of the steps of a method according to the first aspect or another method described herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The embodiments will be better understood in light of the detailed description below and the accompanying drawings, which are given for illustration purposes only and are therefore not limiting of the present disclosure.

[0031] Fig. 1A, already described, illustrates a classification currently used worldwide according to the WHO recommendations (WHO / HGN / FH / CONS / 99.2) based on the DLCN criteria for genetic screening of FH as an example.

[0032] Fig. 1 B shows the risk of (CAD, for Coronary Artery Disease) as a function of LDLc concentration according to an example.

[0033] Fig. 2 shows a statistical distribution of polygenic score (PG score) values ​​for “mutated” and “non-mutated” individuals according to an example.

[0034] Fig. 3 represents the variation in the number of HF screening prescriptions and the percentage of genetically confirmed HF patients.

[0035] Fig. 4 shows the distribution of “mutated” and “non-mutated” patients according to LDLc decile values ​​according to an example.

[0036] Fig. 5 shows the distribution of “mutated” and “non-mutated” patients according to the PG Score deciles according to an example.

[0037] Figs. 6 to 11 represent each of the joint statistical distributions obtained for a subsample of “mutated” patients (left) and a subsample of “non-mutated” patients (right) as a function of their LDLc and PG Score values ​​according to an example.

[0038] Fig. 12 illustrates the method of obtaining classes based on the joint statistical distributions of Figures 8 and 9.

[0039] Fig. 13 shows a flowchart of a classification method according to one or more exemplary embodiments.

[0040] It should be noted that these drawings are intended to illustrate various aspects of the devices, methods and structures used in the exemplary embodiments described herein. The use of similar or identical reference numerals in the various figures is intended to indicate the presence of a similar or identical element or feature. DETAILED DESCRIPTION

[0041] Detailed exemplary embodiments are presented herein. However, the specific structural and / or functional details disclosed herein are merely representative for the purpose of describing the exemplary embodiments and clearly understanding the underlying principles. However, these exemplary embodiments may be practiced without these specific details. These exemplary embodiments may be embodied in many other forms, with various modifications, and should not be construed as being limited to only the embodiments set forth herein. In addition, the figures and descriptions may have been simplified to illustrate elements and / or aspects relevant to a proper understanding of the present invention, while eliminating, for the sake of clarity, many other elements which may be well known in the art or not relevant to an understanding of the invention.

[0042] In this document, by "patient" or "subject", is meant any individual likely to have, for example, hypercholesterolemia, for example severe hypercholesterolemia, it can be for example a human regardless of their sex and / or age. It can be for example a human, for example an adolescent, an adult. It can be for example a human aged from 12 years to 99 years. It can be for example an adolescent with an age between 12 and 17 years, an adult with an age between 18 and 99 years, preferably an adult with an age between 18 and 99 years. In this document, the term "individual" or "patient" will be used interchangeably.

[0043] In this document, hypercholesterolemia or hyperlipidemia means a pathology in which the blood concentration of lipids, for example cholesterol, triglycerides and / or phospholipids, is higher than a reference value in a healthy subject. This may be, for example, a pathology in which the total cholesterol concentration is higher than a reference threshold value in a healthy subject. This may be, for example, hypercholesterolemia or hyperlipidemia as mentioned in the bibliographic reference “Hyperlipidemia in coronary heart disease. II. Genetic analysis of lipid levels in 176 families and delineation of a new inherited disorder, combined hyperlipidemia” JL Goldstein, HG Schrott, WR Hazzard, EL Bierman, AG Motulsky J Clin Invest. 1973 Jul;52(7): 1544-68. PMID: 4718953 or “Familial Hypercholesterolemia”. Levenson AE, de Ferranti SD. 2023 Nov 28.In: Feingold KR, Anawalt B, Blackman MR, Boyce A, Chrousos G, Corpas E, de Herder WW, Dhatariya K, Dungan K, Hofland J, Kalra S, Kaltsas G, Kapoor N, Koch C, Kopp P, Korbonits M, Kovacs CS, Kuohung W, Laferrère B, Levy M, McGee EA, McLachlan R, New M, Purnell J, Sahay R, Shah AS, Singer F, Sperling MA, Stratakis CA, Trence DL, Wilson DP, editors. Endotext [Internet], South Dartmouth (MA): MDText.com, Inc.; 2000-, PMID: 27809433.

[0044] In this document, by hypercholesterolemic “patient” or “individual” severe means an individual, subject or patient, for whom, in the absence of treatment for hypercholesterolemia, the concentration of cholesterol in low-density lipoproteins (LDL cholesterol, LDLc), measured in a biological sample, is higher than the threshold for severe hypercholesterolemia, set at 1.9 gL -1 This threshold corresponds to 90 epercentile of the French population. This is also the threshold for Possible FH according to the DLCN (Dutch Lipid Clinic Network).

[0045] In this document, total cholesterol blood concentration or total cholesterol concentration in a blood sample means the total cholesterol blood concentration comprising the sum of cholesterol concentration in very low density lipoproteins (VLDL), low density lipoproteins (LDL) and high density lipoproteins (HDL). For example, the reference threshold value for total cholesterol concentration is 200 mg.dL -1 For example, the reference threshold value for LDL cholesterol concentration is 130 mg.dL -1 For example, the reference threshold value for HDL cholesterol concentration is 40 mg.dL -1 For example, the reference threshold value for triglyceride concentration is 150 mg.dL -1. For example, reference concentrations of total cholesterol, LDL cholesterol, HDL cholesterol and triglycerides are preferably: total cholesterol: less than 200 mg.dL -1 , LDL cholesterol: less than 130 mg.dL -1 , HDL cholesterol: greater than 40 mg.dL -1 , and triglycerides: less than 150 mg.dL -1

[0046] Herein, the determination of the concentration of cholesterol in low-density lipoproteins (LDL) may be carried out in a biological sample by any suitable method known to the person skilled in the art. This may be, for example, the method described in the document "Methods for measurement of LDL-cholesterol: a critical assessment of direct measurement by homogeneous assays versus calculation." Nauck M, Warnick GR, Rifai N. Clin Chem. 2002 Feb;48(2):236-54. PMID: 11805004.

[0047] As used herein, "biological sample" means any biological fluid, for example, it may be a sample of blood, plasma, serum, synovial fluid, skin, etc. It may preferably be a blood sample.

[0048] Herein, the determination of the cholesterol concentration in low-density lipoproteins (LDL) may be carried out from a biological sample by any suitable method known to the person skilled in the art. For example, the cholesterol concentration in low-density lipoproteins may be measured by any suitable method and may be expressed, for example, in mg.dL -1 , or gL -1 or in mmol.L -1 .

[0049] The concentration noted here as LDLc corresponds to the concentration of cholesterol in low-density lipoproteins (LDL cholesterol, LDLc).

[0050] In this document, by genetic mutation associated with and / or inducing hypercholesterolemia, we mean any mutation of at least one gene capable of modifying the blood concentration of lipids, for example cholesterol and / or triglycerides of a subject or patient. It may be, for example, a genetic mutation of at least one gene from a set of genes. Herein, the gene may be any gene known to the person skilled in the art in which at least one mutation is likely to modify the blood concentration of lipids. Herein, the set of genes may comprise at least one gene chosen from the genes LDLR, APOB, PCSK9, APOE. It may be, for example, at least one mutation in at least one gene chosen from LDLR, APOB, PCSK9, APOE.

[0051] In this document, the term LDLR gene refers to the gene coding for the low-density lipoprotein receptor (LDLR). This is the LDLR gene with the sequence reference NCBI NM_000527.5 (https: / / www.ncbi.nlm.nih.gOv / nucleotide / NM_000527.5)

[0052] In this document, by mutation of the LDLR gene is meant any mutation known to the person skilled in the art capable of modifying the structure and / or biological activity of the low-density lipoprotein receptor. This may be at least one mutation of the LDLR gene as described in "ClinVar database of global familial hypercholesterolemia-associated DNA variants". lacocca MA, Chora JR, Carrié A, Freiberger T, Leigh SE, Defesche JC, Kurtz CL, DiStefano MT, Santos RD, Humphries SE, Mata P, Jannes CE, Hooper AJ, Wilemon KA, Benlian P, O'Connor R, Garcia J, Wand H, Tichy L, Sijbrands EJ, Hegele RA, Bourbon M, Knowles JW; ClinGen FH Variant Curation Expert Panel. Hum Mutât. 2018 Nov;39(11): 1631-1640. doi: 10.1002 / humu.23634.PMID: 30311388; PMCID: PMC6206854 et / ou identifiée selon le procécé décrit dans “Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.” Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, Rehm HL; ACMG Laboratory Quality Assurance Committee. Genet Med. 2015 May;17(5):405-24. doi: 10.1038 / gim.2015.30. Epub 2015 Mar 5. PMID: 25741868; PMCID: PMC4544753 et / ou “The Clinical Genome Resource (ClinGen) Familial Hypercholesterolemia Variant Curation Expert Panel consensus guidelines for LDLR variant classification.''Chora JR, lacocca MA, Tichÿ L, Wand H, Kurtz CL, Zimmermann H, Leon A, Williams M, Humphries SE, Hooper AJ, Trinder M, Brunham LR, Costa Pereira A, Jannes CE, Chen M, Chonis J, Wang J, Kim S, Johnston T, Soucek P, Kramarek M, Leigh SE, Carrié A, Sijbrands EJ, Hegele RA, Freiberger T, Knowles JW, Bourbon M; ClinGen Familial Hypercholesterolemia Expert Panel. Genet Med. 2022 Feb;24(2):293-306. PMID: 34906454. This may for example be a mutation in the LDLR gene as described in the Leiden Open Variation Database (LOVD) (https: / / databases.lovd.nl / shared / genes / LDLR). This could be, for example, a mutation of the LDLR gene as described in the Clingen database (https: / / erepo.clinicalgenome.org / evrepo / ui / summary / classifications?columns=gene&values=LDLR&matchTypes=exact&pgSize=25).

