WO2012163941A2

WO2012163941A2 - Marker for the detection and classification of leukemia from blood samples

Info

Publication number: WO2012163941A2
Application number: PCT/EP2012/060097
Authority: WO
Inventors: Oliver LAULE; Stefan BLEULER; Philip Zimmermann
Original assignee: Nebion Ag
Priority date: 2011-05-30
Filing date: 2012-05-30
Publication date: 2012-12-06
Also published as: WO2012163941A3

Abstract

The invention relates to a method for detecting leukemia from patient samples, preferably blood samples, and to preferably classify it into specific types of leukemia, comprising the steps of: (a) determining the expression level of at least one marker gene; and (b) comparing the expression level of the at least one marker gene to a reference to predict leukemia, preferably a specific type of leukemia; wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI. The invention further relates to a kit to perform the method, to the use of the marker genes and to a reagent for detecting leukemia.

Description

Marker for the detection and classification of leukemia from blood samples

TECH N ICAL FI ELD

The present invention is related to a method and a kit for detecting leukemia from patient samples and to preferably classify it into specific types and subtypes of leukemia. The invention further relates to the use of at least one marker gene for detecting leukemia and a reagent for detecting leukemia.

PRIOR ART

Leukemia is a disease of the bone marrow and is characterized by an abnormal proliferation of white blood cells. Leukemia belongs to the haematological neoplasms. It is subdivided into several clinically and pathologically classified subgroups. A first classification level divides it into its acute and chronic forms (acute leukemia and chronic leukemia) . A second level of classification refers to the affected cell type, such as lymphoblastic or lymphocytic leukemia and myelogenous or myeloid leukemia. From these two levels, four main types of leukemia therefore have been defined: (a) Acute lymphoblastic leukemia (ALL), (b) Acute myeloid leukemia (AM L), (c) Chronic lymphocytic leukemia (CLL), and (d ) Chronic myeloid leukemia (CM L) .

Other types or subtypes of leukemia exist, such as T-cell prolymphocytic leukemia, Hairy cell leukemia, adult T-cell leukemia, and several others based on histopa- thological and staging characteristics. Leukemia is usually diagnosed by physical examination, blood examination, bone marrow puncture, bone marrow biopsy and blood testing . A number of technical approaches exist to identify and classify leukemia from harvested body samples (bone marrow or blood), including complete blood count, blast cell count, hematocrit assays, microscopy examination (cell morphology), cell biochemical staining (immunophenotyping), flow cytometry, and cytogenetic tests. Once the medical diagnosis is made, the potential spread to other organs (metastasis) is usually evaluated using diagnostic imaging tools.

The diagnosis of leukemia therefore involves several steps that, in combination, provide sufficient information to correctly diagnose and classify leukemia such as to direct an appropriate therapy. While some of the methods can be performed on blood, several of them are more invasive, requiring bone marrow aspiration and/or biopsy. Samples of liquid bone marrow and bone marrow biopsies are subsequently examined by specialists, including a pathologist, a hematologist and/or an oncologist. I n particular, they perform several tests involving microscopic examination, cytochemistry, flow cytometry, immunocytochemistry, cytogenetic testing, and immunophenotyping. The precise and reliable diagnosis of leukemia therefore involves several labor-intensive and expensive steps requiring the expertise from multiple persons. Additionally, they generally involve invasive steps that may cause pain to the patient, such as bone marrow biopsies. Finally, the combination of multiple diagnostic steps increases the time needed for the final diagnosis, which can require up to several days. Recent developments in molecular measurement techniques have raised the hopes to detect diseases with minimally invasive methods, such as measuring nucleic acids, proteins or metabolites from a patient's blood samples. Ideally, a single diagnostic test measuring one or a plurality of molecular signatures from blood samples could allow identifying precisely the presence of a disease and its malignancy, progression, metastasis formation, and responsiveness to distinct treatment strategies.

EXPLANATION OF TH E I NVENTION

It is an objective of the invention to provide a method for detecting leukemia, preferably specific leukemia types, with a single, non-invasive test that can be carried out in a short time for immediate and reliable diagnosis.

The objective of the present invention is achieved by a method according to claim 1 for detecting leukemia from patient samples and to preferably classify it into specific types of leukemia. The method of the invention comprises the steps of: (a) determining the expression level of at least one marker gene; and (b) comparing the expression level of the at least one marker gene to a reference, wherein an up-regulated expression level of the at least one marker gene predicts leukemia, preferably a specific type or subtype of leukemia; wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence (SEQ I D NOs) listed in one or more of Tables I to VI . Within the present specifications of the invention, marker genes and contrasting marker genes (or contrasting genes) are represented by probe sets that match to a target DNA sequence (target sequence; cf. Table VI I I ) from said marker gene or contrasting gene. The identifiers of the probe sets of the present invention refer to the identifiers of Affymetrix Probe Sets from Affymetrix, Inc., USA, and are unique. The at least one marker gene can also be identified or is represented either by the probe set or by the target sequence.

With the present invention the expression levels of at least one marker gene in a patient's sample is determined . If the patient's sample is affected with leukemia or a specific type of leukemia the expression level of the marker gene is up-regulated compared to the sample if it would not be affected with the disease. To predict if leukemia or a specific type or subtype of leukemia is present the possibly up- regulated expression level of the at least one marker gene is compared to a reference. The reference can be a control sample, e.g. from a healthy person, where the expression level of the same marker gene is determined . In this case a significant deviation of an up-regulated expression level of the marker gene in the patient's sample to the expression level of the marker gene in a healthy sample predicts the disease.

Alternatively, the reference can be at least one internal reference gene within the patient's sample, which is invariant between leukemic and non-leukemic samples or which is down-regulated in leukemic samples, e.g . a contrasting gene of Table VI I . Such a reference would avoid the need of a control sample from a healthy person. A patient sample is positive if the ratio of expression level between the at least one marker gene and the internal reference gene or contrasting gene is beyond an empirically defined threshold, e.g . 1 0-fold. The threshold would have to be determined during clinical validation of a given kit for detecting leukemia from patient samples. However, there are other statistical methods that could be used to classify healthy from diseased samples.

Another alternative can be a template signature as reference, when determining the expression levels of at least two or more marker genes. The expression levels of the two or more marker genes (also called expression signature) is then compared with the typical expression signatures (template signature) of these genes in the blood of diseased and healthy patients, respectively, wherein the higher similarity of the signature of the tested sample to the typical signature of diseased patients predicts leukemia, preferably a specific type of leukemia.

The method is performed outside of the human body in vitro. The patient sample is preferably a blood, preferably peripheral blood mononuclear cells ( PBMC), or bone marrow sample. I n the case of a control sample as reference, the control is preferably a blood sample, preferably peripheral blood mononuclear cells ( PBMC), or bone marrow sample of a healthy person .

In another aspect of the method of the invention ALL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I and/or is represented by at least one probe set or at least one target sequence listed in Table I . In another aspect of the method of the invention AM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I .

In another aspect of the method of the invention CLL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I I .

In another aspect of the method of the invention CM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table IV and/or is represented by at least one probe set or at least one target sequence listed in Table IV.

In another aspect of the method of the invention hairy cell type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table V and/or is represented by at least one probe set or at least one target sequence listed in Table V.

In another aspect of the method of the invention multiple types leukemia are detected and the at least one marker gene corresponds to at least one of the genes listed in Table VI and/or is represented by at least one probe set or at least one target sequence listed in Table VI . In another aspect of the method of the invention the expression levels of a group of at least two marker genes are determined and compared to a reference, wherein each marker gene of the group of at least two marker genes corresponds to one of the genes listed in Tables I to VI and/or is represented by one probe set or one target sequence listed in Tables I to VI . Each selected maker gene can be a marker gene specific for a different specific type of leukemia. With the method it is possible to identify the specific type of leukemia from one patient's sample.

In another aspect of the method of the invention at least one contrasting gene corresponding to at least one of the genes listed in Table VI I and/or being represented by at least one probe set or at least one target sequence listed in Table VI I is used to increase the discriminating power of the method for detecting leukemia from patient samples. The contrasting genes are genes that are down-regulated in the presence of leukemia or a specific type of leukemia. The contrasting genes can be used as the reference in step b or additionally to the reference in step b. The at least one contrasting gene of Table VI I is preferably used to increase the discriminating power of the method for detecting a specific type of leukemia specified in Table VI I .

The invention further relates to a kit for detecting leukemia from patient samples and to preferably classify it into specific types of leukemia, the kit comprising at least one reagent for determining the expression level of at least one marker gene and optionally at least one contrasting gene, wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI and wherein the at least one contrasting gene corresponds to at least one of the genes listed in Table VI I and/or is represented by at least one probe set or at least one target sequence listed in Tables VI I . The at least one reagent or the reagents can be the listed probe sets, or other probes that can be used to assess the expression level of the at least one marker gene. In particular, the reagents can be probes to measure nucleic acids, preferably gene transcripts ( mRNA), as well as the products (preferably proteins) and regulators (preferably microRNAs and other small RNAs) of the at least one marker gene. The reagents are preferably oligonucleotides that match part of the sequence of the at least one marker gene and are used to determine the expression levels of the at least one marker gene, e.g . by RT-qPCR or microarrays.

Preferably, the kit comprises at least two, preferably three or more, marker genes and preferably one or more contrasting gene. With the kit the expression levels of several marker genes can be measured to determine an expression signature of the selected marker genes. By comparing the expression signature to a reference the specific type of leukemia can be identified from a patient's sample.

