MITOCHONDRIAL GENOTYPIC SCORES: PROGNOSTIC MARKERS IN CHEMO-SENSITIVE CANCER

Abstract

The present invention relates to a prognostic risk score for chemosensitive cancer, in particular acute myeloid leukemia (AML), based on somatic genetic abnormalities affecting certain genes of the mitochondrial genome. Abbreviated figure: Fig. 2A

Description

Title of the invention: MITOCHONDRIAL GENOTYPIC SCORES: PROGNOSTIC MARKERS IN CHEMO-SENSITIVE CANCER Technical field of the invention

[0001] The present invention relates to a prognostic risk score for chemosensitive cancer, in particular acute myeloid leukemia (AML), based on somatic genetic abnormalities affecting certain genes of the mitochondrial genome.

[0002] In the description below, references in parentheses [ ] refer to the list of references presented at the end of the text. State of the art

[0003] Acute myeloid leukemia (AML) is an uncontrolled malignant proliferation of immature myeloid progenitors called leukemic cells, arrested at an early stage of differentiation, accumulating in the bone marrow and possibly in other organs. Despite therapeutic advances, the prognosis is very poor, with a cure rate of 35 to 40% in adult patients under 60 years of age and 5 to 15% in patients over 60 years of age. The prognostic factors currently used in practice are based on age, circulating white blood cell count, and the 2017 European LeukemiaNet (ELN) prognostic classification (Dohner et al., 2017) [1]. The 2017 ELN classification is based on cytogenetic and / or molecular alterations and defines three statuses: favorable, intermediate, and unfavorable. However, this prognostic classification can still be improved.For example, in the favorable group, the 5-year survival rate is only 64% in patients under 60 years of age (35% in those over 60) [2].

[0004] Mitochondria are cellular organelles of bacterial origin present in all eukaryotic cells (except red blood cells). They play an essential role in the regulation of cellular metabolism, in energy supply (Krebs cycle, 3-oxidation of fatty acids, etc.), but also in calcium homeostasis, the generation of reactive oxygen species (ROS), and in triggering apoptosis. Somatic mutations of isocitrate dehydrogenases (IDH1, IDH2), mitochondrial enzymes present on the nuclear genome, lead to dysregulation of the Krebs cycle with the accumulation of a neometabolite, 2-hydroxyglutarate, which contributes to leukemogenesis.

[0005] In AML, the level of OXPHOS pathway activation is directly correlated with the resistance of leukemic cells (Farge et al, 2017) [3]. Furthermore, mitochondrial transfers between stromal cells and leukemic cells have been described. as responsible for the resistance of leukemic cells to chemotherapy. In a prospective study (DRCI LAM38RC13-209), it was shown that dysregulation of reactive oxygen species (ROS) production by the mitochondria of leukemic cells was associated with a poor prognosis (decreased overall survival) independently of the usual prognostic factors of AML (age, WBC / 1, ELN 2017, transplant) (Mondet et al, 2019) [4]. Thus, several lines of inquiry converge on the role of mitochondria in AML chemoresistance.

[0006] However, regarding the mitochondrial genome, few studies have analyzed the impact of mitochondrial genome mutations as a prognostic marker in AML. ND4 mutations, encoding a subunit of complex I, were analyzed in 452 patients using a targeted sequencing approach (different from high-throughput sequencing technology) in AML and are associated with increased overall survival in a univariate analysis (Damm et al., 2012) [5]. However, multivariate analysis did not significantly confirm this result (p=0.089). The article even concludes that this result should be confirmed with further studies. Another study investigated the prognostic impact of ND4 mutations in 121 patients with AML (Chun et al., 2014) [6]. However, the study showed no difference in terms of overall survival or relapse-free survival. The prognostic impact of ND4 is therefore not a certainty for a person skilled in the art.