[0053] In this document, the term APOB gene refers to the gene coding for apolipoprotein B. This is the APOB gene with sequence reference NCBI NM_000384.3 (https: / / www.ncbi.nlm.nih.gOv / nucleotide / NM_000384.3).

[0054] In this document, APOB gene mutation means any mutation known to the person skilled in the art that may modify the structure and / or biological activity of apolipoprotein B. This may be at least one APOB gene mutation as described in the “ClinVar database of global familial hypercholesterolemia-associated DNA variants”. lacocca MA, Chora JR, Carrié A, Freiberger T, Leigh SE, Defesche JC, Kurtz CL, DiStefano MT, Santos RD, Humphries SE, Mata P, Jannes CE, Hooper AJ, Wilemon KA, Benlian P, O'Connor R, Garcia J, Wand H, Tichy L, Sijbrands EJ, Hegele RA, Bourbon M, Knowles JW; ClinGen FH Variant Curation Expert Panel. Hum Mutât. 2018 Nov;39(11): 1631-1640. doi:10.1002 / humu.23634. PMID: 30311388; PMCID: PMC6206854 and / or identified according to the process described in “Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.” Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, Rehm HL; ACMG Laboratory Quality Assurance Committee. Genet Med. 2015 May;17(5):405-24. doi: 10.1038 / gim.2015.30. Epub 2015 Mar 5. PMID: 25741868; PMCID: PMC4544753. This could be, for example, a mutation in the APOB gene as described in the LOVD database (https: / / databases.lovd.nl / shared / genes / APOB).

[0055] In this document, the PCSK9 gene is understood to mean the gene coding for the proprotein convertase subtilisin / kexin type 9. This is the PCSK9 gene with the sequence reference NCBI NM_174936.4 (https: / / www.ncbi.nlm.nih.gov / nucleotide / NM_174936.4)

[0056] In this document, by mutation of the PCSK9 gene is meant any mutation known to the person skilled in the art likely to modify the structure and / or biological activity of the proprotein convertase subtilisin / kexin type 9.

[0057] It may be at least one mutation in the PCSK9 gene as described in “ClinVar database of global familial hypercholesterolemia-associated DNA variants”. lacocca MA, Chora JR, Carrié A, Freiberger T, Leigh SE, Defesche JC, Kurtz CL, DiStefano MT, Santos RD, Humphries SE, Mata P, Jannes CE, Hooper AJ, Wilemon KA, Benlian P, O'Connor R, Garcia J, Wand H, Tichy L, Sijbrands EJ, Hegele RA, Bourbon M, Knowles JW; ClinGen FH Variant Curation Expert Panel. Hum Mutât. 2018 Nov;39(11): 1631 -1640. doi:10.1002 / humu.23634. PMID: 30311388; PMCID: PMC6206854 and / or identified according to the process described in “Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.” Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, Rehm HL; ACMG Laboratory Quality Assurance Committee. Genet Med. 2015 May;17(5):405-24. doi: 10.1038 / gim.2015.30. Epub 2015 Mar 5. PMID: 25741868; PMCID: PMC4544753. This may be, for example, a mutation of the PCSK9 gene as described in the LOVD database (https: / / databases.lovd.nl / shared / genes / PCSK9 or “Mutations and polymorphisms in the proprotein convertase subtilisin kexin 9 (PCSK9) gene in cholesterol metabolism and disease”, Abifadel M, Rabès JP, Devillers M, Munnich A, Erlich D, Junien C, Varret M, Boileau C. Hum Mutât 2009 Apr;30(4):520-9. doi: 10.1002 / humu.20882. PMID: 19191301.

[0058] In this document, the term APOE gene means the gene coding for apolipoprotein E. This is the APOE gene with the sequence reference NCBI NM_000041 .4 https: / / www.ncbi.nlm.nih.gov / nucleotide / NM_000041 .4)

[0059] In this document, APOE gene mutation means any mutation known to the person skilled in the art that may modify the structure and / or biological activity of apolipoprotein E. It may be at least one APOE gene mutation as described in the “ClinVar database of global familial hypercholesterolemia-associated DNA variants”. lacocca MA, Chora JR, Carrié A, Freiberger T, Leigh SE, Defesche JC, Kurtz CL, DiStefano MT, Santos RD, Humphries SE, Mata P, Jannes CE, Hooper AJ, Wilemon KA, Benlian P, O'Connor R, Garcia J, Wand H, Tichy L, Sijbrands EJ, Hegele RA, Bourbon M, Knowles JW; ClinGen FH Variant Curation Expert Panel. Hum Mutât. 2018 Nov;39(11): 1631-1640. doi:10.1002 / humu.23634. PMID: 30311388; PMCID: PMC6206854 and / or identified according to the process described in “Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.” Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, Grody WW, Hegde M, Lyon E, Spector E, Voelkerding K, Rehm HL; ACMG Laboratory Quality Assurance Committee. Genet Med. 2015 May;17(5):405-24. doi: 10.1038 / gim.2015.30. Epub 2015 Mar 5. PMID: 25741868; PMCID: PMC4544753. This could be, for example, a mutation in the APOE gene as described in the LOVD database, https: / / databases.lovd.nl / shared / genes / APOE.

[0060] Herein, obtaining genetic information on the presence of a genetic mutation may be carried out by any suitable method known to the person skilled in the art. This may be, for example, a method comprising, from a biological sample, for example a blood sample, determining and / or detecting a mutation in a gene known to the person skilled in the art. This may be, for example, a method chosen from restriction fragment length polymorphism (RFLP), single-strand conformation polymorphism (SSCP), denaturing gradient gel electrophoresis (DGGE), polymerase chain reaction (PCR) and Sanger sequencing, high-resolution melting curve analysis (HRM), next-generation sequencing (NGS), also referred to as high-throughput sequencing or massive sequencing parallel, and / or long-read sequencing or the MLPA (Multiplex Ligation Dependent Probe Amplification) technique, which can highlight large rearrangements. This could be, for example, high-throughput sequencing as described in "An Overview of DNA Analytical Methods." Arboleda VA, Xian RR. Methods Mol Biol. 2019;1897:385-402. doi: 10.1007 / 978-1-4939-8935-5_31. PMID: 30539459

[0061] In this document, we will discuss: • “mutated” patient or individual to mean that at least one genetic mutation of at least one gene causing hypercholesterolemia has been identified by genetic analysis (for example by sequencing technology); • “non-mutated” patient or individual to mean that no genetic mutation causing hypercholesterolemia has been identified by genetic analysis;

[0062] Furthermore, we will speak of monogenic hypercholesterolemia for a "mutated" patient suffering from hypercholesterolemia, regardless of the number of mutated gene(s) and / or the number of mutations observed.

[0063] One or more exemplary embodiments relate to a method for classifying patients with severe hypercholesterolemia and an associated device.

[0064] The classification process is based on a statistical analysis carried out on a population of individuals with severe hypercholesterolemia.

[0065] Monogenic FH is associated with a high risk of coronary artery disease (CAD). Indeed, this risk has been shown to be higher in “mutated” patients compared to “non-mutated” patients, as shown in Fig. 1 B.

[0066] Fig. 1 B compares the risk of CAD as a function of LDLc concentration in “mutated” versus “non-mutated” patients for a range of LDLc concentrations.

[0067] In Fig. 1 B, we see that, for concentrations above the genetic screening threshold, the risk of CAD is much higher in “mutated” patients compared to “non-mutated” patients, with a multiplicative factor greater than 3.

[0068] Strategies for identifying HF patients based on clinico-biological criteria such as those of the Dutch Lipid Clinic Network (DLCN) have been implemented and present the limitations mentioned in the introduction with reference to Fig. 1A.

[0069] The threshold value for genetic screening of severe hypercholesterolemia is currently set, in the absence of treatment, at 1.9 gL -1 of LDLc in a blood sample. This LDLc level also corresponds to the 90e percentile of the distribution of serum LDLc concentrations in the general French population. Since the prevalence of monogenic FH (mutated individuals) is estimated at 1 / 300 (i.e. 0.33%), this means that a majority of so-called severe hypercholesterolemia (LDLc s 1.9 gL -1 ) are not of monogenic origin.

[0070] A portion of severe hypercholesterolemia should therefore correspond to polygenic forms as suggested by Talmud et al. in

[0001] , by proposing the calculation of a polygenic score (PG Score). This score takes into account a combination of genetic variants. For example, the PG score can advantageously make it possible to determine, based on the value of the score obtained, the origin of the genetic variation, for example monogenic or polygenic.

[0071] In this document, a polygenic score (PG Score), also referred to as a polygenic risk score (or PRS), is understood to mean, for example, a numerical value derived from a genomic analysis. It can be used to predict the probability that an individual will develop a certain disease or characteristic based on their genetic profile. This may be the PG score as described in

[0001] .

[0072] Herein, obtaining a polygenic score (PG Score), also called a polygenic risk score, can be carried out by any suitable method known to the person skilled in the art. The polygenic score can be obtained from any biological sample known to the person skilled in the art for analyzing genomic DNA. This can be, for example, a sample of blood, saliva, appendages, hair, skin and / or cutaneous tissue. The polygenic score (PG Score) can be obtained / determined, for example, by sequencing genomic DNA, for example according to the method as described in

[0001] .

[0073] The PG score can be used to predict the likelihood of an individual developing hypercholesterolemia, this score advantageously allows determining a risk of hypercholesterolemia, for example polygenic.

[0074] The discriminatory nature of the PG score is however difficult to specify as evidenced by the overlap of the distribution curves of the values ​​of this score between “mutated” and “non-mutated” individuals represented in Fig. 2.