The invention further relates to the use of at least one marker gene and/or at least one contrasting gene for detecting leukemia from patient samples, preferably a specific type of leukemia, wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI . In another aspect of the inventive use, ALL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I and/or is represented by at least one probe set or at least one target sequence listed in Table I .

In another aspect of the inventive use, AM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I .

In another aspect of the inventive use, CLL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I I .

In another aspect of the inventive use, CM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table IV and/or is represented by at least one probe set or at least one target sequence listed in Table IV.

In another aspect of the inventive use, hairy cell type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table V and/or is represented by at least one probe set or at least one target sequence listed in Table V. In another aspect of the inventive use, multiple types leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table VI and/or is represented by at least one probe set or at least one target sequence listed in Table VI .

The invention further relates to the a reagent for detecting leukemia from patient samples and to preferably classify it into specific types of leukemia, the reagent being suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI .

The reagent can be one of the listed probe sets, or other probes that can be used to assess the expression level of the at least one marker gene. In particular, the reagent can be a probe to measure nucleic acids, preferably gene transcripts (mRNA), as well as the products (preferably proteins) and regulators (preferably microRNAs and other small RNAs) of the at least one marker gene. The reagent is preferably an oligonucleotide that matches a part of the sequence of the at least one marker gene and are used to determine the expression levels of the at least one marker gene, e.g. by RT-q PCR or microarrays.

In another aspect of the invention, the reagent is for detecting ALL type leukemia and is suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in Table I and/or is represented by at least one probe set or at least one target sequence listed in Table I .

In another aspect of the invention, the reagent is for detecting AM L type leukemia and is suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in Table I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I .

In another aspect of the invention, the reagent is for detecting CLL type leukemia and is suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in Table I I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I I .

In another aspect of the invention, the reagent is for detecting CM L type leukemia and is suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in Table IV and/or is represented by at least one probe set or at least one target sequence listed in Table IV.

In another aspect of the invention, the reagent is for detecting hairy cell type leukemia and is suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in Table V and/or is represented by at least one probe set or at least one target sequence listed in Table V.

In another aspect of the invention, the reagent is for detecting multiple types leukemia and is suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in Table VI and/or is represented by at least one probe set or at least one target sequence listed in Table VI . The marker genes are represented by probe sets that match to a target DNA sequence (target sequence; cf. Table VI I I ) from said marker genes. The identifiers of the probe sets of the present invention refer to the identifiers of Affymetrix Probe Sets from Affymetrix, I nc., USA, and are unique.

In one aspect, the invention relates to at least one molecular signature, comprising at least one marker gene and optionally at least one contrasting gene, measured from human blood samples, to detect leukemia and to classify it into specific types of leukemia. The invention covers several areas of the method, in particular: (a) the marker gene (and optionally the used contrasting gene) is detected in blood and associated with the disease and its specific types, (b) the choice of marker genes allows to distinguish between healthy and diseased samples, including disease types and subtypes; (c) the marker genes and optionally the contrasting genes can be grouped into diagnostic signatures; (d ) the use of the contrasting genes increase the power of the diagnostic test; and (e) the method can be applied in a clinical environment, in particular for the detection and classification of leukemia in blood from human patients. Type of molecule measured in blood samples

The present invention relates to the measurement of at least one marker gene in human blood, preferably from peripheral blood mononuclear cells, to detect signatures associated with leukemia. In particular, it relates to the measurement of nucleic acids, preferably gene transcripts (mRNA), as well as the products (preferably proteins) and regulators (preferably microRNAs and other small RNAs) of these marker genes.

Molecules associated with the disease

The present invention relates to the following marker genes for the detection of leukemia: a) ALL type leukemia. The marker genes from Table I are strongly expressed, or up-regulated, in the blood of ALL type leukemia patients and weakly expressed in the blood of healthy patients (control). These genes may, but must not, be weakly expressed in thymus and bone marrow of healthy patients. The invention relates in particular to the genes represented by the Affymetrix probe sets: 220389_at (ENSG00000149201 , CCDC81, coiled-coil domain containing 81; target sequence: SEQ ID NO:1), 215117_at (ENSG00000175097, RAG 2, recombination activating gene 2; target sequence: SEQ ID NO:2), 208302_at (ENSG00000158497, HMHB1, histocompatibility (minor) HB-1; target sequence: SEQ ID NO:3), 213060_s_at (ENSG00000064886, CHI3L2, chitinase 3-like 2; target sequence: SEQ ID NO:4), 240179_at (uncharacterized LOC100505801; target sequence: SEQ ID N0:5), and 1556598_at, ENSG00000172995, ARPP21, cAMP-regulated phosphoprotein, 21kDa; target sequence: SEQ ID NO:49). The genes represented by the Affymetrix probe sets 220389_at and 240179_at are preferred as single marker genes or in combination with other marker genes from Tables I to VI. The genes represented by the probe sets 215117_at, 208302_at, 213060_s_at, and 1556598_at are preferred in combination with other marker genes indicated in Tables I to VI. b) AMLtype leukemia. The marker genes from Table II are strongly expressed, or up-regulated, in the blood of AML type leukemia patients but are weakly or non- expressed in other leukemia types and in healthy blood samples (control). The invention relates in particular to the genes represented by the Affymetrix probe sets: 1553613_s_at (FOXC1 (forkhead box C1 ); target sequence: SEQ ID NO:6) and 240766_at (Homo sapiens cDNA FLJ43834 fis, clone TESTI4005801 ; target sequence: SEQ ID NO:7), 236738_at (ENSG00000180044, C3orf80, chromosome 3 open reading frame 80; target sequence: SEQ ID NO:8), 236892_s_at (ENSG00000233101, Hs.660088, HOXB-AS3, HOXB cluster antisense RNA 3; target sequence: SEQ ID NO:9). The gene represented by the Affymetrix probe set 236892_s_at is preferred as single marker genes or in combination with other marker genes from Tables I to VI. The genes represented by the probe sets 1553613_s_at, 240766_at, and 236738_at are preferred in combination with other marker genes indicated in Tables I to VI. c) CLL type leukemia. The marker genes from Table III are strongly expressed, or up-regulated, in the blood of CLL type leukemia patients but are weakly or non- expressed in other leukemia types and in healthy blood samples (control). The invention relates in particular to the genes represented by the Affymetrix probe sets: 1562587_at and 241483_at (ENSG00000109684, Cytokine-dependent hematopoietic cell linker; target sequence: SEQ ID NO: 10 and SEQ ID NO:11), 205414_s_at (ENSG00000006740, ARHGAP44, Rho GTPase activating protein 44; target sequence: SEQ ID NO:12), 236184_at (Hs.687695, no description; target sequence: SEQ ID NO:13), 1562754_at (Hs2.436003.1 , hypothetical LOC339260; target sequence: SEQ ID NO:14), 239185_at (ENSG00000154258, ABCA9, ATP-binding cassette, sub-family A, member 9; target sequence: SEQ ID NO:15), 241278_at (GenBank AI674679; target sequence: SEQ ID NO: 16). The genes represented by the Affymetrix probe sets 1562587_at, 241483_at, 236184_at, 1562754_at, and 241278_atare preferred as single marker genes or in combination with other marker genes from Tables I to VI. The genes represented by the probe sets 205414_s_at and 239185_at are preferred in combination with other marker genes indicated in Tables I to VI. d) CML type leukemia. The marker genes of Table IV are strongly expressed, or up-regulated, in the blood of CML type leukemia patients but have significantly lower expression in other leukemia types and in healthy blood samples (control/reference). The invention relates in particular to the genes represented by the Affymetrix probe sets 236031_x_at (ENSG00000164946, FRAS1 related extracellular matrix 1 (FREM1); target sequence: SEQ ID NO:17), 204624_at (ENSG00000123191, ATP7B, ATPase, Cu++ transporting, beta polypeptide; target sequence: SEQ ID NO:18), and 205984_at (ENSG00000145708, CRHBP, corticotropin releasing hormone binding protein (CRH BP); target sequence: SEQ I D NO: 29) . The genes represented by the Affymetrix probe sets 23603 1 _x_at and 204624_at are preferred as single marker genes or in combination with other marker genes from Tables I to VI . The gene represented by the probe set 205984_at is preferred in combination with other marker genes indicated in Tables I to VI . e) Hairy cell type leukemia ( HCL) . The marker genes of Table V are strongly expressed, or up-regulated, in the blood of CM L type leukemia patients but have significantly lower expression in other leukemia types and in healthy blood samples (control/reference) . The invention relates in particular to the genes represented by the Affymetrix probe sets 238009_at ( Hs.434948, no description; target sequence: SEQ I D NO: 1 9 ) and 207336_at ( ENSG000001 34532, SOX5, SRY (sex determining region Y)-box 5; target sequence: SEQ I D NO: 20) . The genes represented by the Affymetrix probe sets 238009_at and 207336_at are preferred as single marker genes or in combination with other marker genes from Tables I to VI . f) Multiple types of leukemia. The marker genes of Table VI are strongly expressed, or up-regulated, in the blood in response to two or more of the types ALL, CLL, AM L, CM L, and/or Hairy Cell type leukemia patients and weakly or non- expressed in the blood of healthy patients (control) . The invention relates in particular to the genes represented by the Affymetrix probe sets 242520_s_at ( ENSG000001 98520, C 1 orf228, chromosome 1 open reading frame 228; target sequence: SEQ I D NO: 2 1 ), 1 566482_at ( Hs.684006, no description; target sequence: SEQ ID NO:22), 214575_s_at (ENSG00000172232, AZU1, azuro- cidin 1; target sequence: SEQ ID NO:23), 230551_at (ENSG00000171435, KSR2, kinase suppressor of ras 2; target sequence: SEQ ID NO:24), 213714_at (ENSG00000165995, CACNB2, calcium channel, voltage-dependent, beta 2 subunit; target sequence: SEQ ID NO:25), 1558871 _at (clone IMAGE:4105785; target sequence: SEQ ID NO:26), 219790_s_at (ENSG00000113389, NPR3, natriuretic peptide receptor C/guanylate cyclase C; target sequence: SEQ ID NO:27), 238206_at (ENSG00000171509, RXFP1, relaxin/insulin-like family peptide receptor 1; target sequence: SEQ ID NO:30), 1552715_a_at (ENSG00000171509, RXFP1, relaxin/insulin-like family peptide receptor 1 ; target sequence: SEQ ID NO:31), 236632_at (ENSG00000248890, hypothetical LOC100505470; target sequence: SEQ ID NO:32), 1559315_s_at (ENSG00000246985, hypothetical protein LOC144481; target sequence: SEQ ID NO:33). The marker gene relating to probe set 230551 _at is predominant in CLL and shows signals also in ALL. The marker gene relating to probe set 1558871 _at is predominant in AML. The marker gene relating to probe set 1552715_a_at is particularly high at blast crisis in bone marrow. The marker genes relating to probe sets 236632_at and 1559315_s_at are predominant in ALL. The genes represented by the Affymetrix probe sets 242520_s_at, 1566482_at, 213714_at, 1558871 _at, 236632_at are preferred as single marker genes or in combination with other marker genes from Tables I to VI. The genes represented by the probe sets 214575_s_at, 230551 _at, 219790_s_at, 203948_s_at, 203949_at, 238206_at, 1552715_a_at, and 1559315_s_at are preferred in combination with other marker genes indicated in Tables I to VI. Table VI also indicates the types of leukemia, in which the expression is highest, but it does not exclude expression in further types of leukemia.