[0007] However, the mitochondrial genome comprises 16 kB, including 37 genes encoding 13 proteins involved in the respiratory chain, as well as 22 tRNAs and 2 rRNAs. Another study using publicly available somatic genetics data from the cbioportal website (https: / / www.cbioportal.org / ) analyzed the impact of mutations on acute myeloid leukemia (Wu et al., 2018) [7]. However, several biases exist in this study. One of them is related to the use of a bioinformatics analysis dedicated to genomic T-DNA, not specific to mitochondrial T-DNA. In this study, only 8% of AML patients have mutations present in the genes (ND1, ND2, ND3, ND4, ND4L, ND5, ND6, CYB, C0X1, C0X2, C0X3, ATP6, ATP8). Errors in the analysis of genetic data lead to an underestimation of the number of variants, thus skewing the final results. For comparison, in the study targeting only the ND4 gene (Damm et al., 2012) [5], the frequency of mutations was 6.4% (29 / 452). Since mitochondrial DNA has its own genetic code, using a common bioinformatics analysis between nuclear genomic mutations and mitochondrial mutations leads to errors in the interpretation of the variants obtained (Caudron-Herger and Diederichs, 2018) [8]. The same analysis bias exists in other pathologies (e.g., cbioportal website as of 08 / 10 / 2021, Diffuse Large B cell Lymphoma study, Duke 2017, 0% present mutations in the ND1, ND2, ND3, ND4, ND4L genes vs. 35% in the study by Zeng et al., . 2018) [9]).

[0008] Apart from the studies by Damm et al. and Chun et al. analyzing only the ND4 gene, and by Wu et al. using an inappropriate bioinformatic analysis, no study has shown the contribution of mitogenome mutations in the stratification of patients with AML.

[0009] Therefore, the study of the mitochondrial genome and the identification of possible new prognostic markers of AML remains a challenge, in order to better classify patients with AML in terms of overall survival, and thus adapt their therapeutic management according to their survival prognosis. Description of the invention

[0010] The Inventors are the very first to have highlighted a score based on combinations of presence and absence of mutations of certain genes of the mitochondrial genome as prognostic markers of acute myeloid leukemia (AML).

[0011] The inventors have thus developed a score, hereinafter referred to as Mitoscore, to predict the response to chemotherapy treatment and the survival of patients with AML. Mitoscore is based on mitochondrial genome sequencing technology (e.g., NGS or other technology). Based in particular on molecular abnormalities of the ND2, ND3, ND4, CYTB, ATP8, COX1, COX2, COX3, and 12S genes, Mitoscore stratifies AML patients at diagnosis into three groups with respective prognoses: favorable, intermediate, and unfavorable in terms of overall survival. Mitoscore provides improved prognostic stratification compared to the European LeukemiaNet 2017 (ELN 2017) prognostic score currently used for AML patients, in order to distinguish good from poor responders and to tailor therapeutic management.The Mitoscore is valid independently of the usual prognostic factors (age, white blood cell count, ELN 2017, transplant), and can be combined with FELN2017 in the form of a score, hereinafter referred to as Mitoscore+.

[0012] The Mitoscore improves prognostic stratification of AML patients at diagnosis. Furthermore, the Mitoscore obtained by sequencing the mitochondrial genome (16 kb) is simpler than the 2017 ELN classification, which requires both culturing leukemic cells for karyotyping and sequencing molecular abnormalities, for example, using NGS (a panel of approximately 100 kb, varying between centers). In addition, karyotyping requires a large number of leukemic cells, which is not always obtained in bone marrow deficiencies. The Mitoscore therefore saves time, reduces costs, offers easier reproducibility (no cell culture, no interpretation difficulties), and requires less biological material.

[0013] The present invention therefore relates to an in vitro method for establishing a prognosis of Mitoscore survival in a patient with chemosensitive cancer, said process comprising the following steps: - detection in a biological sample of said patient, of the presence and / or absence of at least one mutation on one of the nine genes of the following mitochondrial genome: ND2, ND3, ND4, CYTB, ATP8, C0X1, C0X2, C0X3 and 12S; - comparison of detection results with a reference biological sample or a reference sequence or a reference haplogroup / haplotype; - establishment of the survival prognosis of said patient.

