[go: up one dir, main page]

WO2013072346A2 - États discrets destinés à être utilisés en tant que marqueurs biologiques pour des cancers, tel que le cancer à cellules rénales - Google Patents

États discrets destinés à être utilisés en tant que marqueurs biologiques pour des cancers, tel que le cancer à cellules rénales Download PDF

Info

Publication number
WO2013072346A2
WO2013072346A2 PCT/EP2012/072578 EP2012072578W WO2013072346A2 WO 2013072346 A2 WO2013072346 A2 WO 2013072346A2 EP 2012072578 W EP2012072578 W EP 2012072578W WO 2013072346 A2 WO2013072346 A2 WO 2013072346A2
Authority
WO
WIPO (PCT)
Prior art keywords
disease
genes
descriptors
invers
state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/EP2012/072578
Other languages
English (en)
Other versions
WO2013072346A9 (fr
Inventor
Manfred Beleut
Karsten Henco
Holger Moch
Peter Schraml
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zurich Universitaet Institut fuer Medizinische Virologie
PAREQ AG
Original Assignee
Zurich Universitaet Institut fuer Medizinische Virologie
PAREQ AG
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zurich Universitaet Institut fuer Medizinische Virologie, PAREQ AG filed Critical Zurich Universitaet Institut fuer Medizinische Virologie
Publication of WO2013072346A2 publication Critical patent/WO2013072346A2/fr
Publication of WO2013072346A9 publication Critical patent/WO2013072346A9/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6876—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes
    • C12Q1/6883—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material
    • C12Q1/6886—Nucleic acid products used in the analysis of nucleic acids, e.g. primers or probes for diseases caused by alterations of genetic material for cancer
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
    • G16B20/20—Allele or variant detection, e.g. single nucleotide polymorphism [SNP] detection
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
    • G16B40/30—Unsupervised data analysis
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/112—Disease subtyping, staging or classification
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q2600/00—Oligonucleotides characterized by their use
    • C12Q2600/158—Expression markers
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations

Definitions

  • markers which are frequently designated as biomarkers, are at hand being characteristic for the disease in question and relating to relevant mechanisms, relevant clinical endpoints and relevant criteria to select proper treatment.
  • markers may be found on the DNA, the R A or the protein level.
  • molecular markers as a diagnostic tool is relatively straightforward as one can use the aberration on the DNA level to predict whether the disease will develop with a certain probability or not. For example, trinucleotide expansions on the DNA level may be used to predict whether an individual will develop Huntington Chorea. Similarly, mutations in the Survival of Motor Neurons gene can be used to predict whether an individual will develop Spinal Muscular Atrophy.
  • markers of inflammation or ongoing apoptosis markers of metabolic properties or molecular markers derived from mechanistic understanding of tumor induction, induced by deregulated balances between oncogenes such as Ras, Myc, CDKs and tumor suppressor genes such as pi 6, p27 or p53 (see e.g. Hanahan & Weinberg in "The Hallmarks of Cancer” (2000).
  • tumor development mechanisms such as uncontrolled cellular growth, senescence and apoptosis evasion, such as extravasion, invasion, and evasion of immune responses have further accentuated the tumor suppressor gene hypothesis.
  • Cancer for example, is considered as a prime example for multi- factorial diseases which arise from subtle to severe deregulation of complex molecular networks. In most cases, these diseases do not develop from a single gene mutation but rather result from the accumulation from mutations in various genes. Each single mutation may not be sufficient in itself to start disease development. Rather, accumulation of mutations over time seems to increasingly deregulate the complex molecular signaling networks within cells. In these cases, disease development has therefore usually been considered to be a gradual continuous process which cannot be characterized by key events. As a consequence thereof, it is commonly assumed that such diseases cannot be diagnosed or classified by a single bio marker but by a group of markers which ideally would reflect in a simplified manner the complex molecular mechanisms underlying the disease.