Grouping of molecules into diagnostic signatures

The present invention relates to the use of these marker genes (Tables I to VI ) and/or contrasting gene (Table VI I ), alone or in combination, to present a diagnostic prediction with increased specificity and/or sensitivity. It is statistically established that the simultaneous use of multiple markers with independent classification power increases the performance of the diagnostic test. In some cases, a single marker gene may have the highest discriminating power between healthy and diseased samples, whereas in other cases, the combination of a plurality of marker genes offers the highest discriminating power. The present invention makes use of this property to select suitable combinations of marker genes for the detection of leukemia and its specific types.

Contrasting molecules The present invention further relates to the use of contrasting genes (also called reference genes) listed in Table VI I that have opposite properties, such as non- expressed, or down-regulated, in diseased samples but expressed in healthy samples (control/reference), to increase the discriminating power of the diagnostic test. I n particular, it relates to the use of such genes either for correcting the values of the marker genes, or for the generation of ratios between marker genes and contrasting genes. The contrasting genes can also directly be used as refer- ence in step b of the method . Table VI I further indicates the preferred leukemia type for which the contrasting gene is especially well suited .

Application in a clinical environment

The present invention relates to the use of at least one of the aforementioned marker genes and optionally at least one of the contrasting molecules to create a diagnostic test to detect diseases, preferably cell proliferating diseases, preferably leukemia, more preferably specific types of leukemia, in human blood samples.

BRI EF EXPLANATION OF TH E FIG U RES

The invention is described in greater detail below with reference to embodiments that are illustrated in the figures. The figures show:

Fig . 1 Expression evidence for the selected marker genes (Tables l -VI ), based on results obtained from Genevestigator. The plots represent the expression level of individual markers across 5941 Affymetrix Human 1 33 Plus 2.0 arrays. U nder (a) Acute Lymphocytic Leukemia (ALL); (b) Chronic Lymphocytic Leukemia (CLL); (c) Acute Myelocytic Leukemia (AM L); (d) Chronic Myelocytic Leukemia (CM L); (e) Hairy Cell Leukemia ( HCL); and (f) Multiple leukemia types.

Fig . 2 Heat map representing the average log-expression values of all marker genes across different blood tissues, leukemia types and subtypes, and different types of lymphatic tissue neoplasms. The number of arrays taken to process each average is indicated on the right of the heat map. DETAI LED DESCRI PTION OF TH E I NVENTION

Materials and Methods

Identification of marker genes:

Gene expression biomarkers were identified using Genevestigator ( Hruz et al ., 2008) . The search for molecular markers was performed by comparing the average expression level in selected "target" categories with the average expression level in selected "base" categories. "Target" categories refer to individual or multiple diseases in which marker genes are strongly expressed . "Base" categories, which include the target categories, refer to the complete set of categories the analysis is done with. Marker genes that are highly expressed in the "target" categories are expected to be weakly or non-expressed in the other "base" categories. As "base" categories, we chose either:

- Selection A: leukemic and non-leukemic blood and cells extracted from blood;

- Selection B: selection A + leukemic bone marrow samples;

- Selection C: selection B + blood samples having a variety of lymphatic system neoplasms; and

- Selection D: selection C + all cancer samples and all non-cancer tissue samples available from the Genevestigator database in May 201 2. In an initial search, we looked for genes that are highly expressed in leukemia samples but are not expressed (or minimally expressed ) in other cancer types and healthy tissue types. I n this case, the "base" categories comprise all categories, i .e. corresponds to selection D above.

- Target: separately ALL ( B-ALL; c-ALL; T-ALL) or CLL ( B-lineage) or AM L or

CM L or Hairy Cell Leukemia

- Base: all categories of selection D above

In a second search, we identified genes that are expressed in specific types or subtypes of leukemia but are not expressed in healthy blood samples or healthy cells extracted from blood (corresponds to selection A above) and in other hematologic diseases (corresponds to selection C above) . We verified expression of these genes in leukemic and healthy bone marrow samples (corresponds to selection B above) to confirm their response to the leukemia type chosen as target.

- Target: separately ALL ( B-ALL; c-ALL; T-ALL) or CLL ( B-lineage) or AM L or

CM L or Hairy Cell Leukemia

- Base: blood leukemia (all available subtypes), blood, platelet, peripheral blood leukocyte, peripheral blood mononuclear cells (all), monocyte derived dendritic cell, leukocyte, B-lymphocyte, monocyte, natural killer T-cell, T-lymphocytes, macrophages, CD4 - and CD8 T-lymphocytes In a third step, we filtered the candidate marker genes (between 1 0 and 1 00, depending on leukemia subtype) against the Human Perturbations datasets available in Genevestigator to exclude genes that are up-regulated in response to common diseases, infections, drugs or other conditions. This dataset contained at the time of analysis more than 1 ,000 different perturbations, including response to drugs, diseases, hormones, stress, mutations, and a group of miscellaneous conditions. This step is to maximize disease specificity and minimize the probability of false diagnosis caused by other conditions affecting the tested patient. This filtering resulted in a small number of genes that are highly specifically expressed in a chosen type or subtype of leukemia, weakly or non-expressed in any other disease, and weakly or non-responsive to common conditions such as viral or bacterial infection, to commonly used drugs or other common conditions.

The figures 1 and 2 were generated from figures or data exported from the Genevestigator analysis tool ( https://www.genevestigator.com; Nebion AG, Switzer- land ) .

Mappings of probe set identifiers to gene model identifiers were obtained using publicly available Bioconductor packages. Transcript sequences were obtained from the array manufacturer (Affymetrix, NetAffx™ Analysis Center) .

Blood collection and Peripheral Blood Mononuclear Cell separation: The collection of blood, the separation of peripheral blood mononuclear cells ( PBMC) and their storage were performed according to the COLOX Clinical Study protocol (www.diagnoplex.com) . CPT VACUTAI N ER TU BES ( Becton Dickinson ) were used to collect blood in collection centers. Within 6 hours, PBMC were extracted by centrifugation and washed twice in 1 x PBS. PBMC were stored in RNA later ( [kit provider] ) at -20 °C until shipment to the test laboratory, where they were stored at -80°C.

RNA extraction:

An equivalent of 700Ό00 to 2Ό00Ό00 PBMC were used for total RNA extraction using RNeasy Mini Kit and Qiacube (Qiagen) for automatization . 600 ng to 3 g of total RNA were obtained in a final elution volume of 30 μΙ_. A Bioanalyzer (Agilent) and a NanoDrop (Thermo Scientific) were performed for quality and quantity check. RNA was stored at -80 °C.

Reverse Transcription:

For reverse transcription, we used 200 ng of total RNA and Superscript VI LO cDNA Synthesis Kit ( Invitrogen) . The final volume of cDNA was 20 μΙ_. The positive control was 200 ng of Human Universal Reference Total RNA (Clontech) . The negative control was H₂0 ( Roche) . The program in the Thermocycler was set according to the Superscript VI LO Technical Data Sheet. cDNA was diluted 1 : 2 in H₂0 ( Roche) and stored for 24 hours at - 20 °C

Real-Time PCR: Primers (forward and reverse) and TaqMan probes ( FAM/TAM RA) for the tested marker genes were designed, whereas the TaqMan assays related to the reference genes were purchased from Life Technologies. I n each well, 2 μΜ of each primer and 2.5 μΜ of probe was added . Taq Man Fast Advanced Master Mix ( Life Technologies) was used at 1 x concentration, and the final reaction volume per well was 1 0 μί. Plate Loading was automated with the CAS 1 200 (Corbett Robotics) adapted to 384-wells plates and real-time PCR was performed using the VI IA7 instrument platform ( Life Technologies) . The PCR program with 45 cycles was set according to the Technical Data Sheet of the Master Mix. Ct values were obtained after threshold adjustment to 0.05 Delta Rn.