[0075] Fig. 2 shows the statistical distribution of PG Score values ​​of severely hypercholesterolemic individuals suspected of familial hypercholesterolemia (FH), on the one hand for “mutated” individuals (FH / M+) and on the other hand for “non-mutated” individuals (FH / M-) (source: cf

[0001] ). For “mutated” individuals (FH / M+), the mean score is 0.95 and the standard deviation is 0.20. For “non-mutated” individuals (FH / M-), the mean score is 1.00 and the standard deviation is 0.21.

[0076] Molecular diagnosis of HF can be performed using new sequencing technologies (NGS). It is thus possible to identify rare mutations causing monogenic forms (autosomal dominant / recessive) but also to calculate the PG Score in order to assess the share of the polygenic component of these severe hypercholesterolemias.

[0077] Furthermore, a sharp increase from year to year in the number of prescriptions for severe hypercholesterolemia was observed while a steady decline of the percentage of HF confirmed by mutation identification occurs over the same period.

[0078] This phenomenon is illustrated by Fig. 3 which represents the number of HF screening prescriptions (left scale, vertical bars) and the percentage of patients (right scale, curve) genetically confirmed HF by identification of rare mutations.

[0079] These opposing trends are largely explained by the increase in the proportion of non-monogenic and non-hereditary forms among the growing requests for analysis addressed to structures carrying out the genetic diagnosis of FH. This is probably linked to a certain inadequacy of the criteria used.

[0080] Indeed, these clinical-biological criteria, derived from the DLCN classification, were described 26 years ago (1998), when genetic diagnosis was not very developed and screening for HF was carried out in expert centers and by a few specialists.

[0081] However, this screening is now offered by many more diverse structures, in particular to respond to an increase in demand which is justified by the need for proof of monogenic HF to be able to access certain innovative and expensive hypocholesterolemic therapies.

[0082] In any case, one of the deleterious effects observed is the increase in the number of negative genetic results that do not contribute in any way to patient care. These negative results, on the contrary, generate a feeling of failure of this diagnosis without definitively eliminating the part of a hereditary component in the severe hypercholesterolemia of these patients.

[0083] Currently, no molecular data on severe hypercholesterolemia in France are available. Consequently, it appears that it is not possible to clearly identify the different forms of heredity (monogenic versus polygenic) and non-hereditary forms, nor to measure their possible difference in distribution according to LDLc values ​​or their respective level of cardiovascular risk.

[0084] In response to these gaps, an analysis of molecular data from a cohort of severely hypercholesterolemic patients was performed and made it possible to generate the joint distribution of untreated LDLc concentrations, on the one hand, and PG scores, on the other hand for this cohort of individuals. On this basis, it is possible to • determine a range of untreated LDLc values ​​in a population of severely hypercholesterolemic individuals comprising both “mutated” and “non-mutated” individuals in order to determine the levels (e.g., deciles) of untreated LDLc allowing to specify the percentage of “mutated” and “non-mutated” individuals according to these deciles, on the one hand; • determine a range of PG Score values ​​in the subsample comprising only “mutated” severely hypercholesterolemic individuals in order to define the levels (e.g., deciles) of PG Score allowing to describe the distribution of “non-mutated” patients, on the other hand.

[0085] From these values ​​(untreated LDLc deciles and PG score in a cohort of severely hypercholesterolemic individuals), the joint statistical distribution of untreated LDLc concentrations and PG scores of "mutated" and "non-mutated" individuals was generated and analyzed. Based on this new statistical distribution, a method for classifying patients is introduced, which is based on their heredity, whether monogenic, polygenic, or undetermined.

[0086] The patient classification method allows: • to identify the most distinctive clinical-biological characteristics of these forms of heredity in order to improve the selection of patients who are candidates for genetic screening for HF and to propose a redefinition of the “DLCN” type criteria in the population of severe hypercholesterolemia. • to modify the methods of rendering results by specifying more clearly the probable polygenic origin of HF; • to study the respective cardiovascular risk of these different forms, which makes it possible to improve their management to prevent the occurrence of cardiovascular events.

[0087] Study population and methods

[0088] Since 2015, 4358 patients (level “possible HF” according to DLCN corresponding to a maximum LDLc value without treatment > 1.9 gL -1 ) were routinely analyzed by NGS technique (ADH MASTR v2, Agilent / Multiplicom NV) allowing (i) to identify mutations in the coding regions of LDLR, PCSK9, APOE, and the APOB binding domain and (ii) to calculate a polygenic score (PG Score) based on 12 SNPs (Single Nucleotide Polymorphism) (GS12; cf.

[0001] ).

[0089] Among this initial population of 4358 patients, a selection of 3188 index and related cases was studied for whom lipid assessments without treatment were available and who met the following selection criteria: maximum LDLc without treatment s 1.9 gL -1 , triglycerides < 4 gL -1 and age 18 years.

[0090] These 3188 patients, representative of the severe hypercholesterolemic population, were screened for HF in a structure carrying out genetic diagnosis in France and the results obtained were as follows: • 1574 did not present any molecular abnormalities (“non-mutated”). • 1407 presented a molecular anomaly (“mutated”), including o 1275 in the LDLR gene o 97 in the APOB gene o 26 in the PCSK9 gene o 9 in the APOE gene • 207 presented a molecular anomaly of unknown significance which did not allow not taken into account and were not considered in the generation of the joint statistical distribution.

[0091] In total, the results presented in Figures 4 to 11 correspond to the analysis of a sub-sample of 1407 “mutated” individuals and a sub-sample of 1574 “non-mutated” individuals, the cohort studied ultimately being made up solely of these two sub-samples.

[0092] For 1355 of them (667 "mutated" and 688 "non-mutated"), the CAC score (Coronary Artery Calcium) value was available and reflects the involvement of their coronary arteries and their risk of CAD; for 2022 of them (1061 "mutated" and 961 "non-mutated"), lipid profiles under treatment were available.

[0093] The proportions of "mutated" and "non-mutated" individuals were estimated by decile of LDLc, PG Score and by crossing (joint distribution) these two ordinal variables. A statistical model was developed to analyze the density of "mutated" and "non-mutated" patients according to the joint distribution of LDLc and PG Score.

[0094] For this, the entire sample was divided into subsets of individuals according to: (i) quantiles (e.g., deciles) of LDLc concentration (based on the distribution of LDLc concentration values ​​across the entire cohort studied, consisting of a first subsample of “mutated” individuals and a second subsample of “non-mutated” individuals) and (ii) quantiles (e.g., deciles) of PG Score (based on the distribution of the PG score in the first subsample of “mutated” individuals only).

[0095] Thus, the LDLc concentration quantiles were determined for the entire cohort studied, by dividing the range of LDLc concentration values ​​into N levels (or N sub-ranges of values), each level corresponding to a quantile, for example to a decile if N = 10, i.e. 100 / N = 10% of the cohort consisting of the sub-sample of “mutated” individuals and the sub-sample of “non-mutated” individuals. Other values ​​can be used for the number N. The number N can for example be chosen equal to 12, 15, 20 or 100. For N = 100, the range of values ​​will be divided into percentiles.

[0096] As is known, the quantiles of a statistical distribution F as a function of a variable X correspond to a sequence of N values ​​of X which make it possible to divide the population (i.e. the area under the curve F) for which this distribution is established into N (> 2) equal parts and thus define sub-ranges of values. These sub-ranges are also called deciles.

[0097] The PG score quantiles were determined within the subsample of “mutated” individuals only, by dividing this range of PG score values ​​into N levels (or sub-ranges of values), each level corresponding to a quantile, for example to a decile if N = 10, i.e. 100 / N = 10% of the subsample of “mutated” individuals. This same scale was applied to the subsample of "non-mutated" individuals. Other values ​​can be used for the number N. The number N can, for example, be chosen equal to 12, 15, 20 or 100. For N = 100, the range of values ​​will be divided into percentiles.

[0098] In the implementation examples described in this document, it is assumed, for simplicity of presentation, that the number N is chosen equal to 10 for the scale of LDLc concentrations and PG scores, which results in decile scales. But these examples can be generalized to any number of levels.

[0099] The joint statistical distribution was generated for the subsample of “mutated” individuals based on the LDLc concentration deciles and the PG score deciles, each value of this joint statistical distribution corresponding to the number of “mutated” individuals for a pair of values ​​composed of an LDLc concentration decile and a PG score decile, according to the decile scales established as indicated above.

[0100] Similarly, another joint statistical distribution was generated for the subsample of “non-mutated” individuals based on the LDLc concentration deciles and the PG score deciles, with each value in this joint statistical distribution corresponding to the number of “non-mutated” individuals for a pair of values ​​consisting of an LDLc concentration decile and a PG score decile, according to the decile scales established as indicated above, so that the two joint statistical distributions for the subsample of “mutated” individuals and the subsample of “non-mutated” individuals use the same decile scales for LDLc concentration and PG score.

[0101] Each of the two joint statistical distributions was normalized relative to the total number of individuals in the sample concerned. Thus, the joint statistical distribution of the subsample of "mutated" individuals was related to the total number of individuals in this subsample, and similarly for the subsample of "non-mutated" individuals. After normalization, each point of the joint statistical distribution corresponds a proportion of "mutated" individuals and a proportion of "non-mutated" individuals (these proportions can be expressed as a percentage or as a decile or percentile) to a pair of values ​​composed of a decile of LDLc concentration and a decile of PG score.

[0102] The proportions obtained for the joint statistical distributions after normalization were modeled as a function of the LDLc concentration and PG score deciles using a generalized additive model (GAM) regression method, in order to obtain a non-linear smoothing of the calculated proportions and therefore of the joint statistical distributions after normalization. The proportions predicted by this model for each pair of LDLc concentration decile and PG score decile were ordered in ascending order and their cumulative distribution was calculated, thus making it possible to present, by means of smoothed statistical distributions (hereinafter called "cumulative distributions"), a gradient of coupled concentration decile values LDLc and PG score decile covering an increasing proportion of individuals belonging to the “mutated” and “non-mutated” samples, respectively.