[0014] For the purposes of this invention, "chemosensitive cancer" means a pathology selected from the group consisting of acute myeloid leukemia (AML), sarcomas, testicular cancer (germ cell cancer in general), choriocarcinomas, hematological malignancies, ovarian cancer, and breast cancer. Preferably, it is acute myeloid leukemia (AML).

[0015] For the purposes of this invention, the term "reference biological sample" means a biological sample from a healthy subject, for example, a DNA sample from bone marrow, blood, or tissue.

[0016] For the purposes of this invention, "reference sequence" means a reference mitochondrial DNA sequence, for example a sequence present in the Mitomap / Mitomaster, gnomAD databases.

[0017] The term “reference haplogroup / haplotype” according to the present invention means the variations in composition relative to the reference sequence (CRS) defining a haplogroup or haplotype according to the classification of Richards and Macaulay of 1998, which can be determined by several tools (MitoTool, HaploFind...).

[0018] According to a particular embodiment of the in vitro method for establishing a Mitoscore survival prognosis in a patient with chemosensitive cancer according to the present invention: - the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, ND4 and the absence of mutation in the genes COX1, COX2, COX3, 12S, is an indication of a favorable survival prognosis; - The presence of at least one mutation in one of the genes COX1, COX2, COX3, 12S, and the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, indicates an unfavorable survival prognosis; and - either the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S, or the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, ND4 and at least one mutation in one of the genes COX1, COX2, COX, 12S, is the indication of an intermediate survival prognosis.

[0019] According to a particular embodiment of the in vitro process for establishing a Mitoscore survival prognosis in a patient with chemosensitive cancer according to the present invention: - the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, and the absence of mutation in the genes COX1, COX2, COX3, 12S, is an indication of a favorable survival prognosis; - The presence of at least one mutation in one of the genes C0X1, C0X2, C0X3, 12S, and the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, indicates an unfavorable survival prognosis; and - either the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, C0X1, C0X2, C0X3, 12S, or the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, ND4 and at least one mutation in one of the genes C0X1, C0X2, COX, 12S, is the indication of an intermediate survival prognosis.

[0020] For the purposes of this invention, a “favorable”, “intermediate” or “unfavorable” survival prognosis means the gradation of hypotheses made on the evolution of the pathology based on mitochondrial variants (i.e. in terms of chances of survival, risks of complications and / or death).

[0021] The present invention further relates to an in vitro method for establishing a Mitoscore+ survival prognosis in a patient with acute myeloid leukemia (AML) with no mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes (i.e., a patient from a subgroup called Mitonaïf), said method comprising the method for establishing a Mitoscore survival prognosis as defined above and the determination of the EuroLeukemiaNet 2017 (ELN 2017) prognostic score in said patient.

[0022] According to a particular embodiment of an in vitro method for establishing a Mitoscore+ survival prognosis in a patient with acute myeloid leukemia (AML) according to the present invention: - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes and the absence of mutation in the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and a "favorable" classification according to the ELN 2017 score, is the indication of a favorable survival prognosis; - either the presence of at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes and the absence of mutation in the ND2 / ND3 / ATP8 / CYTB / ND4 genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and an "unfavorable" classification according to the ELN 2017 score, is an indication of an unfavorable survival prognosis; - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes and at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, C0X1, COX2, COX3, 12S and an "intermediate" ranking according to the ELN 2017 score, is an indication of an intermediate survival prognosis.

[0023] According to a particular embodiment of an in vitro method for establishing a Mitoscore+ survival prognosis in a patient with acute myeloid leukemia (AML) according to the present invention: - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB genes and the absence of mutation in the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and a "favorable" classification according to the ELN 2017 score, is the indication of a favorable survival prognosis; - either the presence of at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes and the absence of mutation in the ND2 / ND3 / ATP8 / CYTB / ND4 genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and an "unfavorable" classification according to the ELN 2017 score, is an indication of an unfavorable survival prognosis; - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes and at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and an "intermediate" classification according to the ELN 2017 score, is the indication of an intermediate survival prognosis.