Quality Control:

The quality control of the experiment was performed by checking the positive controls for each run of RT-q PCR. The 7 1 subjects were divided in 54 runs of RT- qPCR. Therefore 4 independent runs of RT were performed with 1 positive and 1 negative control per run.

Results

Marker genes and contrasting genes identified from Genevestigator are listed in Tables I - VI I . The corresponding target sequences relating to the genes are listed in Table VI I I .

Fig . 1 shows the expression evidence for the selected marker genes, based on results obtained from Genevestigator ( Nebion AG, Switzerland ) . U nder Fig. 1 (a) for Acute Lymphocytic Leukemia (ALL); Fig.1 (b) for Chronic Lymphocytic Leukemia (CLL); Fig.1(c) for Acute Myelocytic Leukemia (AML); Fig.1(d) for Chronic Myelocytic Leukemia (CML); Fig.1 (e) for Hairy Cell Leukemia (HCL); and Fig.1 (f) for Multiple leukemia subtypes. The numbers of the probe sets refer to the identifiers of Affymetrix Probe Sets from Affymetrix, Inc., USA, and are unique. The plots represent the expression level of individual markers across 5941 Affymetrix Human 133 Plus 2.0 arrays, with the following order of the samples: 1-1344: Non-malignant (blood, 1344 samples); 1345-1590: ALL (blood, 246 samples); 1591-1688: CLL (blood, 98 samples); 1689-2338: AML (blood, 640 samples); 2339-2343: JMML (blood, 5 samples); 2344-2347: Hairy Cell Leukemia (blood, 4 samples); 2348-2947: Lymphomas (blood, 600 samples); 2948-4027: ALL (bone marrow, 1079 samples); 4028-4449: CLL (bone marrow, 422 samples); 4450-5753: AML (bone marrow, 1304 samples); 5754-5788: APL (bone marrow, 35 samples); 5789-5843: CML (bone marrow, 55 samples); 5844-5876: JMML (bone marrow, 33 samples); and 5877-5941: CML (bone marrow, 65 samples).

Fig. 2 shows a heat map representing the average log-expression values of all marker genes across different blood tissues, leukemia types and subtypes, and different types of lymphatic tissue neoplasms. The number of arrays taken to process each average is indicated on the right of the heat map. The circles and the roman numbers indicate the marker genes specific for types and subtypes of leukemia (Table I to V).

Output from quality control: As a proof of principle, one novel marker gene (C 1 orf228, see Table VI ), two published marker genes ( IG LL1 and VPREB 1 ) and three reference genes ( NACA, RPLPO, TPT1 ) were selected for quality control experiments and validation using RT-q PCR. For each target gene, Ct and Standard Deviation (StDev) for positive control ( POS) were calculated and results are shown in Table 1 . The very low variations observed between the 4 independent runs show high analytical reproducibility using the defined protocol.

Table 1 : Ct and StDev values for POS and four independent runs. Over-expressed biomarkers (first three columns) and reference genes (last three columns) .

Gene expression results:

Table 2 P-values of t-tests between the expression in the clinical groups comprising healthy samples (CON ) and leukemic samples (ALL, CLL, AM L, CM L) . The test was performed on the gene expression results from one tested marker gene selected from Table VI (C1orf228), two known biomarker genes (IGLL1 and VPREB1) and three reference genes (NACA, RPLPO, TPT1). The normalization of the biomarker expression was obtained by Delta Ct methodology using the aver- age of these 3 reference genes.

References:

(1) Hruz T, Laule O, Szabo G, Wessendrop F, Bleuler S, Oertle L, Widmayer P, Gruissem W and Zimmermann P (2008) Genevestigator V3: A Reference Expres- sion Database for the Meta-Analysis of Transcriptomes. Adv Bioinformatics vol. 2008, Article ID 420747, 5 pages. DOI: 10.1155/2008/420747 Tables I to VII: Marker genes and contrasting genes are represented by probe sets that match to a target DNA sequence (target sequence; cf. Table VIII) from said marker or contrasting genes. The identifiers of the probe sets of the present invention refer to the identifiers of Affymetrix Probe Sets from Affymetrix, Inc., USA, and are unique. The sequences of the target sequences (SEQ ID NO: 1 to 49) of Tables I to VII are listed in Table VIII.

Table I: ALL type leukemia

Table II: AMLtype leukemia

Gene Gene Description Target Se¬

Probe set ID Gene Identifier quence

1553613_s_at ENSG00000054598 FOXC1 forkhead box C1 SEQ ID NO:6

240766_at Hs.449281, Homo sapiens SEQ ID NO:7

Entrez ID: 80314 cDNA FLJ43834

fis, clone TES- TI4005801 Gene Gene Description Target Se¬

Probe set ID Gene Identifier quence

236738_at ENSG00000180044 C3orf80 chromosome 3 SEQ ID NO:8 open reading

frame 80

236892_s_at ENSG00000233101, HOXB- HOXB cluster an- SEQ ID NO:9

Hs.660088 AS3 tisense RNA 3

Table III: CLLtype leukemia

Gene Gene DescripTarget Se¬

Probe set ID Gene Identifier tion quence

1562587_at ENSG00000109684 CLNK Cytokine- SEQ ID NO:10 dependent hematopoietic cell

linker

241483_at ENSG00000109684 CLNK Cytokine- SEQ ID NO:11 dependent hematopoietic cell

linker

205414_s_at ENSG00000006740 ARHGAP44 Rho GTPase SEQ ID NO:12 activating protein 44

236184_at GenBankAI021923, -- no description SEQ ID NO:13

Hs.687695

1562754_at Hs2.436003.1, -- hypothetical SEQ ID NO:14

Entrez ID: 339260 LOC339260

239185_at ENSG00000154258 ABCA9 ATP-binding SEQID NO:15 cassette, subfamily A, member 9

241278_at GenBankAI674679 -- no description SEQ ID NO:16 Table IV: CMLtype leukemia

Table VI: Multiple types of leukemia. The leukemia types of column 5 indicate the types of leukemia in which the expression is highest, but it does not exclude ex^¬ pression in further types of leukemia.

Probe set Gene Gene Gene Description Leukemia Target SeNo. Identifier type quence

242520_s_ ENSG0000 C1orf22 chromosome 1 open ALL/AML SEQ ID at 0198520 8 reading frame 228 NO:21

1566482_at Hs.684006 -- no description AML/CML SEQ ID

NO:22 Probe set Gene Gene Gene Description Leukemia Target SeNo. Identifier type quence

214575_s_ ENSG0000 AZU1 azurocidin 1 ALL/AML/ SEQ ID at 0172232 CML NO:23

230551_at ENSG0000 KSR2 kinase suppressor of CLL SEQ ID

0171435 ras 2 NO:24

213714_at ENSG0000 CACNB2 calcium channel, voltCLL SEQ ID

0165995 age-dependent, beta NO:25

2 subunit

1558871 _at clone IMAML/CML SEQ ID

AGE: NO:26 4105785

219790_s_ ENSG0000 NPR3 natriuretic peptide AML/CML SEQ ID at 0113389 receptor C/guanylate NO:27

cyclase C

238206_at ENSG0000 RXFP1 relaxin/insulin-like AML/CML SEQ ID

0171509 family peptide recepNO:30

tor 1

1552715_a ENSG0000 RXFP1 relaxin/insulin-like AML/CML SEQ ID _at 0171509 family peptide recepNO:31

tor 1

236632_at ENSG0000 -- hypothetical ALL/AML SEQ ID

0248890 LOC100505470 NO:32

1559315_s ENSG0000 -- hypothetical protein ALL/AML SEQ ID _at 0246985 LOC144481 NO:33

Table VII: Contrasting markers or reference genes. The leukemia types of column 5 indicate the types of leukemia for which the contrasting gene is especially well suited.

Probe set Gene Gene Gene DescripLeuk. Target Seq.