[0103] Each of the two cumulative distributions thus obtained constitutes a reference statistical distribution. These statistical distributions make it possible to estimate the expected proportion of "mutated" individuals (or respectively of "non-mutated" individuals) corresponding to a set of pairs of LDLc quantile values ​​and PG score. This proportion of individuals can be expressed in quantiles (deciles or percentiles) or as a percentage of the population concerned.

[0104] These reference statistical distributions lead to the definition, for each of the “mutated” and “non-mutated” populations, of a third scale giving the proportion of individuals corresponding to each pair of LDLc concentration quantile values ​​and PG score decile.

[0105] Results

[0106] The results obtained by the method are described in more detail with regard to figures 4 to 11. In the figures and in the text, we note “Dec” for decile.

[0107] 1 ère step: definition of LDLc deciles in the cohort of severely hypercholesterolemic patients

[0108] Fig. 4 shows the statistical distribution of “mutated” and “non-mutated” patients according to LDLc decile values.

[0109] The scale of LDLc concentration values ​​is included here (Fig. 4) between 1.9 and 8.5 and divided into 10 deciles, each decile corresponding to 10% of the entire sample studied, consisting, as a reminder, of the sub-sample of "mutated" individuals and the sub-sample of "non-mutated" individuals. A statistical distribution in percentile or other quantile could be obtained on the same principle.

[0110] It should be noted that with a quantile distribution (especially decile), the sub-ranges of values ​​corresponding to the deciles can be of variable length. For example, the first sub-range corresponding to the first decile is between 1.9 and 2.07, i.e. a sub-range of range 2.07-1.9 = 0.17; the second sub-range corresponding to the second decile is between 2.07 and 2.18, i.e. a sub-range of range 2.18-2.07 = 0.11;

[0111] In this distribution, we observe that the proportion of “non-mutated” patients decreases as the LDLc concentration increases, while that of “mutated” patients increases.

[0112] This statistical distribution is used to define a quantile scale (e.g., decile scale) of cholesterol concentration levels in low-density lipoproteins. This scale will be referred to hereinafter as S_LDL and, for simplicity of presentation, it will be assumed to be a decile scale.

[0113] 2ème step: definition of the polygenic score deciles (PG Score) in the verses.

[0114] This step takes as a reference only "mutated" individuals (also called "monogenic" here). Indeed, in these patients, we can consider that their LDLc value is mainly determined by the presence of a mutation with a major functional effect and not by a predominant polygenic origin.

[0115] Fig. 5 shows the distribution of “mutated” and “non-mutated” patients according to PG Score deciles.

[0116] The scale of PG score values ​​here is between -0.003 and 1.47 and divided into 10 deciles, each decile corresponding to 10% of the subsample of "mutated" individuals. A statistical distribution in percentiles or other quantiles could be obtained on the same principle. The definition of these deciles makes it possible to show an increase in the proportion of "non-mutated" patients with the increase in the PG score.

[0117] This statistical distribution is used to define a quantile scale (e.g., decile scale) of polygenic risk score levels. This scale will subsequently be denoted S_SPG and for simplicity it will be assumed to be a decile scale.

[0118] 3 ème step: joint statistical distributions of individuals according to the decile scales of LDLc and PG Score, within the sub-samples of “mutated” patients from a

[0119] Fig. 6 represents the joint statistical distribution, in number of individuals, for the subsample of “mutated” patients (Fig. 6A) on the one hand, and the subsample of “non-mutated” patients (Fig. 6B) on the other hand, according to their LDLc and PG Score values. In these joint statistical distributions, the S_LDL and S_SPG scales in deciles, described with regard to Figs. 4 and 5, are used for the LDLc and PG Score values.

[0120] Fig. 7 represents, according to their LDLc and PG Score values: - in Fig. 7A, the normalized joint statistical distribution, in proportion to the subsample of “mutated” patients, this distribution being obtained from that of Fig. 6A; - in Fig. 7B, the normalized joint statistical distribution in proportion to the subsample of “non-mutated” patients, this distribution being obtained from that of Fig. 6B. In each of the joint statistical distributions in Fig. 7, the S_LDL and S_SPG decile scales, described in relation to Figs. 4 and 5, are used for the LDLc and PG Score values. In Fig. 7, the vertical scale giving the population proportions is expressed as a percentage, with values ​​ranging between 0 and 3% of the subsample considered.

[0121] For the "mutated" patients (Fig. 7A) the distribution is essentially a function of the LDLc value (LDL Dec 1 to LDL Dec 10). However, an effect of the PG Score is observed in this subsample. Indeed, patients in the highest decile (GS Dec 10) distribute with a higher proportion in the three highest deciles of LDLc (LDL Dec 8, 9 and 10).

[0122] For “non-mutated” patients (FIG. 7B) the distribution bias clearly shows an enrichment of the proportion of patients according to decreasing values ​​of LDLc (LDL Dec 10 towards LDL Dec 1) and increasing values ​​of PG Score (GS Dec 1 towards GS Dec 10).

[0123] 4 ème step: prediction of the cumulative distribution of individuals according to the deciles of LDLc and PG Score (statistical regression model) within the subsamples of “mutated” or “non-mutated” patients.

[0124] From each of the joint statistical distributions in Fig. 7, a cumulative distribution is predicted, based on a predictive statistical model, to serve as a reference statistical distribution. This distribution can be obtained by regression, as described in this document.

[0125] Fig. 8 represents, as a function of LDLc and PG Score values: - in Fig. 8A, the cumulative distribution (in percentiles) obtained for the joint distribution of “mutated” patients from the distribution of Fig. 7A; and - in Fig. 8B, the cumulative distribution (in percentiles) obtained for the joint distribution of “non-mutated” patients from the distribution of Fig. 7B.

[0126] In each of the cumulative statistical distributions in Fig. 8, the S_LDL and S_SPG decile scales, described in relation to Figs. 4 and 5, are used for the LDLc and PG Score values. In Fig. 8, the vertical scale giving the population proportions is expressed in percentiles, with values ​​ranging between 0 and 100% of the subsample considered.

[0127] Fig. 9 represents, as a function of LDLc and PG Score values: - in Fig. 9A, the cumulative distribution (in deciles) obtained for the distribution of patients “mutated”; and - in Fig. 9B, the cumulative distribution (in deciles) obtained for the distribution of “non-mutated”.

[0128] In each of the cumulative statistical distributions in Fig. 9, the S_LDL and S_SPG decile scales, described in relation to Figs. 4 and 5, are used for the LDLc and PG Score values. In Fig. 9, the vertical scale is a quantile scale. It is conventionally called the third scale, denoted SD_DJ if it is in deciles or SP_DJ if it is in percentiles. This third scale gives the proportions of the population and is expressed in the example of Fig. 9 in deciles, the values ​​ranging between 1 and 10 deciles relative to the subsample considered.

[0129] In this decile distribution, we clearly see a bias in the distribution of “non-mutated” individuals (Fig. 9B). Indeed, there is an over-representation of patients from the 5 highest deciles (6, Yellow to 10, dark red), or 50% of the individuals in this subsample, in 26% of the distribution (26 squares out of 100).

[0130] 5 ème step: assessment of cardiovascular risk and LDLc variation during patient follow-up

[0131] The assessment of cardiovascular risk (see Fig. 10) and LDLc variation during follow-up (see Fig. 11) can be performed for severely hypercholesterolemic patients from the predictive model in Figs. 9A and 9B of the distributions of “mutated” or “non-mutated” patients according to the given combinations of LDLc and PG Score. In each of the joint statistical distributions in Fig. 10, the same S_LDL and S_SPG decile scales, described with respect to Figs. 4 and 5, are used for the LDLc and PG Score values.

[0132] Fig. 10 represents, along the vertical axis, for each pair of LDLc and PG Score values, the median value of the statistical distribution of the CAC score (Coronary Artery Calcium) values ​​obtained for the category of patients corresponding to this pair of LDLc and PG Score values.

[0133] Furthermore, this distribution shows the relationship between these median CAC values ​​and the quantile scale (in deciles in the example of Fig. 10) giving the proportion of individuals corresponding to each pair of LDLc concentration quantile values ​​and PG score decile, proportion given by Fig. 9A or 9B for the population concerned.

[0134] This distribution is obtained based on the deciles of the distribution prediction of “mutated” (Fig. 10A, with n = 667 individuals) and “non-mutated” (Fig. 10B, with n = 688 individuals) patients according to the LDLc and PG Score values.

[0135] As used herein, the term "coronary artery calcium score" means a score obtained from coronary imaging data that assesses the amount of calcium in the walls of the blood vessels of the heart. This score can be measured by noninvasive computed tomography of the heart, advantageously without the injection of contrast material. The amount of calcium is indicated on a numerical scale. A score of zero indicates that there is no calcification in the course of the coronary arteries. For example, a coronary artery calcium score can be obtained by the method as described in Noninvasive definition of anatomy coronary artery disease by ultrafast computed tomographic scanning: a quantitative pathologic comparison study. Simons DB, Schwartz RS, Edwards WD, Sheedy PF, Breen JF, Rumberger JA. J Am Coll Cardiol. 1992 Nov 1;20(5):1118-26. doi: 10.1016 / 0735-1097(92)90367-v.PMID: 1401612 ou dans Cardiac calcification as a marker of subclinical atherosclerosis and predictor of cardiovascular events: A review of the evidence. Faggiano P, Dasseni N, Gaibazzi N, Rossi A, Henein M, Pressman G. Eur J Prev Cardiol. 2019 Jul;26(11):1191-1204. doi: 10.1177 / 2047487319830485. Epub 2019 Mar 7. PMID: 30845832 ou dans Mortensen MB,. Cainzos-Achirica M, Steffensen FH, Botker HE, Jensen JM, Sand NPR, Maeng M, Bruun JM, Blaha MJ, Sorensen HT, Pareek M, Nasir K, Norgaard BL. Association of Coronary Plaque With Low-Density Lipoprotein Cholesterol Levels and Rates of Cardiovascular Disease Events Among Symptomatic Adults.