[0024] For the purposes of this invention, a “favorable”, “intermediate” or “unfavorable” survival prognosis means the gradation of hypotheses made on the evolution of the pathology based on mitochondrial variants (i.e. in terms of chances of survival, risks of complications and / or death).

[0025] According to a particular embodiment of a method according to the present invention, the detection steps are carried out by next-generation sequencing (NGS) of the mitochondrial genome.

[0026] The present invention further relates to an in vitro method for predicting or evaluating the efficacy and / or benefit of a treatment for a chemosensitive cancer, in particular acute myeloid leukemia (AML), in a patient with said cancer, comprising the following steps: - determination of a Mitoscore or Mitoscore+ survival prognosis according to a method according to the present invention from a biological sample of said patient before treatment; - determination of a Mitoscore or Mitoscore+ survival prognosis according to a method according to the present invention from a biological sample of said patient after treatment; - comparison of Mitoscore or Mitoscore+ survival prognoses before and after treatment ; - determination of either resistance to treatment when the prognosis of survival of said patient after treatment is the same as or worse than the prognosis of survival of said patient before treatment, or of sensitivity to treatment when the prognosis of survival of said patient after treatment is better than the prognosis of survival of said patient before treatment.

[0027] For the purposes of this invention, "chemosensitive cancer" means a pathology selected from the group consisting of acute myeloid leukemia (AML), sarcomas, testicular cancer (germ cell cancer in general), choriocarcinomas, hematological malignancies, ovarian cancer, and breast cancer. Preferably, it is acute myeloid leukemia (AML).

[0028] For the purposes of this invention, “identical survival prognosis before and after treatment” means a favorable, intermediate, or unfavorable survival prognosis before treatment which remains, respectively, a favorable, intermediate, or unfavorable survival prognosis after treatment.

[0029] For the purposes of the present invention, “worse survival prognosis after treatment than before treatment” means, for example, a favorable or intermediate survival prognosis before treatment which becomes, respectively, an intermediate or unfavorable prognosis after treatment.

[0030] For the purposes of the present invention, "better survival prognosis after treatment than before treatment" means, for example, an intermediate or unfavorable survival prognosis before treatment which becomes, respectively, a favorable or intermediate survival prognosis after treatment. BRIEF DESCRIPTION OF THE FIGURES

[0031] [Fig.l] represents the technical workflow carried out for the application of Mitoscore and Mitoscore+, defining in particular the criteria used for the selection of variants.

[0032] [Figure 2] represents the Kaplan-Meier survival curves (log-rank test p<0.05) representing overall survival over time in patients with AML according to [Fig.2A] the Mitoscore or [Fig.2B] the Mitoscore+.

[0033] [Fig.3] represents the Hazard Ratio (HR) (95% CI) of the risk of death as a function of the Mitoscore and Mitoscore+ calculated according to the Cox model in univariate (UV) or multivariate (MV) analysis. The covariates used in the MV analysis are age at diagnosis, white blood cell (WBC) count in G / L, ELN 2017 classification and / or bone marrow transplant. EXAMPLES

[0034] EXAMPLE 1: DEVELOPMENT OF A MITO- GENOTYPIC SCORE CHONDRIAL IN THE CONTEXT OF PROGNOSIS FOR ACUTE MYELOID LEUKEMIA (AML)

[0035] Abstract: The Inventors sequenced the entire mitochondrial genome using next-generation sequencing (NGS) of patients with acute myeloid leukemia (AML). From this sequencing, a score called Mitoscore was defined, which made it possible to predict patient survival. The usefulness of Mitoscore was confirmed in multivariate analysis independently of the usual prognostic factors for AML (i.e., age, circulating white blood cell (WBC) count, cytogenetic and molecular abnormalities, transplantation). Furthermore, Mitoscore was functionally characterized by comparison with ROS emission data with and without inhibitors of mitochondrial complexes (antimycin A, rotenone). Definition of the selected variants:

[0036] Mitochondrial genome sequencing was performed on an S5 automated system after PCR amplification of the mitochondrial genome (two 8 kDa fragments). Mitochondrial genome sequencing can also be performed using other sequencing technologies (Illumina). Variant analysis was performed using Mitomaster. Variants affecting non-coding regions (e.g., D-loop), heteroplasmy levels strictly below 3%, and silent mutations were eliminated. The Mitomaster website defined a mitochondrial haplogroup. Variants with a frequency in the haplogroup <0.5% or those described in cancers were included. Any sequencing errors were checked using IGV. For haplogroups with a sample size <100, the variant frequency was checked in gnomAD v3 and included if <0.5% or if described in cancers ([Fig. 1]). Development of the Mitoscore:

[0037] From the survival curves produced individually by mitochondrial genome gene in the 64 patients of the study who received induction chemotherapy, certain genes were determined to be of “good prognosis” (ND2 / ND3 / ND4 / ATP8 / CYTB) or of “poor prognosis” (COX1 / COX2 / COX3 / 12S).

[0038] Despite good separation of survival curves, no significant difference was observed in the univariate analysis of these genes taken individually.

[0039] Consequently, the “Mitoscore” was created, stratifying patients into 3 groups of respective prognostic favourable, intermediate, unfavorable ([Fig.2A]).

[0040] This prognostic classification based on the “Mitoscore” is independent of the usual prognostic factors of AML (age, WBC / 1, ELN 2017, bone marrow transplant).

[0041] However, 30% of patients with no mutations in any of the Mitoscore genes, i.e., patients in the subgroup called Mitonaïf, require a re classification. Therefore, the Mitoscore+ was created by combining the ELN 2017 prognostic classification with the Mitoscore in Mitochondrial-naïve patients ([Fig. 2B]). This allows for a reduction in the use of the ELN 2017 prognostic classification of approximately 70%.

[0042] The usefulness of Mitoscore and Mitoscore+ was confirmed with a Cox model in univariate (UV) analysis, independent of the usual prognostic factors for AML, and in multivariate (MV) analysis ([Fig. 3]). The covariates used in the MV analysis were age, circulating white blood cell (WBC) count in G / L, ELN 2017 classification, and / or bone marrow transplantation.

[0043] Furthermore, the Mitoscore was functionally characterized by comparison with ROS emission data. Interestingly, leukemic cells from patients in the "unfavorable" Mitoscore subgroup emit significantly less ROS under antimycin / rotenone stimulation (targeting mitochondrial ROS) than those in the "favorable" and "intermediate" Mitoscore subgroup with COX1 / COX2 / COX3 / 12S and ND2 / ND3 / ATP8 / CYTB / ND4 mutations.

[0044] Within mitochondria, complexes I and III of the respiratory chain are the main sources of reactive oxygen species (ROS) emission. The genes involved in the "favorable" Mitoscore affect subunits of complexes I (ND2 / ND3 / ND4) and III (CYTB), as well as ATP synthase (ATP8). The ROS data therefore support the hypothesis of the "favorable" Mitoscore's function, where mutations affect ROS emission, particularly under stimulation conditions such as those induced by chemotherapy. Thus, high ROS emission triggers the apoptosis threshold and leads to cell death in leukemic cells.

[0045] In the "unfavorable" Mitoscore subgroup, fewer ROS are emitted. This is because complex IV of the respiratory chain, encoded by the COX1 / COX2 / COX3 genes, is not a site of superoxide ion emission. Furthermore, the mitochondrial-derived peptide MOTS-c, encoded by the 12S gene, appears to play a role in metabolic adaptation to stress. Further functional studies are needed to evaluate its impact and to develop therapeutic strategies targeting MOTS-c.