No. Identifier tion Type

204897_at ENSG0000 PTGER4 prostaglandin E hairy cell SEQ ID NO:34

0171522 receptor 4 leukemia

215096_s_ ENSG0000 -- esterase D hairy cell SEQ ID NO:35 at 0139684 mRNA, complete leukemia Probe set Gene Gene Gene DescripLeuk. Target Seq. No. Identifier tion Type

cds

212074_at ENSG0000 SUN 1 Sad1 and UNC84 CML SEQ ID NO:36

0164828 domain containing 1

212124_at ENSG0000 ZMIZ1 zinc finger, MIZ- CLL SEQ ID NO:37

0108175 type containing 1

225673_at ENSG0000 MYADM myeloid- CLL SEQ ID NO:38

0179820 associated differentiation marker

229560_at ENSG0000 TLR8 toll-like receptor ALL SEQ ID NO:39

0101916 8

220005_at ENSG0000 P2RY13 purinergic recepALL SEQ ID NO:40

0181631 tor P2Y, G- protein coupled,

13

221766_s_ ENSG0000 FAM46 family with seALL SEQ ID NO:41 at 0112773 A quence similarity

46, member A

204912_at ENSG0000 IL10RA interleukin 10 ALL SEQ ID NO:42

0110324 receptor, alpha

227167_s_ Hs.643605 -- CLONE=IMAGE: t-ALL SEQ ID NO:43 at 504304

202270_at NM 0020 GBP1 guanylate bindall leukemias SEQ ID NO:44

53 ing protein 1 ,

interferon - inducible, 67kDa

226272_at ENSG0000 RCAN3 RCAN family all leukemias SEQ ID NO:45

0117602 member 3

201877_s_ ENSGOOOO PPP2R5 protein phosAML SEQ ID NO:46 at 0078304 C phatase 2, regulatory subunit B^',

gamma

208091 _s_ ENSGOOOO VOPP1 vesicular, overex- AML SEQ ID NO:47 at 0154978 pressed in cancer, prosurvival

protein 1

212149_at ENSGOOOO EFR3A EFR3 homolog A CML SEQ ID NO:48

0132294 (S. cerevisiae) Probe set Gene Gene Gene DescripLeuk. Target Seq. No. Identifier tion Type

201 877_s_ ENSG0000 PPP2R5 protein phosAM L SEQ ID NO:28 at 0078304 C phatase 2, regulatory subunit B',

gamma

Table VI I I : target sequences corresponding to the genes of Tables I to VI I :

SEQ ID NO Target Sequence

1 gaacgggcttcagacaagctgtttctcctagaccagtgtgagaagtatcggcgctgcaag caatgccagaggcgcacctccaatgtgggcgagagcaacctgtggcccctgaacaagttc ctgcctggctcccggttgcttgtgtaaaactcaaagtttggctcttcgtttcccggggaa agtttttatcttttacatgtttgggggtgattgcgaaactgcgtatttttacctcagaga aaaaaatcattgtttaggtttggtgctttcattagattgcttgttaagcccttattgaat tcactcctgctttcctcccacccccaattatttcctatactagtttctgatggcagtgaa ggtgtctgaatggtcctgagggctagaacctgctgcacaggggctgggaatgggatccag cttcattatggctgctggggtgctgctgacccagccctgctgctgctccatgactttgtg tacaaaa

2 gtattactgcaatgagcatgtggagatagcaagagctctacacactccccaaagagtcct acccttaaaaaagcctccaatgaaatccctccgtaaaaaaggttctggaaaaatcttgac tcctgccaagaaatcctttcttagaaggttgtttgattagttttgcaaaagcctttcaga ttcaggtgtatggaatttttgaatctatttttaaaatcataacattgattttaaaaatac atttttgtttatttaaaatgcctatgttttcttttagttacatgaattaagggccagaaa aaagtgtttataatgcaatgataaataaagtcattctanaccctananattttgaaaata ttttacccaaatactcaatttactaatttattcttcactgaggatttctgatctgatttt ttattcaacaaaccttaaacacccagaagcagtaataatcatcgaggtat

3 tgccccgcatctcagaagccgggcaggccctgagccttctgacctcacatcctctgccac accacagtggagaaaccagaactggaggagcagccagaatgcagagaagaaaaaagaggt tctctgcatgtttggaagtcggaattggttgaagttgaagatgatgtgtatctgaggcac agctcttccctgacttataggctttgacactgctgttgaggtttgactcgaagcccagag ttttggtgtggatgagcagggacaaattgctgagcatgaagaagagtaaaattaagcaag tggaacatatgccctt

4 ctctgcagaaaccaccgtgggggcccctgcctctggccctggagctgctggacccatcac agagtcttcaggcttcctggcctattatgagatctgccagttcctgaaaggagccaagat cacgcggctccaggatcagcaggttccctacgcagtcaaggggaaccagtgggtgggcta tgatgatgtgaagagtatggagaccaaggttcagttcttaaagaatttaaacctgggagg agccatgatctggtctattgacatggatgacttcactggcaaatcctgcaaccagggccc ttaccctcttgtccaagcagtcaagagaagccttggctccctgtgaaggattaacttaca gagaagcaggcaagatgaccttgctgcctggggcctgctctctcccaggaattctcatgt gggattccccttgccaggnnggcctttggatctctcttccaagcctttcctgacttcctc ttagatcatagattggacctggttttgttttcctgcagctgttgacttgttgccctgaag ta SEQ ID

NO Target Sequence

cttggcctcacccattagaaggtaagctccataaaaggtgactatgctgtatcaaggaat

5

cttgtatagtgtctggtatttaggtaacagacaataaatattaaagaatgaatagcatat ctctattttttaacttgagttaaaccaaactccataggtgagacttctaagagcctttaa aatgctttttgaattttcataatggaggcatagtaagaattatagcctaaacatagaatc ctggtagcatcttagatggtggttaatgctttgtgtaacacactctggcacttttaaaat tgtaaatcagtttaatttttatcacaaatgtccaacacttttctcttatgaa aacctgcaagccatgagcctgtacgcggccggcgagcgcgggggccacttgcagggcgcg

6

cccgggggcgcgggcggctcggccgtggacgaccccctgcccgactactctctgcctccg gtcaccagcagcagctcgtcgtccctgagtcacggcggcggcggcggcggcggcggggga ggccaggaggccggccaccaccctgcggcccaccaaggccgcctcacctcgtggtacctg aaccaggcgggcggagacctgggccacttggcgagcgcggcggcggcggcggcggccgca ggctacccgggccagcagcagaacttccactcggtgcgggagatgttcgagtcacagagg atcggcttgaacaactctccagtgaacgggaatagtagctgtcaaatggccttcccttcc agccagtctctgtaccgcacgtccggagctttcgtctacgactgtagcaagttttgacac accctcaaagccgaactaaatcgaaccccaaagcaggaaaagctaaaggaacccatcaag gcaaaatcgaaa

agtgcatatcttggtaaggctctgtgaggaaatgacctaatctggagagaagctgctggg

7

acactcntctgtgtgtggcctgtggctcacaggggcaggtattctcagaacgcacatggg ataatcaccatcctggtgtgtccaggctctgggaacaggctcttcctcgtggctttcaca tgagtttcctggtgggctatgtagtcacttttctgcatccttcttttaaaggctcatgaa tgtcaaagtaacacagaccctgagatgaggcaggaaagttgtatcggaatgttttcagac tatcaaccagaccaaacgttctggaatccataa gtaatgtacatatgcatattgtctatgtactatatacacattgtttataatattgttatt

8

acagtttgtcattcacctgaaaggagaaggaaaagtacgaaaggtttctctgcttgggaa aaggtgaatgttgttatatcaagaacgattacacacatggntctcatacatgnttttaga acattgttcttctgattgaagaagtctgatgctcctgaaaaatcttaaaatatctgactt gtattgaagaaaattatttaattaaatttttaaaggctggttgaaaaagtctgacagttc tgagaattttttaaatgtctgaattgtaataaaaaaatggtttactttaaacttctaaaa actaatgaccttgtgactaa ggtagacttcgcttgaaaagatcgagaatggaggaggggatcggcactcacttgcagaca

9

tcaacagtttccagaaaattcggttcaattccttcaccacggacccaccaaccaaggagc tggcaaaacccactaaccaaccaggagaacacagacccggtttgtcttttgacaaccgag tggaactgacggg

aacattctagtaactggccactggaaaataaacaaaaataaaaactagggttttaaaagt

10

atcttctaaaaaacaacaacaaaaaatactataaacatagccattatgctcatgatacag gcgagcagcaaagggcaccagaagctgttgcttaaatgtttgcagccagtgcaagacaag tctatgggaaattcccaaatctgtgctctttacaggacactgcgctgcctttatgtcagt tgttgggccttacctatctacaatgtgtg

gtactgcatgtgtattcattactaagcactattgtagatgttgaaaatagagcagtgaat

11

aagagagatagatgacaaataaatatacaaacaaatnagataannnaanncagtttccca ttactggattggatttctattttaacacttatttctgagtacttgtggggtgatgccaac attaaatgttttgaggaagctcctttgagatggctttgaaaagtcctaagtgagaaaant ttaaaanttttcccaaacttatttatgtggggtgtgcatgtgtgtgcctgtgcctgtgtg cgtgtgtgtgtgtgtgtttttaagtaaccgncatcatggctgtaatcagttaagtctcct gttctaattactacggaggaatcaaatgaaccccattgtcgaatgtagccaattatgtat gtgtgtacacaaatacaaatttttatgatagcctatattgtatattagcattatatttgt atatcttttttattctgagcaatatttgttctagctaagca SEQ ID

NO Target Sequence

attacgttacccaaagcatgtggaggaaaagtgaaaccagccacggagacgctggcccac

1 2

ggctcggcctgcggtgtggcctgctttgctcaccagcgtcagccgctcatttccttctca tgaagtcccatctggtcatggggacgagggccgggagggcaccgggtagccttttcacac ttggggattaggggagtgagaaaagatttgggccatgcatgcaaagtcaaagtttaaaat tttatccttttcaaatagatgatataatatacctatacatgatataatatttgtatatat gaaatctctctatatttgtttaatttgagccattcaatctaaaccaatgtacaggtgtac aatgaaaaatttaaatgcttagttatttttcccaacacagtgtaaagtcaccctcctctg agagtgggatgtgcagagttttgatgttgcagctttgctcacttcctggcaagggcaggt catgcctcaatttgtaatgggagtctggggtaagggtg

aaatcagtcttaacgctcagagataaataaccacattcatgacaatgagattatgtaatt

1 3

gttttttattcacaaatattttctggcacacacattaattcacagaaagaaagaacagta gtcacgtgggcacaatatttaaccctttcagatactgaggtttatcaccgcttgatttct tagttggaaatgttgcagtttggtttcaggcaggatttccaggctatcacagcctgctcc gatggcatgattattaggagctcatctgcatcactggtgaatttctttggtcacttctgg acagtttttaaacataagtccatgcccttcttt aactttcaggtctcggcccaaacatcgggggtctcctctgaccacacagtagtactgggc