[0136] CAC score values ​​are correlated with CAD risk. In other words, cardiovascular risk can be assessed based on the CAC score value. By analyzing their distributions within the two subsamples, we observe that in the "non-mutated" (Fig. 10B) the lowest median CAC score values ​​(lowest risk of CAD) are found in patients in the last 5 deciles (6, Yellow to 10, dark red) while patients in the 1 er decile (1, dark blue) show the highest values ​​(higher risk of CAD).

[0137] Fig. 11 shows the distribution of the means of variation in LDLc values ​​during follow-up (% reduction), this variation being evaluated between 2 moments during the patient's follow-up. The distribution is determined according to the distribution prediction deciles of "mutated" (Fig. 11 A, n = 1061 individuals) and "non-mutated" (Fig. 11 B, n = 961 individuals) patients according to the values ​​of LDLc and PG Score, with the same scales of LDLc S_LDL and PG Score S_SPG as those for the distributions in Figures 6 to 10. The distributions in Fig. 11 give, along the vertical axis, the mean value of the variation in LDLc values ​​during follow-up, averaged for the individuals corresponding to a pair of values ​​of quantiles of LDLc concentration and decile of PG score.

[0138] Furthermore, this distribution shows the relationship between these mean values ​​and the quantile scale (in deciles in the example of Fig. 11) giving the proportion of individuals corresponding to each pair of LDLc concentration quantile values ​​and PG score decile, proportion given by Fig. 9A or 9B for the population concerned.

[0139] Regarding the percentage variation of LDLc values ​​during follow-up, it is observed that the "non-mutated" patients of the 1 er decile (1, dark blue, Fig. 11 B) behave (-50% reduction) like the “mutated” patients of the highest deciles (8, dark orange to 10, dark red, Fig. 11A).

[0140] Conclusion and perspectives

[0141] Based on the results described above, a method for a new classification of severe hypercholesterolemia based on the statistical regression model (Fig. 8 and 9) is proposed.

[0142] This method leads to the individualization of classes of individuals with distinct genetic but also clinical-biological characteristics.

[0143] Fig. 12 illustrates a method for obtaining classes based on the statistical distributions of Figures 8 (with an SP_DJ scale in percentiles) and / or 9 (with an SP_DJ scale in deciles) according to an exemplary embodiment using scales in deciles for LDLc concentration (S_LDL scale) and / or PG score (S_SPG scale).

[0144] Percentile or other quantile scales can also be used for LDLc concentration (called second scale, denoted S_LDL) and / or PG score (called first scale, denoted S_SPG) and / or population proportion (third scale, denoted SD_DJ if it is in deciles or SP_DJ if it is in percentiles). To simplify the presentation, we assume that all scales are in deciles.

[0145] The method for obtaining classes is based on defining quantiles (e.g., deciles) of joint statistical distributions according to the previously described predictive models, for “mutated” and “non-mutated” individuals, respectively.

[0146] Each of the joint statistical distributions gives distribution values ​​of the individuals of the population considered according to a third scale SD_DJ, SP_DJ in quantiles.

[0147] According to these statistical distributions, approximately 10% of individuals fall into each decile (within rounding errors in the calculation of deciles). It can be expected that, in the population of patients to whom these distributions will be applied, the distribution will be identical or similar.

[0148] According to one or more embodiments, a first threshold and a second threshold are defined in the third scale and classes are defined relative to these thresholds.

[0149] A first class is the class of “mutated” individuals. This first class corresponds to the deciles from 1 to 10 previously described in the subsample of “mutated” individuals (Fig. 9A) (zone 121 in Fig. 12A). This first class corresponds to the subsample of “mutated” individuals, whose pairs of LDLc and PG Score values ​​are all in the red box in Fig. 12A. Subcategories can be defined according to the third scale in this first class, notably on the basis of the statistical distribution in Fig. 10A (“mutated” individuals), depending on the risk levels obtained for this class.

[0150] A second class is defined for “non-mutated” individuals. This second class corresponds to the 5 highest deciles of the third scale (deciles from 6 to 10, or approximately 50% of the distribution in Fig. 9B) for the subsample of “non-mutated” individuals (Fig. 9B). This second class corresponds to “non-mutated” individuals whose pairs of LDLc and PG Score values ​​are in the white box in Fig. 12B (hatched area 122). The low threshold, corresponding in this example to 6 ème decile, may vary depending on the subsample of “non-mutated” individuals considered.

[0151] This second class is a class called individuals with polygenic heredity in that: - their hypercholesterolemia, although severe, corresponds to the first half of the deciles of LDLc (the lowest) and that their PG Score is the highest (second half of the deciles: 6 to 10) signifying their association with a strong combinatorial of genetic variants of the PG Score (cf.:

[0001] ); - their distribution according to the predictive statistical model shows a distribution bias insofar as this class of individuals which represents 50% of “non-mutated” patients is only distributed over 26% of the distribution. - their cardiovascular risk, as assessed by their CAC values, shows the lowest median CAC score values ​​for patients in this class, which corresponds to the lowest cardiovascular risk.

[0152] A third class is defined for “non-mutated” individuals. This third class corresponds to the 4 middle deciles of the third scale (deciles from 2 to 5) for the subsample of “non-mutated” individuals (Fig. 9B). This third class corresponds to “non-mutated” individuals whose pairs of LDLc and PG Score values ​​are in the green box in Fig. 12C (hatched area 123). The low threshold, corresponding in this example to the 2 ème decile, may vary depending on the subsample of “non-mutated” individuals considered.

[0153] This third class is the so-called class of individuals with undetermined heredity in that: - their hypercholesterolemia is distributed over almost all LDLc deciles (1 to 9) and over all PG Score deciles (1 to 10) showing no particular association with the combination of genetic variants of the PG score (cf.:

[0001] ); - their distribution according to the predictive statistical model does not show any significant distribution bias insofar as this class of individuals which represents 40% of “non-mutated” patients is equally distributed over 45% of the distribution.

[0154] A fourth class is defined for “non-mutated” individuals. This fourth class corresponds to 1 er decile of the third scale (decile 1) for the subsample of “non-mutated” individuals (Fig. 9B). This fourth class corresponds to “non-mutated” individuals whose pairs of LDLc and PG Score values ​​are in the blue frame in Fig. 12D (hatched area 124).

[0155] This fourth class is the so-called class of individuals with monogenic heredity, the cause of which remains to be identified in that: - their severe hypercholesterolemia is distributed over the highest deciles of LDLc (5 to 10) and on all PG Score deciles (1 to 10) showing no particular association with the combination of genetic variants of the PG score; - their distribution according to the predictive statistical model shows a distribution bias insofar as this class of individuals which only represents 10% of “non-mutated” patients occupies 29% of the distribution; - their cardiovascular risk, as assessed by their CAC values, as well as the variations in their LDLc values ​​during follow-up (% reduction) show the highest values ​​for patients in this class corresponding to those found in “mutated” patients.

[0156] The two thresholds delimiting the second, third and fourth classes may vary depending on the distribution of individuals in the sub-sample of “non-mutated” individuals considered, in particular depending on the presence or absence of distribution bias in the different classes defined by these thresholds.

[0157] The example thresholds given above are related to the subsamples obtained. When the SD_DJ scale is in decile, the second threshold (low threshold) can correspond to the 2 ème (or 3 ème ) decile and the first threshold (high threshold) can correspond to 6 ème (or 7 ème ) decile for a decile scale.

[0158] Similarly, the second threshold can thus correspond to H ème (or 21 ième ) percentile and the first threshold can correspond to 51 ème (or 61 ème) percentile for a SP_DJ percentile scale. For a percentile scale, the thresholds can be adjusted more finely, the second threshold can for example be between 1 1 ème and 21 ième percentile and the first threshold can for example be between the 5i ème and the 61 ème percentile.

[0159] Based on the LDLc value pair and the PG score obtained for a patient and / or the class to which an individual belongs and the distribution in Fig. 1 1 B, it is possible to predict a rate of change in the cholesterol concentration in low-density lipoproteins. This change may be the consequence of a drug treatment of the patient or be related to another factor. This rate of change corresponds, for example, to a mean, minimum or maximum value of the reduction rate calculated on the basis of the values ​​of the reduction rates of the points of the distribution in Fig. 1 1 B which are in the box (see Fig. 12B to 12D) corresponding to the class concerned.

[0160] It is possible, based on the pair of LDLc values ​​and the PG score obtained for a patient and / or the class to which an individual belongs and via the distribution of CAC medians in Fig. 10B, to predict a level of risk of cardiovascular accident for patients. This level of risk corresponds, for example, to a mean, minimum or maximum CAC value, calculated on the basis of the CAC values ​​of the points of the distribution in Fig. 10B which are in the frame (see Fig. 12B to 12D) corresponding to the class concerned.

[0161] It is possible, for each of the classes of individuals, through the distribution of CAC medians in Fig. 10A, to predict a level of risk of cardiovascular accident for patients belonging to this class. This level of risk corresponds, for example, to a mean, minimum or maximum CAC value or a range of values, calculated on the basis of the CAC values ​​of the points of the distribution in Fig. 10A which are located in the frame (see Fig. 12A) corresponding to the class concerned.