[0046] Mitoscore+ was also functionally characterized in the same way as Mitoscore. In the subgroup of patients called Mitonaïf, patients whose leukemic cells are able to produce more ROS in the presence of antimycin / rotenone and DPI (diphenyleneiodonium) have significantly better overall survival compared to patients whose leukemic cells produce less ROS. Definition of the Mitoscore:

[0047] Mitoscore allowed the 64 patients with AML to be stratified into 3 prognostic groups in terms of overall survival based on variant combinations following:

[0048] A “favorable” Mitoscore corresponds to the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes or at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB genes, and the absence of mutation in the following genes COX1 / COX2 / COX3 / 12S.

[0049] An “unfavorable” Mitoscore corresponds to the presence of at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes, and the absence of mutation in the following genes ND2 / ND3 / ATP8 / CYTB / ND4.

[0050] The "intermediate" Mitoscore corresponds to: - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes and at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes. This is the subgroup of patients called "mutant intermediate"; - or to the absence of mutations in the ND2 / ND3 / ATP8 / CYTB / ND4 / COX1 / COX2 / COX3 / 12S genes. This is the subgroup of patients called "Mitonaïf", corresponding to about 30% of patients with AML.

[0051] [Table 1] Mitoscore Variants selected according to the Work Flow fflmi «afK SUV Definition of Mitoscore+:

[0052] In the subgroup of patients called Mitonaïf of the "intermediate" Mitoscore, the ELN 2017 stratification was applied to determine the Mitoscore+ which made it possible to reclassify these 30% of patients with AML into 3 prognostic groups in terms of overall survival:

[0053] A “favorable” Mitoscore+ corresponds to:

[0054]

[0055] - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes or at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB genes, and the absence of mutation in the following genes COX1 / COX2 / COX3 / 12S; - or to Mitonaifs patients classified as "favorable" according to the ELN 2017 score. An "unfavorable" Mitoscore+ corresponds to: - either the presence of at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes, and the absence of mutation in the following genes ND2 / ND3 / ATP8 / CYTB / ND4; - or to Mitonas patients classified as "unfavorable" according to the ELN 2017 score. The "intermediate" Mitoscore+ corresponds to: - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes and at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes; - or to Mitonaid patients classified as "intermediate" according to the ELN 2017 score. [Tables 2] Mit o scorePlus Selected variants seîonie Work F(ow ^^:¾¾¾¾¾¾¾¾¾¾¾¾ ^0)777777777777777777777777777^ ^OSBBiBBBBB^B^BB^BBBBBBBB^BIBBBBBBBBBBBB COXVCOK2 / CX3 / 12S and No valid te* .genes

[0056]

[0057] List of references 1. Dohner et al., Blood, 129(4):424-447, 2017 2. Herold et al, Leukemia. 2020 Dec;34(12):3161-3172 3. Farge et al., Cancer discov. 2017 Jul;7(7):716-735 4. Mondet et al, Hematologica 2019 Sep;104(9):e393-e397 5. Damm et al., Leukemia, 26: 289-295, 2012 6. Chun et al., Chin. J. Hematol., 35(8): 708-712, 2014 7. Wu et al., Scientific reports, 8: 13301, 2018 8. Caudron-Herger and Diederichs, Biology, 15: 62-69, 2018 9. Zeng et al., Scientific reports, 8: 3623, 2018

Claims

Demands

1. An in vitro method for establishing a survival prognosis in a patient with chemosensitive cancer, said method comprising the following steps: - detection in a biological sample of said patient, of the presence and / or absence of at least one mutation on one of the nine genes of the following mitochondrial genome: ND2, ND3, ND4, CYTB, ATP8, C0X1, C0X2, C0X3 and 12S; - comparison of detection results with respect to a reference biological sample or a reference sequence or a reference ha-plogroup / haplotype; - establishing the survival prognosis of said patient; where: - the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, ND4 and the absence of mutation in the genes C0X1, C0X2, C0X3, 12S, is the indication of a favorable survival prognosis; - The presence of at least one mutation in one of the genes C0X1, C0X2, C0X3, 12S, and the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, indicates an unfavorable survival prognosis; and - either the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, C0X1, C0X2, C0X3, 12S, or the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, ND4 and at least one mutation in one of the genes C0X1, C0X2, COX, 12S, is the indication of an intermediate survival prognosis.