1 4

cctcctttgtactgctagtgttactttcttttgaaacttttaattacagaacttaccaca catatacatagagaggactgcacaatgatgccctcctcatccattaccagcttcacaagt catttaaattctgccttgtactttcccaaccagttgttttggcttgtgtattttaaagca aagccaagacctcatatcattttacctgcagatactgggccataaagatctgtaaaacat aacccaaatatcctatcactgccagcaataatttattaacatcatctaatattcaggcca tatccaattttccctgattgtcttgaaaactatccattgcttgctgttctttttcataat ctgtctttttccttgatagaac

taccatttaagtttgcatctgaatgaaaggtgtgatccagagagtataacatcactggtt

1 5

aagcagcacatctctgatgccaaattgacagcncaaagtgaagaanaacttgtatatatt ttgcctttggaaaggacaaacaaatttccagaacttnacagggatcttgatagatgttct aaccaaggcattgaggattatggtgtttccataacaactttgaatgaggtgtttctgaaa ttagaaggaaaatcaactattgatgaatcagatattggaatttggggacaattacaaact gatggggcaaaagatataggaagccttgttgagctggaacaagttttgtcttccttccac gaaacaaggaaaacaatcagtggcgtggcgctctgg gtaataattcccagtcacacaaacccatacattctctggattttgccatggactcttgtt

1 6

ctccaatgattccagactattagaacacttatgacattttggttaataataatccttcct ataggccaaacttgggtaatgatttaaggaggtattaaaatctgggaatgccagacagtg tgctgaaacagagactcgagtcttgaacaagaaaatatcatttgtttaaccttgaataag tcttcttactagtcttgtgcttattttcctcccttataaagtaattgagtcagcaaatca tgggggtataagaattctggatctcgcaccactctccatcaaaagtatttaacacatacc ccacatatat

taccacctcaggaaaaggcccactgtgtccaggggcgtcagtccagctcaatagtgtcta

1 7

gcttttcagaatcttctgattgtttggtctttaacaaaaacatagcatctatttgctaac aataaacttaagcaacatatgataatcatttcatctccttccaagaagtggtgaaaatga accaggtgacattaatttaacttctgtttttgtttatttttaatacaaccaaaaagatat gcacatattttaatttacattatggtactgtataatttaaaccataaactggtgctgatt aatttctgcaacaaagcagcatattaaatgtactacatctagtgttgta

gtgaaggcttcactcttattgtccttctaccttgaatagaagttttcctgataagaataa

1 8

acgaggaaaaggtccttgcctcctggaagaacaaatctaccaggtgatctattcattgtt tcaactcagaatgcacttgattcaggaggtcatctgaccttcaccttggatggttagttt cactttttacatatagtttttgcagggttttattttataaaatccaagcgcgctgttgat SEQ ID

NO Target Sequence

tgtgttttccttgttttcagccccccgactccagcccgcagcacatttccgctgtccgtc agtaattgtgtcctctctttatgcttgcttggggaatgttgttttctgactaggctgatc gtatcctataactatcaacttcccaggttgaagacgatgtgttgagtttcctactgattg

19

attgattcctgtcctccccagtgtttccgtcactggttcactaaaacagtatttatatag ctccactggctctaaagctcttagtccttctaatattttggattttacaagtaaaaatgg aaaaaaaatagaaaagagncaatcaaatgcctggagcttaaaacaaagtatgtgcaacct accatctcacttgaaatttaataaaataataagtaattatgtaaatataacatagagtta tagatttatattttgttcataacacatagtgtaatataagttgtatattttcatgttttt ggttttatgttatcattcatgccacaataaaaataaaacaggagtttatgtgctcttaaa aaaaagatgtgggttgccaccaacctgtttttcgtttttgttttttgtttattttatttt atttttttgcattctcctttttcagtattactgccatg agccatcgccatggctgggatgccctcccctcacctgccctcggagcactcaagcgtgtc

20

tagcagcccagagcctgggatgcctgttatccagagcacttacggtgtgaaaggagagga gccacatatcaaagaagagatacaggccgaggacatcaatggagaaatttatgatgagta cgacgaggaagaggatgatccagatgtagattatgggagtgacagtgaaaaccatattgc aggacaagccaactgataagggtcaaaagattgttgtgaccttaggacttaaagaagccc taactggttcatccttaccagtggccaagcacattaactttctcatacactgactgttac tttaactgttagtcttaaatagttgggacatcagctgactaatagacctcagcctcaaaa ggcttggaaagaaaaaacaaatacaacaagcaaacaacaatatcaacaacaagagattga aataagctatgggtaaaataatgccagtaattcagctgctacatccaagcactgaagtct tacccgtcaacttt

agccatcgagttgatcaaagacctggtcggttacgatgtgcgccaggcgctgctcaaggg

21

cctcgtggcgctgctgataccgtcggtcaaggagatctccaaactgcaggcc gtctatggacacttgatttcttgctttatggatactctcaggaagttctggtctcaactt

22

tgagtatctcattgtcctcttactttagaatgtggggtatgggggaaaacttctatttga aaatttgagccatagcggaaattttggaatgacagggaaacaccttctcagaatcctctt acaataaaatttatatatgttatgccagacaatgtactttatactaagactggactttta aatagtctcatggattagaagtcagatacaataaataaatactttttaaaaattatattt ttaggtatatgagcattgttgttgggactacagtgctattttaaagtgaaatttagttga tttgcatgtgagtatacacctacaggtctctgttatagccgtgacatgtcttctgttact ataaattacataacaaatccttatggtttcttgggtcacttattttatttaaagacaaaa acatctttcagtacatatgtaagacttaaggcattattgaaaataagcctgtgctacatg gta

atgagcgagaatggctacgacccccagcagaacctgaacgacctgatgctgcttcagctg

23

gaccgtgaggccaacctcaccagcagcgtgacgatactgccactgcctctgcagaacgcc acggtggaagccggcaccagatgccaggtggccggctgggggagccagcgcagtgggggg cgtctctcccgttttcccaggtttgtcaacgtgactgtgacccccgaggaccagtgtcgc cccaacaacgtgtgcaccggtgtgctcacccgccgcggtggcatctgcaatggggacggg ggcacccccctcgtctgcgagggcctggcccacggcgtggcctccttttccctggggccc tgtggccgaggccctgacttcttcacccgagtggcgctcttccgagactggatcgatggt gttctcaacaacccgg

acagaccaacgcggaccgatggtccatattcttttctccatgtttgtttgtgtcgactgt

24

tgcagagtgagttgatgtaagaaggtggtttcatgttttcttttggggggccacttcttt cttttcctttccgtttctatctctgtctgtctccgtctctgtctctctctttctcacttg ttgccaatgtcttgttgctgagaaggaacactctattcagatgccatccaggaaagaaag gccagatggatggagggtccactgtgtatatttgtacagaatcatttataaagttttcta cacttctaaaaaaagtatttctaacttgggctgttggacagtttgtgtgcctctgagccg gcaagcctgttgtttcttgtgtttagtgc SEQ ID

NO Target Sequence gttctactccatacagttcacactgattgtgacacattcttagtagctagtgtctgttct agtcactgcactggagtctacgagccggaactcgctatatgcacgtgtgtgtgtccgtat gtaagaaagtgtgcaccgagtgactgaatggttgagatgaattggaatgctgaagactaa cgaagaaactagagactgatatcgagcattctgcccacctcgctctgtatttaattaatt gtgctatatgttgctttaacaacccattgagcagtcagggaatgtgagtaagcttgctgc

26 tggtttcaagtgaccctcacacgtcagcctcctaaagagntgaagttataagcatgaacc atntgtggccagcggtgggactgatccttacnnngntcttgtgtgcttcttatctctctt ggaaccacgtcatggccatgaaaacaagctcaagctagcctactgaaggatacaagaacc atagaggaaagctgagtcactttagctgaggccataatagactgatncagccccctgcca acaaccagtgatggcagatgcttgatcgggccccagaaagttcatcagaaatgnagtgag tcaaggtcacnaataactgatcagctgaaccatggactcaggaaaaattttttttaaagg tagttgtttgttttgttttttaacataccagtagctacatagcccatattgcaaatga ctctacgtcttggctctacatgaagtactcagagctggttacagcaaaaaggatggaggg aaaattatacagcagacttggaacagaacatttgaaggtatcgccgggcaggtgtccata gatgccaacggagaccgatatggggatttctctgtgattgccatgactgatgtggaggcg ggcacccaggaggttattggtgattattttggaaaagaaggtcgttttgaaatgcggccg aatgtcaaatatccttggggccctttaaaactgagaatagatgaaaaccgaattgtagag catacaaacagctctccctgcaaatcatgtggcctagaagaatcggcagtgacaggaatt gtcgtgggggctttactaggagctggcttgctaatggccttctactttttcaggaagaaa tacagaataaccattgagaggcgaacccagcaagaagaaagtaaccttggaaaacatcgg gaattacgggaagattccatcagatcccatttttcagt

tccgtctttagctgatagattctcaaatatccttgattttggatgttcagtatgtttgtg agagaggtttctgggaagactctctttttgccctcgggaaaaagcaaaatatcaatgttt gggtgactgtgtaaagctcagtgtgtaagaacatctttttgtctaggttttctttctgct ctttattgaagacaaacactcaccaaaaagaaaaataaaagttttcagagaaactaattt tctttggcaagagtattacttaatattttggcctcctaaagtttccctagttagtactcg gactcctgtgctaattgtcagcttacatatcattgtatagagactgtttattctgtacca aactgatttcaaaagtactacattgaaaataaaccggtgactgtttttcttcataaagtt ctgcgtttggcatcttcactctttccaaaatgtatctgtacatcagaaatgtcactattc caagtgtctttttagtgtggcctttagtatggcttccttttaatat