[0162] The predicted risk level for a patient can also be calculated as the median value of the CAC score provided by the statistical distribution of CAC scores (Fig. 10) obtained for the subset of individuals having the same level (expressed in deciles) of polygenic risk score as the patient and the same level of cholesterol concentration in low-density lipoproteins, this subset of individuals being either a subset of the first subsample of “mutated” individuals if the patient is himself a “mutated” individual, or a subset of the second subsample of “non-mutated” individuals if the patient is himself a “non-mutated” individual.

[0163] Indeed, for patients currently considered "non-mutated", who represent more than 70% of molecular diagnoses in 2022 (Fig. 3), it becomes possible, for half of them (cf.: deciles 6 to 10, Fig. 9B) to inform them of their polygenic inheritance, the associated risk of CAD of which seems intermediate between that of monogenic and non-hereditary forms.

[0164] An expansion of molecular analyses can be proposed to patients in the fourth (1 er decile of Fig. 9B) whose characteristics suggest that they behave like patients with monogenic inheritance but whose molecular etiology remains to be identified.

[0165] Finally, clarifying the distinction between the different etiologies of severe hypercholesterolemia allows for better characterization of the variations in their LDLc values ​​during follow-up and their respective cardiovascular risks for better medical management of patients.

[0166] As used herein, "cardiovascular event" means any cardiovascular event known to the person skilled in the art. This may be, for example, coronary artery disease, a stroke, or any vascular complication related to atherosclerosis known to the person skilled in the art. This may be, for example, a cardiovascular event as described in "The risk of various types of cardiovascular diseases in mutation positive familial hypercholesterolemia; a review." Hovland A, Mundal LJ, Veierod MB, Holven KB, Bogsrud MP, Tell GS, Leren TP, Retterstol K. Front Genet. 2022 Dec 6;13:1072108. doi: 10.3389 / fgene.2022.1072108. PMID: 36561318; PMCID: PMC9763610.

[0167] As used herein, "cardiovascular risk" means any cardiovascular risk known to the person skilled in the art. This may, for example, be a cardiovascular risk as described in "The risk of various types of cardiovascular diseases in mutation positive familial hypercholesterolemia; a review." Hovland A, Mundal LJ, Veierod MB, Holven KB, Bogsrud MP, Tell GS, Leren TP, Retterstol K. Front Genet. 2022 Dec 6;13:1072108. doi: 10.3389 / fgene.2022.1072108. PMID: 36561318; PMCID: PMC9763610. As used herein, cardiovascular risk may, for example, be determined by the value of the CAC score.

[0168] In this document, the term "treatment of hypercholesterolemia" means any drug suitable for the treatment of hypercholesterolemia known to the person skilled in the art and / or commercially available. This may be, for example, a drug for the treatment of hypercholesterolemia, for example, any drug against excess cholesterol as described in the Vidal dictionary and / or commercially available. This may be, for example, a drug chosen from the group comprising drugs of the statin class, for example, atorvastatin, rosuvastatin, pitavastatin, pravastatin, simvastatin, fibrates, for example, bezafibrate, ciprofibrate, fenofibrate, cholesterol or bile acid absorption inhibitors, for example, ezetimibe, colestyramine, PCSK9 action inhibitors, for example, Alirocumab, Evolocumab.Examples include adapted dietary regimens, such as a diet and / or dietary regimen as described in “Recent advances in the management and implementation of care for familial hypercholesterolaemia.” Lan NSR, Bajaj A, Watts GF, Cuchel M. Pharmacol Res. 2023 Aug;194:106857. doi: 10.1016 / j.phrs.2023.106857. Epub 2023 Jul 17. PMID: 37460004, and / or “Advances in Treatment of Dyslipidemia.” Dybiec J, Baran W, Dqbek B, Fularski P, Mlynarska E, Radzioch E, Rysz J, Franczyk B. Int J Mol Sci. 2023 Aug 27;24(17):13288. doi: 1 ,0.3390 / ijms241713288. PMID: 37686091 ; PMCID: PMC10488025., of LDL apheresis, for example as described in Lipoprotein Apheresis. Feingold KR. 2023 Feb 19.In: Feingold KR, Anawalt B, Blackman MR, Boyce A, Chrousos G, Corpas E, de Herder WW, Dhatariya K, Dungan K, Hofland J, Kalra S, Kaltsas G, Kapoor N, Koch C, Kopp P, Korbonits M, Kovacs CS, Kuohung W, Laferrère B, Levy M, McGee EA, McLachlan R, New M, Purnell J, Sahay R, Shah AS, Singer F, Sperling MA, Stratakis CA, Trence DL, Wilson DP, editors. Endotext [Internet], South Dartmouth (MA): MDText.com, Inc.; 2000-. PMID: 28402616, for pharmacological procedures in development, for example what describes in Current Options and Future Perspectives in the Treatment of Dyslipidemia. Muscoli S, Ifrim M, Russo M, Candido F, Sanseviero A, Milite M, Di Luozzo M, Marchei M, Sangiorgi GM. Current Options and Future Perspectives in the Treatment of Dyslipidemia. J Clin Med. 2022 Aug 12;11 (16):4716. doi: 10.3390 / jcm11164716. PMID: 36012957; PMCID: PMC9410330.

[0169] The information provided by the distributions in Figure 8 or 9 makes it possible in particular to predict a class to which the patient belongs as described in relation to Figure 12.

[0170] The information provided by the distributions in Figures 10 and 11 makes it possible to predict, or at least estimate, for a patient, on the basis of the pair of LDLC values ​​and PG score obtained for this patient, a level of risk of cardiovascular accident and / or a variation in LDLc values.

[0171] Thus, a hypercholesterolemic profile can be determined for the patient. A hypercholesterolemic profile here corresponds to a category or status of a patient with severe hypercholesterolemic. This category can be determined on the basis of one or more of the parameters considered here, namely: - the patient class (as described for example with reference to FIG. 12); and / or - the estimated level of risk of cardiovascular accident; and / or - a variation in LDLc values ​​during monitoring.

[0172] Each of these parameters is linked to the pair of LDLc values ​​and PG score obtained for the patient. This pair of values ​​determines a position in a statistical distribution corresponding to the patient, this corresponding statistical distribution is: - either the statistical distribution (Fig. 8A or 9A) of the subsample of “mutated” individuals if the patient is a “mutated” individual; this position therefore determines a subset of individuals in the subsample of “mutated” individuals presenting the same pair (for example, expressed in deciles) of LDLc value and PG score as that obtained for the patient; - either the statistical distribution (Fig. 8B or 9B) of the subsample of “non-mutated” individuals if the patient is a “non-mutated” individual); this position therefore determines a subset of individuals in the subsample of “non-mutated” individuals presenting the same pair (for example, expressed in deciles) of LDLc value and PG score as that obtained for the patient.

[0173] Considering the patient's class: the first class corresponds to a hypercholesterolemic profile with confirmed so-called monogenic heredity, the second class corresponds to a hypercholesterolemic profile with so-called polygenic heredity, the third class corresponds to a hypercholesterolemic profile with so-called undetermined heredity, the fourth class corresponds to a hypercholesterolemic profile with so-called monogenic heredity of unidentified cause.

[0174] Within the first class, patient categories can be determined according to the pair of LDLc values ​​and PG score obtained for the patient, and / or according to another parameter such as the level of estimated cardiovascular accident risk and / or a variation in LDLc values.

[0175] In addition to class determination or as a replacement for class determination, determination of the hypercholesterolemic profile may include determination of an index, called the Hypercholesterolemia Polygenic Index (or "HPI"), for the patient.

[0176] This HPI index corresponds to the value given, according to the third scale, by the cumulative statistical distribution corresponding to the patient, at the position determined for the patient in the corresponding cumulative statistical distribution. This cumulative statistical distribution serves as a reference distribution for the patient. This reference distribution is either the cumulative statistical distribution of the subsample of "mutated" individuals if the patient is a "mutated" individual, or the cumulative statistical distribution of the subsample of "non-mutated" individuals if the patient is a "non-mutated" individual.

[0177] The third scale, used to represent the index, can be a scale (continuous or discrete) from 0 to 10 if the corresponding cumulative statistical distribution gives decile values ​​(SD_DJ scale), or on a scale (continuous or discrete) from 0 to 100 if the corresponding cumulative statistical distribution gives percentile values ​​(SP_DJ scale).

[0178] This HPI index is used to code, by a single value, synthetic information on the hypercholesterolemic profile. The HPI index can be calculated by interpolation in 2 dimensions, (for example by a polynomial formula, by a bilinear function, etc.) allowing to estimate the exact position (for example given in percentile) of any new patient in the reference cumulative statistical distribution from the exact values ​​of his polygenic risk score and his level of cholesterol concentration in low density lipoproteins.

[0179] This index is determinable independently of the patient's class and is more precise than the class. In each of the subsamples of "mutated" and "non-mutated" individuals respectively, a class corresponds to a range of index values.

[0180] For example, the first class corresponds to all index values ​​(values ​​0 to 10, or 0 to 100) for “mutated” patients (see Fig. 12A).

[0181] For example, for “non-mutated” patients: the second class (see Fig. 12B) corresponds to index values ​​in the interval [6;10] for a SDJDJ decile scale (or [51;100] for a SP_DJ percentile scale) for “non-mutated” patients when the first threshold is set at 6 ème decile (the beginning of the 6 ème decile corresponding to 51 èmepercentile). The third class (see Fig. 12C) corresponds to index values ​​of [2; 5[ for an SDJDJ decile scale (or [1 1; 50[ for an SPJDJ percentile scale) for non-mutated patients when the second threshold is set at 2 ème decile and the first threshold is set at the beginning of the 6th ème decile (the beginning of the 2 ème decile corresponding to 1 1 ème percentile); the fourth class (see Fig. 12D) corresponds to index values ​​of [0; 1[ for a SDJDJ decile scale (or [0; 10[ for a SPJDJ percentile scale) for “non-mutated” patients when the second threshold is set at 2 ème decile.