  • the human genome project together with all its spin-off projects such as analysis of individual genome varieties between individuals or just individual cells affected by a disease, analyses of respective transcriptomes, proteomes etc. were assumed to directly provide a large variety of useful biomarkers. Interestingly, most of these approaches have tried again to link the phenotypic differences observed for disease with distinct molecular pathways.
  • the present invention provides a strategic and direct approach to global and functional biomarkers of clinical relevance for essentially all kinds of tumors, such as for renal cell cancer or breast cancer and potentially non-tumor diseases, too.
  • tumors such as renal cell cancer or breast cancer being associated with discrete stable or meta-stable states which can be of clinical relevance
  • one is now able to define methods allowing the skilled person to not only identify and prove the existence of such discrete states for any kind of tumor such as renal cell cancer or breast cancer, but to assign such states with descriptors and signatures associated with such states.
  • the technology allows to identify a minimum of those descriptors which unequivocally identify and discriminate each such discrete state from alternative states in a given tumor cell sample such as for renal cell cancer.
  • the invention is thus based on the surprising finding that diseases such as renal cell cancer or breast cancer can be characterized by discrete states, which reflect the underlying molecular mechanisms. Interestingly, these discrete states are distinct from one another so that disease development does not seem to be characterized by a continuous process. Rather, a discrete state seems to be maintained until a certain threshold level is reached when a switch to another discrete state occurs. Further, it seems that the discrete states may be linked to clinically and pharmacologically important parameters. However, they do not necessarily seem to coincide with standard histological classification schemes or other classification schemes.
  • a signature is a pattern reflecting the qualitative and/or quantitative appearance of at least one descriptor.
  • a signature is a pattern reflecting the qualitative and/or quantitative appearance of multiple descriptors.
  • Descriptors may in principle be any testable molecule, function, size, form or other parameter that can be linked to a cell. Descriptors may thus be e.g. genes or gene-associated molecules such as proteins and RNAs. The expression pattern of such molecules may define a signature. Such descriptors may also be designated as markers and marker sets.
  • the invention thus relates to at least one discrete disease-specific state for use as a diagnostic and/or prognostic marker in classifying samples from patients, which are suspected of being afflicted by renal cell cancer.
  • the invention further relates to at least one discrete disease-specific state for use as a diagnostic and/or prognostic marker in classifying cell lines of renal cell cancer.
  • the invention also relates to at least one discrete disease-specific state for use as a target for
  • the invention in one embodiment relates to at least one signature for use as a diagnostic and/or prognostic marker in classifying samples from patients which are suspected to be afflicted by a disease such as renal cell cancer.
  • the invention also relates to at least one signature for use as a diagnostic and/or prognostic marker in classifying cell lines of a disease such as renal cell cancer.
  • the invention further relates to at least one signature for use as a read out of a target for development, identification and/or screening of pharmaceutically active compounds.
  • the invention also relates to sets of descriptors which have been found to be predictive for a given discrete disease-specific state such as a renal cancer- or breast cancer-specific state, and to methods of identifying such sets or predictive descriptors for all states currently known for a specific disease.
  • These sets of predictive descriptors may relate to measurable properties such as determining expression by PCR, optionally by qPCR for a set of genes which is then considered to be the set of predictive descriptors. It is disclosed herein that a set of at least 6 genes for each state of renal cell cancer may be sufficient to assign a patient to one of the three known states in renal cell cancer with an accuracy of at least 65%.
  • the invention relates to methods of diagnosing a disease such as renal cell cancer or breast cancer by making use of signatures and discrete disease- specific states.
  • the invention also relates to methods of determining the responsiveness of a test population suffering from a disease such as a renal cell cancer or breast cancer towards a pharmaceutically active agent by making use of signatures and discrete disease-specific states.
  • the invention relates to methods of predicting the responsiveness of patients suffering from a disease such as renal cell cancer or breast cancer in clinical trials towards a pharmaceutically active agent by making use of signatures and discrete disease-specific states.
  • the invention also relates to methods of determining the effects of a potential pharmaceutically active compound by making use of signatures and discrete disease- specific states.