2. The method according to claim 1, where: - the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, and the absence of mutation in the genes C0X1, C0X2, C0X3, 12S, is an indication of a favorable survival prognosis; - The presence of at least one mutation in one of the genes C0X1, C0X2, C0X3, 12S, and the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, indicates an unfavorable survival prognosis; and - either the absence of mutations in the genes ND2, ND3, ATP8, CYTB, ND4, C0X1, C0X2, C0X3, 12S, or the presence of at least one mutation in one of the genes ND2, ND3, ATP8, CYTB, ND4 and at least one mutation in one of the genes C0X1, C0X2, COX, 12S, is

3.

4.

5. the indication of an intermediate survival prognosis. A method according to any one of claims 1 or 2, wherein the method further comprises determining the EuroLeu-kemiaNet 2017 (ELN 2017) prognostic score in the subgroup of patients with no mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes. Method according to claim 3, where: - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes and the absence of mutation in the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and a "favorable" classification according to the ELN 2017 score, is the indication of a favorable survival prognosis; - either the presence of at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes and the absence of mutation in the ND2 / ND3 / ATP8 / CYTB / ND4 genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and an "unfavorable" classification according to the ELN 2017 score, is an indication of an unfavorable survival prognosis; - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB / ND4 genes and at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and an "intermediate" classification according to the ELN 2017 score, is the indication of an intermediate survival prognosis. Method according to claim 3, where: - either the presence of at least one mutation in one of the ND2 / ND3 / ATP8 / CYTB genes and the absence of mutation in the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and a "favorable" classification according to the ELN 2017 score, is the indication of a favorable survival prognosis; - either the presence of at least one mutation in one of the COX1 / COX2 / COX3 / 12S genes and the absence of mutation in the ND2 / ND3 / ATP8 / CYTB / ND4 genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, COX1, COX2, COX3, 12S genes and an "unfavorable" classification according to the ELN 2017 score, is an indication of an unfavorable survival prognosis; - or the presence of at least one mutation in one of the genes ND2 / ND3 / ATP8 / CYTB / ND4 and at least one mutation in one of the The absence of mutations in the COX1 / COX2 / COX3 / 12S genes, or the absence of mutations in the ND2, ND3, ATP8, CYTB, ND4, C0X1, C0X2, C0X3, 12S genes and an "intermediate" classification according to the ELN 2017 score, is an indication of an intermediate survival prognosis.

6. A method according to any one of claims 1 to 5, wherein the detection steps are carried out by high-throughput sequencing of the mitochondrial genome.

7. An in vitro method for predicting or evaluating the efficacy and / or benefit of a treatment for a chemosensitive cancer in a patient with said cancer, comprising the following steps: - determining a survival prognosis according to a method as defined in any one of claims 1 to 6 from a biological sample of said patient before treatment; - determining a survival prognosis according to a method as defined in any one of claims 1 to 6 from a biological sample of said patient after treatment; - comparing the survival prognoses before and after treatment;- determination of either resistance to treatment when the prognosis for survival of said patient after treatment is the same as or worse than the prognosis for survival of said patient before treatment, or of sensitivity to treatment when the prognosis for survival of said patient after treatment is better than the prognosis for survival of said patient before treatment.

8. A method according to any one of claims 1 to 7, wherein the chemosensitive cancer is selected from the group consisting of acute myeloid leukemias (AML), sarcomas, testicular cancer, choriocarcinomas, hematological malignancies, ovarian cancer, and breast cancer.