[0182] Fig. 13 shows a flowchart of a classification method according to one or more exemplary embodiments.

[0183] The steps of the method may be implemented by a computing device according to one of the examples described herein.

[0184] Although the steps are described sequentially, the person skilled in the art will appreciate that certain steps may be omitted, combined, performed in a different order and / or in parallel.

[0185] The method is an in vitro method for determining a hypercholesterolemic profile for a severely hypercholesterolemic patient. The method can be implemented by computer.

[0186] In a step 110, the method comprises obtaining a set of input data for the patient. This obtaining can be done by reading this input data from a memory.

[0187] Input data includes, for the patient: • genetic information on the presence or absence of a genetic mutation inducing hypercholesterolemia for at least one gene among a set of genes. • a polygenic risk score associated with hypercholesterolemia. • a concentration of cholesterol in low-density lipoproteins, in the absence of treatment.

[0188] In a step 111, the method comprises obtaining a first joint statistical distribution as a function of a quantile scale of polygenic risk score levels (hereinafter the first S_SPG scale) and a quantile scale of cholesterol concentration levels in low-density lipoproteins (hereinafter the second S_LDL scale). The first joint statistical distribution gives, as a function of these first and second S_SPG, S_LDL scales, a distribution of the individuals of a first sub-sample of severely hypercholesterolemic individuals having at least one genetic mutation causing hypercholesterolemia for at least one gene from a set of genes (sub-sample of “mutated” individuals). This obtaining can be done by reading data from a memory defining the first joint statistical distribution.

[0189] The set of genes may include at least: LDLR, APOB, PCSK9, APOE.

[0190] In a step 112, the method comprises obtaining a second joint statistical distribution as a function of the first scale in quantiles S_SPG of levels of the polygenic risk score and the second scale in quantiles S_LDL of levels of cholesterol concentration in low-density lipoproteins. The second joint statistical distribution gives as a function of these first and second scales a distribution of individuals from a second subsample of severely hypercholesterolemic individuals not presenting a genetic mutation in the set of genes, (subsample of “non-mutated” individuals). This can be obtained by reading data from a memory defining the second joint statistical distribution.

[0191] The first and second joint statistical distributions give distribution values ​​of individuals according to a third scale SD_DJ, SP_DJ in quantiles (for example, in deciles or percentiles).

[0192] The second quantile scale S_LDL of low-density lipoprotein cholesterol concentration levels used in the first and second statistical distributions is obtained based on an original statistical distribution, which is the statistical distribution of low-density lipoprotein cholesterol concentrations obtained for individuals in the total population consisting of the first and second subsamples of individuals (see the example of the original statistical distribution in Figure 4).

[0193] The first S_SPG scale of polygenic risk score levels used in the first and second statistical distributions is obtained based on an original statistical distribution, which is the statistical distribution of polygenic risk scores obtained for individuals in the first subsample of individuals (see the example of the original statistical distribution in Figure 5).

[0194] Thus each of the first and second scales S_LDL, S_SPG is a scale derived from an original statistical distribution and encodes information about the quantile distribution of individuals in this original statistical distribution. For example, the correspondence between a given quantile value in the first scale S_SPG and the polygenic risk score value is given by the original statistical distribution (see example in Fig. 5). Similarly, the correspondence between a given quantile value in the second scale S_LDL and the cholesterol concentration value in low-density lipoproteins is given by the original statistical distribution (see example in Fig. 4).

[0195] Furthermore, by combining these two particular scales which are the basis of the two statistical distributions, it is possible to determine a hypercholesterolemic profile for a severely hypercholesterolemic patient based on the patient's position in the statistical distribution of "mutated" individuals (or respectively "non-mutated" depending on the patient's profile).

[0196] The first joint statistical distribution can be a cumulative statistical distribution obtained by regression from an initial statistical distribution obtained from the first subsample of individuals.

[0197] The second joint statistical distribution can be a cumulative statistical distribution obtained by regression from an initial statistical distribution obtained the second subsample of individuals.

[0198] The second S_LDL scale in quantiles of low-density lipoprotein cholesterol concentration levels may, for example, be a decile scale whose deciles are calculated on the basis of a statistical distribution of low-density lipoprotein cholesterol concentrations obtained from biological samples for individuals in the total population consisting of the first and second subsamples of individuals.

[0199] The first S_SPG scale in quantiles of polygenic risk score levels can for example be a decile scale whose deciles are calculated on the basis of the statistical distribution of polygenic risk scores obtained from biological samples for individuals in the first subsample of individuals.

[0200] Polygenic score deciles are defined on a general population for which subjects above 1.9 gL -1LDLc represent only 10% of the population, yet this is the entire population of interest. The "mutated" individuals in the population of interest constitute a representative subsample of the random distribution of the PG score in the severe hypercholesterolemic population because it is due to the presence of a mutation that they belong to this population.

[0201] In a step 115, the method comprises determining, for the patient, a position in one of the statistical distributions as a function of the polygenic risk score and the cholesterol concentration in the low-density lipoproteins. The position is determined in the first joint statistical distribution when the genetic information obtained for the patient indicates the presence of at least one mutation (“mutated” patient) and in the second joint distribution when the genetic information obtained for the patient indicates the absence of mutation (“non-mutated” patient).

[0202] In a step 120, the method comprises determining the hypercholesterolemic profile of the patient as a function of the determined position.

[0203] Determination of the patient's hypercholesterolemic profile may include determination of an HPI index corresponding to the given value, at the position determined for the patient in the third scale: (i) either by the first joint statistical distribution (first cumulative statistical distribution) when the genetic information obtained for the patient indicates the presence of at least one mutation (patient “mutated”); (ii) or by the second joint statistical distribution (second cumulative statistical distribution) when the genetic information obtained for the patient indicates the absence of mutation (“non-mutated” patient). Obtaining this index has been described in detail above.

[0204] Determining the patient's hypercholesterolemic profile may include determining a patient's class based on the position obtained.

[0205] The class can be - either a first class when the genetic information indicates the presence of a genetic mutation; - either a second class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint statistical distribution presenting a value greater than or equal to a first threshold according to the third scale; - or a third class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint statistical distribution indicate a value greater than or equal to a second threshold according to the third scale and strictly less than the first threshold according to the third scale; - or a fourth class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint distribution indicating a value strictly lower than the second threshold.

[0206] When the first and second joint statistical distributions give values ​​of the distribution of individuals according to a third SD_DJ scale in deciles: the first threshold corresponds to 6 ème or 7 ème decile and the second threshold corresponds to 2 ème or 3 ème decile.

[0207] Belonging to one of the classes makes it possible to define hypercholesterolemic profiles. The first class corresponds to a hypercholesterolemic profile with confirmed so-called monogenic heredity. The second class corresponds to a hypercholesterolemic profile with so-called polygenic heredity. The third class corresponds to a hypercholesterolemic profile with so-called undetermined heredity. The fourth class corresponds to a hypercholesterolemic profile with so-called monogenic heredity of unidentified cause.

[0208] Thus, by means of the classes defined for "non-mutated" patients, it is possible to distinguish patients who may be potentially monogenic from patients who may be polygenic within "non-mutated" patients. As a result, the two joint statistical distributions make it possible to distinguish different subclasses of severe hypercholesterolemia, in particular through the biases appearing in the statistical distributions thus constituted.

[0209] Determining the patient's hypercholesterolemic profile may include a prediction of a level of risk of cardiovascular accident for the patient on the basis of the position obtained and a third statistical distribution (see: fig. 10) function of the same S_SPG scale of polygenic risk score levels and the same S_LDL scale of cholesterol concentration levels in low-density lipoproteins, the third statistical distribution giving a level of risk of cardiovascular accident as a function of these scales.

[0210] The risk level is calculated as a median value of the CAC score, Coronary Artery Calcium of the statistical distribution of the CAC scores obtained for the subset of individuals having the same level of polygenic risk score as the patient and the same level of cholesterol concentration in low-density lipoproteins, this subset of individuals being either a subset of the first subsample of “mutated” individuals when the genetic information obtained for the patient indicates the presence of at least one mutation, or a subset of the second subsample of “non-mutated” individuals when the genetic information obtained for the patient indicates the absence of a mutation.

[0211] One or more or all of the steps of one or more methods described in this document may be implemented by software or computer program and / or by hardware, for example by circuit, programmable or not, specific or not.

[0212] The functions, steps, and methods described in this document may be implemented by software (e.g., via software on one or more processors, for execution on a general-purpose or special-purpose computer) and / or be implemented by hardware (e.g., one or more circuits, and / or any other hardware component, whether based on electronics, optics, or other technologies).

[0213] The present description thus relates to a software or computer program, capable of being executed by a host device, by means of one or more data processors, this software / program comprising instructions to cause the execution by this host device of all or part of the steps of one or more methods described in this document. These instructions are intended to be stored in a memory of the host device, loaded and then executed by one or more processors of this host device so as to cause the execution by this host device of the method in question.

[0214] This software / program may be coded using any programming language, and may be in the form of source code, object code, or code intermediate between source code and object code, such as in a partially compiled form, or in any other desirable form.

[0215] The host device may be implemented by one or more physically distinct machines. The host device may generally have the architecture of a computer, including one or more components of such architecture: data memory(ies), processor(s), communications bus, hardware interface(s) for connecting this host device to a network or other equipment, user interface(s), etc.

[0216] In one embodiment, all or part of the steps of an operation triggering method or of another method described in this document are implemented by a device provided with means for implementing these steps of this method.

[0217] These means may include software means (for example, instructions of one or more components of a program) and / or hardware means (for example, circuit(s), data memory(ies), processor(s), communication bus, hardware interface(s), etc.).

[0218] These means may comprise, for example, one or more circuits configured to execute one or more or all of the steps of one of the methods described herein. These means may comprise, for example, at least one processor and at least one memory comprising program instructions configured to, when executed by the processor, cause the device to execute one or more or all of the steps of one of the methods described herein.