  • the invention also relates to methods for identifying signatures, discrete disease specific states and sets of predictive descriptors in samples which may be derived from patients or which may e.g. be cell lines.
  • the present invention discloses specific sets of descriptors, properties of which may be used to determine whether a specific state is present within a disease such as renal cell cancer. These properties may e.g. be the expression patterns of the descriptors which are described
  • the expression may be determined e.g. on the R A or protein level.
  • the invention as described herein is not to limited to these specific descriptor and descriptor sets. While determining the expression levels of the descriptors and descriptor sets as described hereinafter may provide a straightforward approach for classifying hyper-proliferative diseases such as renal cell cancer or breast cancer according to a new classification scheme, one can use different type of descriptors and read outs to determine states. Methods for generally detecting states in hyper-proliferative diseases such as renal cancer, colorectal cancer etc. are described in PCT/EP201 1/057691 and Beleut et al., BMC cancer (2012), 12:310.
  • All of these embodiments of the invention can be used in the context of diseases including hyper-proliferative diseases such as cancer and preferably in the context of renal cell cancer or breast cancer.
  • FIG. 12 Breast cancer-specific state-based patient stratification in combination defines responder cohort.
  • a group is defined to comprise at least a certain number of embodiments, this is also to be understood to disclose a group, which preferably consists only of these embodiments.
  • an indefinite or definite article is used when referring to a singular noun, e.g. "a”, “an” or “the”, this includes a plural of that noun unless something else is specifically stated.
  • Terms like “obtainable” or “definable” and “obtained” or “defined” are used interchangeably. This e.g. means that, unless the context clearly dictates otherwise, the term “obtained” does not mean to indicate that e.g. an embodiment must be obtained by e.g. the sequence of steps following the term “obtained” even though such a limited understanding is always included by the terms “obtained” or “defined” as a preferred embodiment.
  • the terms "about” or “approximately” denote an interval of accuracy that the person skilled in the art will understand to still ensure the technical effect of the feature in question.
  • the term typically indicates deviation from the indicated numerical value of ⁇ 10%, and preferably of ⁇ 5%.
  • sample this always preferably refers to an extracorporeal sample.
  • histological phenotypes e.g. cancers such as lung cancer with specific expression patterns assuming that the different detectable phenotypes reflect continuous and progressive disease development.
  • Another example is renal cell cancer, where histological characterization has led to identification of clear cell, papillary and other types of renal cell cancer. The present invention is not using these standard approaches of the prior art.
  • a disease is characterized by switching to discrete disease-specific states. This suggests that de-regulation of regulatory networks within a cell can occur to a certain a threshold level without the overall discrete state being affected. However, once the threshold level has been exceeded cells seem to switch to another specific discrete state. These states can therefore be considered as stable or meta- stable in that they may allow for a certain degree of variation before they may switch. We understand a discrete state to reflect the flow and extent of interactions between and within different regulatory networks.
  • the extent and flow of interactions between and within different regulatory networks may be detectable by e.g. the expression level of e.g. proteins within such regulatory networks either on the RNA or protein level.
  • the molecular entities, which are looked at can be designated as descriptors.
  • the pattern, which is detected for a set of descriptors, can be considered as a signature.
  • the signature will be the expression pattern of proteins, which function as the descriptors.
  • One may thus look at expression levels of genes on the RNA level.
  • One may look at the regulation of miRNAs and one may even look at the qualitative distribution of descriptors such as the cellular localization of certain factors or the shape of a cell.
  • Identification of disease specific descriptors such as biomarkers may then be performed using SAM (Tusher et al, Proc Natl Acad Sci USA (2001) 98(9):5116-5121).
  • SAM Session et al, Proc Natl Acad Sci USA (2001) 98(9):5116-5121).
  • This approach can be used to identify signatures and states not only for renal cell cancer, but also for breast cancer, colorectal cancer, lung cancer, etc.
  • the approach is thus generally applicable by subjecting expression data obtained from different patients to this unsupervised two-way hierarchical clustering approach.