29 gaaactgcagcttctccataatttatcctgtggtgatcaaaatatctgatcttaccctgg gacacgtaaatggtcttcagttaaagaaatcctcagcaggttgcgagggaataggagact ttgtggagctgctggagggaactggattggacccttccaagatgacgcctttagctgatc tctgctacccctttcatggcccggcccagatgaaagttggctgtgacaacactgtggtgc gcatggtctccagtggaaaacacgtaaatcgtgtgacttttgagtatcgtcagctggagc cgtacgagctggaaaacccaaatggaaacagtatcggggaattctgtttgtctggtcttt gaataaccaacccagtgatttacatgctgatagctaagtgagtttttaatggccattgtg tatgattttgatgcacaactagttaaaagcctttcataccagtcagtatttcccagcc

30 tggtttaattcacccactttagatgggtgaatgttatggtgtgtgaaatatctcagtaaa gcagttaaaaggaaaaagagctggaatgcactgattcaggaacttaatttcaggaaggaa aggtctgtatgtacacatttcactttaagcagaaaatctttcttcaagaaatgactttac tttctctttgcactgccagcacgtgagatactaactttttaactagttgttcttctctag tctctacgttattagaattttttgctttcataatgtgaaacctttaagcaggagaagaaa atgttttcagatagtttcaaatacaccaaaaatgtttgaaacacaaaaatactggaatca aaccataatgcacttatt SEQ ID

NO Target Sequence

aatggagtatgcttccctcttcattcagaagatacagaaagtattggagcccagatttat

3 1

tcagtggcaatttttcttggtattaatttggccgcatttatcatcatagttttttcctat ggaagcatgttttatagtgttcatcaaagtgccataacagcaactgaaatacggaatcaa gttaaaaaagagatgatccttgccaaacgttttttctttatagtatttactgatgcatta tgctggatacccatttttgtagtgaaatttctttcactgcttcaggtagaaataccaggt accataacctcttgggtagtgatttttattctgcccattaacagtgctttgaacccaatt ctctatactctgaccacaagaccatttaaagaaatgattcatcggttttggtataactac agacaaagaaaatctatggacagcaaaggtcagaaaacatatgctccatcattcatctgg gtggaaatgtggccactgcaggagatgccacctgagttaatgaagccggaccttttcaca tac

ccagacaaatcaaatcggccctttccagcaaggagtggaacagcccaagggtggctagat

32

cacatcttctgctcacaccaccactgagcaaccctctcnaggactgccacctangcatag gtcatgccctgtgcaagagcacataaagtgggagctgaagtcctgcccaggctccattca ccaagcagagcttatgttgttgtgtgtctgtgtctacccagaggatgtgggtccttttta aaatgtgcaccaatgcatcttgtatgagctggcaccagctcggaccatcccaccctcata ccctggtcctatccagggttatattccc

aacaatttacaacagtaagcacctgtacacaagccccatctccagtcctgtttctagggc

33

ttccaatttaacacacctagaggctgggcata agcagcttattgtttctctgaaagtgtgtgtagttttactttcctaaggaattaccaaga

34

atatcctttaaaatttaaaaggatggcaagttgcatcagaaagctttattttgagatgta aaaagattcccaaacgtggttacattagccattcatgtatgtcagaagtgcagaattggg gcacttaatggtcaccttgtaacagttttgtgtaactcccagtgatgctgtacacatatt tgaagggtctttctcaaagaaatattaagcatgttttgttgctcagtgtttttgtgaatt gcttggttgtaattaaattctgagcctgatattgatatg tgatcatagctactacttcattgcaacctttattactgaccacatcagacatcatgctaa

35

atacctgaatgcatgaaaaaactccaaataagagaatctcttcaggattataaaagttgt aaaatgcaactgtattgctgagcaaaa aatgagatatctacaaggcacttaaagtgttacagatgttttaccttaagaattatttaa

36

gttgtgttgggttaagacagttttcagtgtaccgtaaatgttgtgttttcagaaaaagac aaaacgatggtgctgactggttttctgtatattgcacaacagtcctcaaatacactgatg tatgaaactattcatacatcaagcagcatttttttcactctccttagaattggaactatg cagttaaggcagataaaatgtacagatgtttcatatattacaggttacatatataaatca aaatttcctatataaaactgatttgggatttggggtggaaatattttgaatattaattta tttttaaagatgcaagataggactttgtgcaatgtatttttgtaaatgcttttcaaaata tctgtctttggtagtgcttctgctgctgccaccaaattgataagatg tgtctctctcgctcatgtaatatactctgaccctgagtggaaaggggttttngttctgtt

37

tttattttacctacatgtactatttagcttcagtgtactagtcctgccacctgtgtattt ttagggtgctatggaaataatgaaaagaaacggggatttcagaagaaaattgtaaccaaa ttcatactttgtataatttttgatatcatgatcacaggtgattcacacgtacacacataa acacacccaccagtgcagcctgaagtaactcccacagaaaccatcatcgtctttgtacat cgtatgtacaatgcaatcatttcatactttaaactggtcaaaaaactaattgtgatttct agtcttgcaaagctg gagggacagcgatgtggccctctgtgttaagaataacgtgtcctgctttggcagagagaa

38

gaaaatagccactgcccgctttcaaggcaagatcgaccttttctgttttgttttgttttt ctttctttttcctggccatgaggacaaaaattactgagtggcccttaaagagggaagttt SEQ ID

NO Target Sequence

gttttcagctgttctcttttgcccgtaggtgggagggtggggattgctgcgtcctagcta gaggaatggctttgcttgaatgtgtagtgcacacgcacgggtgtttctgtgtgctagttg cttcttgctgctgcttcctgcttgtctgggactcacatacataacgtgnnatatatatat atatatataaatgtataaatatatattttatttttttttaaatccttggagcttctggtt cctatcagttcctgttgttaatcgtaga ggtgagctctgattgcttcagttggtcatcaactattttcccttgacntgctgtcctngg

39

gatggcctgctatcttgatgatagattgtgaatatcaggaggcagggatcactgtggacc atcttagcagttgacctaacacatcttcttttcaatatctaagaacttttgccactgtga ctaatggtcctaatattaagctgttgtttatatttatcatatatctatggctacatggtt atattatgctgtggttgcgttcggttttatttacagttgcttttacaaatatttgctgta acatttgacttctaaggtttagatgccatttaagaactgagatggatagcttttaaagca tcttttacttcttaccattttttaaaagtatgcagctaaattcgaagcttttggtctata ttgttaattgccattgctgtaaatctt ttcttgcatccttctgtgattcaaaaaagtaaaatgtggctttctgaaatgatggataag

40

agtctacatcttctagaaaaaatacataaaggagtagttaagctctgtaaatgtgccacg agctccaacacgaccatcgtagggtgaagcccacgttttcttccatggcctcaaaggccc tagaacttgcctacctttctggccttacctcctagctacttatccatctcttgaacttta tactcttgtataaatttctaactttcagaaaatgccatactctgttttggcaccacacat gtatatttccccctggtacacttggaagactcttatccatctgtgaaaccctatgttgtc atcacttggtccatgaaatattacctggccaatatcccaccatcacctcaaacccaatca ccccctcctctgtatgctgtcacacctatattattaaacttatcacattgcattgtaatt acttcctgacctttgtatctactcttttagt gatttccaggaagcctttgatcacctttgtaacaagatcattgccaccaggaacccagag

41

gaaatccgagggggaggcctgcttaagtactgcaacctcttggtgaggggctttaggccn gcctctgatgaaatcaagacccttcaaaggtatatgtgttccaggtttttcatcgacttc tcagacattggagagcagcagagaaaactggagtcctatttgcagaaccactttgtggga ttggaagaccgcaagtatgagtatctcatgacccttcatggagtggtaaatgagagcaca gtgtgccntgatgggacatgaaagaagacagactttaaaccttatcaccatgctggctat ccgggtgttagctgaccaaaatgtcattcctaatgtggctaatgtcacttgctattacca gccagccccctatgtagcagatgccaactttagcaattactacattgcacaggttcagcc agtattcacgtgccagcaacagacctactccacttggctaccctgcaattaaga taggccatttggactctgccttcaaacaaaggcagttcagtccacaggcatggaagctgt