[0219] Means implementing a function or set of functions may correspond in this document to a software component, a hardware component or a combination of hardware and / or software components, capable of implementing the function or set of functions, according to what is described below for the means concerned.

[0220] The present description also relates to an information medium readable by a data processor, and comprising instructions of a program as mentioned above.

[0221] The information carrier may be any material means, entity or device, capable of storing the instructions of a program as mentioned above. Usable program storage media include ROM or RAM memories, magnetic storage media such as magnetic disks and magnetic tapes, hard disks or optically readable digital data storage media, or any combination of these media.

[0222] In some cases, the computer-readable storage medium is not transient. In other cases, the information medium may be a transient medium (e.g., a carrier wave) for the transmission of a signal (electromagnetic, electrical, radio, or optical signal) carrying the program instructions. This signal may be conveyed via a suitable transmission medium, whether wired or wireless: electrical or optical cable, radio or infrared link, or by other means.

[0223] An embodiment also relates to a computer program product or a computer-readable storage medium on which are stored instructions for program, the program instructions being configured to cause the host device (e.g., a computer) to implement some or all of the steps of one or more methods described herein when the program instructions are executed by one or more processors and / or one or more programmable hardware components of the host device.

[0224] In this description, the terms "means configured to perform one or more functions" or "means for performing one or more functions" may correspond to one or more functional blocks comprising circuits adapted to perform or configured to perform the function(s) concerned. The block may perform this function itself or cooperate and / or communicate with other blocks to perform this function. The "means" may correspond to or be implemented as "one or more modules", "one or more devices", "one or more units", etc.

[0225] The means may comprise at least one processor and at least one memory comprising at least one memory storing instructions which, when executed by at least one processor, cause a device to perform the at least one function in question. The means may comprise circuitry (e.g., processing circuitry) configured to perform the at least one function in question.

[0226] Although the terms first, second, etc. may be used herein to describe various elements, such elements should not be limited by these terms. These terms are used only to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of this disclosure. As used herein, the term "and / or" includes all combinations of one or more of the listed elements.

[0227] The terminology used herein is for the sole purpose of describing particular embodiments and is not limiting. In this document, the singular forms "a", "a" and "the" also include the plural, unless the context clearly indicates otherwise. It is understood that the terms "comprises", "includes", "includes" and / or "including", when used herein, specify the presence of given features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups of such features, integers, steps, operations, elements, components and / or groups of such elements.

[0228] Although certain aspects of the present disclosure have been particularly illustrated and described with reference to the above embodiments, those skilled in the art will understand that various additional embodiments may be contemplated by modifying the devices and methods described.

[0229] LIST OF MAIN ABBREVIATIONS

[0230] CAC Coronary Artery Disease

[0231] DLCN Dutch Lipid Clinic Network

[0232] HF familial hypercholesterolemia

[0233] NGS Next Generation Sequencing

[0234] SNP Single Nucleotide Polymorphism

[0235] CITED REFERENCE

[0001] “Use of low-density lipoprotein cholesterol gene score to distinguish patients with polygenic and monogenic familial hypercholesterolaemia: a case-control study”. Talmud PJ, Shah S, Whittall R, Futema M, Howard P, Cooper JA, Harrison SC, Li K, Drenos F, Karpe F, Neil HA, Descamps OS, Langenberg C, Lench N, Kivimaki M, Whittaker J, Hingorani AD, Kumari M, Humphries SE. Lancet. 2013 Apr 13;381 (9874): 1293-301. PMID: 23433573

Claims

CLAIMS 1. In vitro method for determining a hypercholesterolemic profile for a severely hypercholesterolemic patient, the method being implemented by computer and comprising: - obtaining (111) a first joint statistical distribution giving a distribution of individuals according to the combination of a first quantile scale (S_SPG) of polygenic risk score levels and a second quantile scale (S_LDL) of cholesterol concentration levels in low-density lipoproteins, the first joint statistical distribution giving the distribution, according to the first and second scales, of the individuals of a first sub-sample of severely hypercholesterolemic individuals presenting at least one genetic mutation causing hypercholesterolemia for at least one gene among a set of genes; - obtaining (112) a second joint statistical distribution giving a distribution of individuals according to the combination of the first scale in quantiles of levels of the polygenic risk score and the second scale in quantiles of levels of cholesterol concentration in low-density lipoproteins, the second joint statistical distribution giving the distribution, according to the first and second scales, of the individuals of a second sub-sample of severely hypercholesterolemic individuals not presenting a genetic mutation in the set of genes; - obtaining (110), by reading from a memory, data determined from at least one biological sample for the patient, the data comprising: o genetic information on the presence or absence of at least one genetic mutation causing hypercholesterolemia for at least one gene among the set of genes; o a polygenic risk score; o a cholesterol concentration in low-density lipoproteins in the absence of treatment; - a determination (115), for the patient, of a position in one of the joint statistical distributions as a function of the polygenic risk score and the cholesterol concentration in the low-density lipoproteins, the position being determined in the first joint statistical distribution when the genetic information obtained for the patient indicates the presence of at least one mutation and in the second joint statistical distribution when the genetic information obtained for the patient indicates the absence of mutation; - a determination (120) of the hypercholesterolemic profile of the patient as a function of the determined position; in which the first and second joint statistical distributions give distribution values ​​of individuals according to a third scale, the second scale (S_LDL) being obtained on the basis of a quantile distribution of the cholesterol concentrations in low-density lipoproteins obtained for the individuals of the total population consisting of the first and second sub-samples of individuals; the first scale (S_SPG) being obtained on the basis of a quantile distribution of the polygenic risk scores obtained for the individuals of the first sub-sample of individuals.

2. Method according to claim 1, in which the determination (120) of the hypercholesterolemic profile of the patient comprises a determination of an index corresponding to the given value, according to the third scale, at the position determined for the patient, either by the first joint statistical distribution when the genetic information obtained for the patient indicates the presence of at least one mutation, or by the second joint statistical distribution when the genetic information obtained for the patient indicates the absence of mutation.

3. Method according to claim 1 or 2, wherein the determination of the hypercholesterolemic profile of the patient comprises a determination of a class to which the patient belongs on the basis of the position obtained, the class being - either a first class when the genetic information indicates the presence of a genetic mutation; - either a second class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint statistical distribution presenting a value greater than or equal to a first threshold in the third scale; - or a third class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint statistical distribution indicate a value greater than or equal to a second threshold in the third scale and strictly less than the first threshold in the third scale; - or a fourth class when the genetic information indicates the absence of a genetic mutation and the values ​​of the polygenic score and the concentration of cholesterol in low-density lipoproteins obtained for the patient correspond to a position in the second joint distribution indicates a value strictly lower than the second threshold in the third scale.

4. Method according to claim 3, in which the first class corresponds to a hypercholesterolemic profile with confirmed so-called monogenic heredity, in which the second class corresponds to a hypercholesterolemic profile with so-called polygenic heredity, in which the third class corresponds to a hypercholesterolemic profile with so-called undetermined heredity, in which the fourth class corresponds to a hypercholesterolemic profile with so-called monogenic heredity of unidentified cause.

5. Method according to any one of the preceding claims, in which the first joint statistical distribution is a cumulative statistical distribution obtained by generalized additive model type regression from an initial statistical distribution giving a distribution of the individuals of the first sub-sample of individuals according to the combination of the first and second quantile scales.

6. Method according to any one of the preceding claims, in which the second joint statistical distribution is a cumulative statistical distribution obtained by generalized additive model type regression from an initial statistical distribution giving a distribution of the individuals of the second sub-sample of individuals according to the combination of the first and second quantile scales.

7. A method according to any preceding claim, wherein the second scale is a quantile scale, the quantiles of which are calculated on the basis of the statistical distribution of cholesterol concentrations in low-density lipoproteins obtained from biological samples for individuals in the total population consisting of the first and second sub-samples of individuals.

8. A method according to any preceding claim, wherein the first scale is a quantile scale whose quantiles are calculated based on the statistical distribution of polygenic risk scores obtained from biological samples for individuals in the first subsample of individuals.

9. Method according to claim according to any one of claims 1 to 8, in which the third scale (SD_DJ, SP_DJ) is a quantile scale.

10. Method according to claims 3 and 9, in which the second threshold corresponds at 2 ème or 3 ème decile and the first threshold corresponds to 6 ème or 7 ème decile when the third scale (SD_DJ) is in deciles.

11. Method according to any one of the preceding claims, in which the set of genes comprises at least: LDLR, APOB, PCSK9, APOE.

12. Method according to any one of claims 1 to 11, in which the determination of the hypercholesterolemic profile of the patient comprises: - a prediction of a level of risk of cardiovascular accident for the patient on the basis of the position obtained and a third joint statistical distribution depending on the first quantile scale (S_SPG) of polygenic risk score levels and the second quantile scale (S_LDL) of cholesterol concentration levels in low-density lipoproteins, the third joint statistical distribution giving, as a function of these first and second scales, a level of risk of cardiovascular accident.

13. The method of claim 12 wherein the risk level is calculated in the third statistical distribution as a median value of the CAC score, Coronary Artery Calcium, of a statistical distribution of the CAC scores obtained for the subset of individuals having the same level of polygenic risk score as the patient and the same level of cholesterol concentration in low-density lipoproteins as the patient, this subset of individuals being either a subset of the first subsample of individuals when the genetic information obtained for the patient indicates the presence of at least one mutation, or a subset of the second subsample of individuals when the genetic information obtained for the patient indicates the absence of a mutation.

14. A device comprising means for implementing the steps of a method according to any one of the preceding claims.

15. Device according to claim 14, wherein the means comprise - at least one processor; - at least one memory storing instructions which, when executed by at least one processor, cause the device to execute the method.

Images