  • Identification of signatures and steps may be best performed by first extracting descriptors such as expressed genes for certain pathway using the Panther software as described in PCT/EP201 1/057691 and Beleut et al, BMC cancer (2012), 12:310 and subjecting these pathway specific sets of genes to unsupervised two-way hierarchical clustering.
  • the groups of descriptors, e.g. the genes identified for the different pathways may then be combined and again subjected to a unsupervised two-way hierarchical clustering approach against the same tumor sets. This two-fold unsupervised two-way hierarchical clustering will reveal in a straightforward manner whether a certain disease can be classified into different disease-specific states as describe herein.
  • the an unsupervised two-way hierarchical clustering approach and preferably the two-fold application thereof as described in PCT/EP2011/057691 and Beleut et al, BMC cancer (2012), 12:310 allows identification of disease-specific states in different diseases such as renal cell cancer, breast cancer, ovarian cancer, colorectal cancer, lung cancer, prostate cancer, brain cancer, hepato cellular carcinoma, acute myeloma, pheochromocytoma, Burkitt's lymphoma, myeloma or Parkinson's disease.
  • the set of descriptors such as e.g. the set of expressed genes which can be used to distinguish the different states can be determined by this approach.
  • a set of predictive descriptors such as a set of genes can be identified which upon analysis by PCR analysis, optionally by quantitative PCR (qPCR) analysis allows assignment of a discrete disease-specific state in a patient sample.
  • qPCR quantitative PCR
  • This information may be obtainable by the unsupervised two-way hierarchical clustering approach, and preferably by the two-fold application thereof as described in PCT/EP201 1/057691 and Beleut et al., BMC cancer (2012), 12:310. From the set of descriptors which are identifiable by this approach, one can then select the set of predictive descriptors for a given disease-specific state, which are e.g. testable by PCR, optionally by qPCR, following the selection criteria mentioned hereinafter.
  • papillary RCCs of different patients may be characterized by different discrete molecular states and that the patients may thus have different survival expectations even though their cancers have been classified as comparable by histological standards. It follows from the invention as laid out hereinafter that the same discrete state can be characterized through different signatures. Thus, a novel interpretation of renal cell cancer is suggested, based on the signatures described hereinafter. The finding that a hyper-proliferative disease such as renal cell cancer can be characterized by different discrete renal cell cancer-specific states has important implications.
  • the discrete disease-specific state(s) may be used to classify patients and samples thereof as falling within distinct groups. As the discrete renal cell cancer-specific state may moreover be linked to clinically important parameters such as survival time or responsiveness to distinct drugs, this will help selecting therapeutic regimens.
  • the discrete molecular state(s) may thus be used as diagnostic and/or markers providing a new way of classifying renal cell cancer into clinically relevant subgroups etc.
  • a drug can be shown to act preferentially only in a selected group of patients which suffer from e.g. a subtype of renal cell cancer or breast cancer and which are characterized by the same discrete disease-specific state of interacting molecular networks, then this drug may be tested in other patients which suffer from a different disease, but are characterized by the same discrete molecular state. Further, clinical trials, which led to ambiguous results for a disease such as renal cell cancer or breast cancer, may be reassessed by regrouping patients according to their status as described herein.
  • the discrete states thus provide a stratifying tool for the testing of pharmacological treatments as it allows grouping of patients for clinical trials. Assuming a drug candidate is identified which is expected or hoped to positively influence the critical parameter of survival time in renal cell cancer substantially, this needs to be proven by clinical trials in order to receive FDA approval. Future drugs will likely focus on mechanistic intervention. If the mechanistically active drug is successful for the clinical end point parameter "survival time", it probably interacts selectively with mechanisms linked to the parameter "survival time”. These mechanistic subgroups are exactly those defined by e.g. the discrete molecular states enabled by this invention. It is thus fair to believe, that most probably one subgroup of patients reacts positively to a different degree than another subgroup does.
  • the knowledge about discrete disease-specific states may also allow using these states as targets during development of pharmaceutical products.