42

gaggggacaggcctgtgcgtgccatccagagtcatctcagccctgcctttctctggagca ttctgaaaacagatattctggcccagggaatccagccatgacccccacccctctgccaaa gtactcttaggtgccagtctggtaactgaactccctctggaggcaggcttgagggaggat tcctcagggttcccttgaaagctttatttatttattttgttcatttatttattggagagg cagcattgcacagtgaaagaattctggatatctcaggagccccgaaattctagctctgac tttgctgtttccagtggtatgaccttggagaagtcacttatcctcttggagcctcagttt cctcatctgcagaataatgactgacttgtctaattcatagggatgtgaggttctgctgag g gattgtctgaataggcatcctcatctatatttacccaaaacctcgcttactgtcatgtgc

43

actacaaattgcaatttggaaacctactgtattgaaattctgtcagtttatggttcttga agactgatgtcctttcccaaacactggttactgcagcagcatttttaatgtgtaagtgaa gaaaaaaggccactaaggccaaagattttttaagaatcattgtacaaatcattatgttaa actatctaagctttgctgtaatactgttttctcttcaatatgtgatggtacaggaaggat gttaaatgaaggggtggtattgcaggagagca SEQ ID

NO Target Sequence

atcctcactgatgatttcaagctaaagcaaaccaccttatacagagatctagaatctctt

44

tatgttctccagaggaaggtggaagaaaccatgggcaggagtaggaattgagtgataaac aattgggctaatgaagaaaacttctcttattgttcagttcatccagattataacttcaat gggacactttagaccattagacaattgacactggattaaacaaattcacataatgccaaa tacacaatgtatttatagcaacgtataatttgcaaagatggactttaaaagatgctgtgt aactaaactgaaataattcaattacttattatttagaatgttaaagcttatgatagtctt ttctaattcttaacactcatacttgaaatctttccgagtttccccagaagagaatatggg attttttttgacatttttgacccatttaataatgctcttgtgtttacctagtatatgtag actttgtcttatgtgtcaaaagtcctaggaaagtggttgatgtt ctgttttctgtttagtcctcaatcttcctaactcaaattggggactgaggagagagaaag

45

gtggttacccctgttaccgtgccatattcttcttgctgcttttcaaccccacgtgattgt tgattgacggttctgctataatgtgcgtgcccttcaagtttcagaaaactttcccaatca tttcacttcaatcttaattgaacccaagagtcaaagttattattttctccgaacgtgttt gtgatcttctgttatattttggggcatgttacctttatggtatataagctgtagtgcata ctctttgtattgcaaaaaactggtcagtaatttatgtacatgtattccacattttagtgt gcttgaagtgacaatccatagtttgtagtagtttgttatttgtcaactttaccctgtgtt ttaaggacatctaaacattccttgtcctatcaagatgacaaaagcagaatgtaatttttt tttggaagcttcgtgattacctgtaaca tccgtctttagctgatagattctcaaatatccttgattttggatgttcagtatgtttgtg

46

agagaggtttctgggaagactctctttttgccctcgggaaaaagcaaaatatcaatgttt gggtgactgtgtaaagctcagtgtgtaagaacatctttttgtctaggttttctttctgct ctttattgaagacaaacactcaccaaaaagaaaaataaaagttttcagagaaactaattt tctttggcaagagtattacttaatattttggcctcctaaagtttccctagttagtactcg gactcctgtgctaattgtcagcttacatatcattgtatagagactgtttattctgtacca aactgatttcaaaagtactacattgaaaataaaccggtgactgtttttcttcataaagtt ctgcgtttggcatcttcactctttccaaaatgtatctgtacatcagaaatgtcactattc caagtgtctttttagtgtggcctttagtatggcttccttttaatat gacaactgcgtgggtccaaacactcctcttcctccaggtcatttgttttgcatttttaat

47

gtctttattttttgtaatgaaaaagcacactaagctgcccctggaatcgggtgcagctga ataggcacccaaaagtccgtgactaaattccgtttgtctttttgatagcaaattatgtta agagacagtgatggctagggctcaacaattttgtattcccatgtttgtgtgagacagagt ttgttttcccttgaacttggttagaattgtgctactgtgaacgctgatcctgcatatgga agtcccactttggtgacatttcctggccattcttgtttccattgtgtggatggtgggttg tgcccacttcctggagtgagacagctcctggtgtgtagaattcccggagcgtccgtggtt cagagtaaac

atcattctctttttgctgtatagaattgcttatatcactctttctttcatgacattggtt

48

aacatttaaatgttcctcctgtacttgtgttgtctgtgaccgcttatagagttttattgt tattggtgtttacctgaatacctatgcgtacacacacacatatttccttaatacttctga actcattatcttttagaataataatactactacactttaccagcaattaacttctcccta cccaaaatgtttttcttcctgtctgaaaatggaactaatttgtcttattcgtgcttatat ctgtat

ttgtgtatgtattaccacagggcttnaatatcttaataagattttagttaatgtcatttt

49

catgaaagacagtcatttggaagccaagagctctatctatgttttgttcaaatagccgag taatctactatcttggcagtaattacttctgctgcat

Claims

1. A method for detecting leukemia from patient samples and to preferably classify it into specific types of leukemia, comprising the steps of: a. determining the expression level of at least one marker gene; and b. comparing the expression level of the at least one marker gene to a reference, wherein an up-regulated expression level of the at least one marker gene predicts leukemia, preferably a specific type of leukemia; wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI .

2. Method according to claim 1 , wherein ALL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I and/or is represented by at least one probe set or at least one target sequence listed in Table I .

3. Method according to claim 1 , wherein AM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I .

4. Method according to claim 1 , wherein CLL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I I .

5. Method according to claim 1 , wherein CM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table IV and/or is represented by at least one probe set or at least one target sequence listed in Table IV.

6. Method according to claim 1 , wherein hairy cell leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table V and /or is represented by at least one probe set or at least one target sequence listed in Table V.

7. Method according to claim 1 , wherein multiple types of leukemia are detected and the at least one marker gene corresponds to at least one of the genes listed in Table VI and/or is represented by at least one probe set or at least one target sequence listed in Table VI .

8. Method according to claim 1 , wherein the expression levels of a group of at least two marker genes are determined, wherein each marker gene of the group of at least two marker genes corresponds to one of the genes listed in Tables I to VI and/or is represented by one probe set or one target sequence listed in Tables I to VI .

9. Method according to any one of the preceding claims, wherein at least one contrasting gene corresponding to at least one of the genes listed in Table VI I and/or being represented by at least one probe set or at least one target sequence listed in Table VI I is used to increase the discriminating power of the method for detecting leukemia from patient samples.

10. Method according to any one of the preceding claims, wherein the patient sample is blood sample, preferably peripheral blood mononuclear cells ( PBMC), or bone marrow.

1 1 . Method according to any one of the preceding claims, wherein the reference is a control sample from a healthy person and the marker gene in the patient sample is up-regulated compared to the control sample.

1 2. Method according to any one of the preceding claims, wherein the reference is an internal reference gene within the patient sample, which is invariant between leukemic and non-leukemic samples or which is down- regulated in leukemic samples.

1 3. Method according to claim 8, wherein the reference is a template signature of the group of marker genes derived from expression levels of the group of marker genes from non-leukemic samples.

14. A kit for detecting leukemia from patient samples and to preferably classify it into specific types of leukemia, the kit comprising at least one reagent for determining the expression level of at least one marker gene and optionally at least one contrasting gene, wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI and wherein the at least one contrasting gene corresponds to at least one of the genes listed in Table VI I and/or is represented by at least one probe set or at least one target sequence listed in Tables VI I .

1 5. Kit according to claim 1 4, comprising at least two, preferably three, marker genes and preferably one or more contrasting gene.

16. Use of at least one marker gene for detecting leukemia from patient samples, preferably a specific type of leukemia, wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI .

1 7. Use according to claim 1 6, wherein ALL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I and/or is represented by at least one probe set or at least one target sequence listed in Table I .

18. Use according to claim 1 6, wherein AM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I .

19. Use according to claim 1 6, wherein CLL type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table I I I and/or is represented by at least one probe set or at least one target sequence listed in Table I I I .

20. Use according to claim 1 6, wherein CM L type leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table IV and/or is represented by at least one probe set or at least one target sequence listed in Table IV.

21 . Use according to claim 1 6, wherein hairy cell leukemia is detected and the at least one marker gene corresponds to at least one of the genes listed in Table V and/or is represented by at least one probe set or at least one target sequence listed in Table V.

22. Use according to claim 1 6, wherein multiple types of leukemia are detected and the at least one marker gene corresponds to at least one of the genes listed in Table VI and/or is represented by at least one probe set or at least one target sequence listed in Table VI .

23. Use according to claim 1 6, wherein a group of at least two marker genes are selected, wherein each marker gene of the group of at least two marker genes corresponds to one of the genes listed in Tables I to VI and/or is represented by one probe set or one target sequence listed in Tables I to VI

24. Use of at least one contrasting gene corresponding to at least one of the genes listed in Table VI I and/or being represented by at least one probe set or at least one target sequence listed in Table VI I to increase the discriminating power of a method for detecting leukemia from patient samples.

25. A reagent for detecting leukemia from patient samples and to preferably classify it into specific types of leukemia, the reagent being suitable for determining the expression level of at least one marker gene, wherein the at least one marker gene corresponds to at least one of the genes listed in one or more of Tables I to VI and/or is represented by at least one probe set or at least one target sequence listed in one or more of Tables I to VI .