  • different renal cell cancer specific states may be linked to clinically relevant parameters such as survival time or response rate to a certain drug. If an agent is shown to switch the discrete disease- specific state in a sample or in a cell line from a state, which is linked to short survival time, into a state with long survival time, such a switch may be used as an indication that the agent may be therapeutically effective in treating the disease in question.
  • assays can be designed which make use of the correlation between a discrete renal cell cancer-specific state and e.g. the associated clinical parameter. The fact that one now knows that e.g.
  • discrete renal cell cancer- or discrete breast cancer-specific states exist and drive disease development in at least some of its aspects allows one to identify signatures of descriptors, which can then be used in a diagnostic test to classify renal cell cancer or breast cancer.
  • signatures of descriptors thus serve as a read-out for the classification of a disease or its subtype.
  • a preferred read-out for signatures and states of renal cell cancer or breast cancer may be the expression of the descriptors and descriptor sets described herein. From a practical perspective, the read out may be implemented in the form of ELISA assay, array technology, kits and all other types of devices and methods that allow determining expression of the descriptors and descriptor sets as described herein.
  • the invention thus also relates to such kits, assays, arrays etc. and as well as to the use of such kits, assays, arrays etc. as mentioned herein.
  • the read out for such states may be sets of predictive descriptors such as genes which can be tested by PCR, optionally by qPCR. The assignment of a disease-specific state based on the PCR- or qPCR-measurements is then done based on the calculations described hereinafter.
  • A, B, C and D are e.g. the expression patterns, i.e. the signatures of a limited set of descriptors, i.e. genes.
  • Each state may be best described by a signature arising from properties of a group of descriptors, which may also be designated as a descriptor set, such as the expression pattern of the group of genes described in Tables 1, 2, 3, and 4.
  • Each group of genes defines a descriptor set.
  • the expression pattern of each group of genes further provides a signature which is indicative of a renal cell cancer-specific disease state.
  • the expression pattern may be determined by different methods such as ELISA, Western Blotting, RNA expression analysis. It is to be understood that the nomenclature A, B and C refers to the same types of states as described in PCT/EP201 1/057691, even though they are described by different signatures, namely in the present case by the expression pattern of different sets of descriptors.
  • the discrete disease-specific states may reflect the aggressiveness of the tumor.
  • the read-out for these four discrete molecular states which are designated hereinafter as A, B, C and D are e.g. the expression patterns, i.e. the signatures of a limited set of descriptors, i.e. genes.
  • Each state may be best described by a signature arising from properties of a group of descriptors, which may also be designated as a descriptor set, such as the expression pattern of the group of genes described in Tables 1, 2, 3, and 4.
  • Each group of genes defines a descriptor set.
  • the expression pattern of each group of genes further provides a signature which is indicative of a breast cancer-specific disease state.
  • the expression pattern may be determined by different methods such as ELISA, Western Blotting, RNA expression analysis.
  • State means a stable or meta-stable constellation of a cell and/or cell population which is identifiable in at least two biological samples from at least two patients and which can be described by means of a single descriptor or multiple descriptors on the cellular or molecular level referenced against a standard state. As explained hereinafter, such state can be characterized by at least one or various signatures. Such signatures may be reflected by the expression of genes relative to each other.
  • different states refer to different stabile and metastabile constellations of a cell meaning that these constellations are distinct from each other in terms of the kind and extent of molecules of at least two regulatory networks interacting within a cell.
  • Different states can be characterized by a limited set of descriptors giving rise to different signatures. They may therefore also be designated a "discrete molecular state”.
  • a state is indicative of a disease, it may be designated as "disease specific molecular state" such as renal cell cancer-specific state.
  • a disease specific state may be linkable to clinically relevant parameters such as survival rate, therapy responsiveness, and the like.
  • a state which can be found in healthy human or animal subjects may be designated as "healthy state”.
  • discrete disease specific state preferably allows distinguishing different subtypes of a disease according to a new classification scheme which links the subtype being characterized by a discrete disease specific state to clinically or pharmacologically important parameters.
  • clinical or pharmacological relevant parameter preferably relate to efficacy-related parameters as they will be typically analyzed in clinical trials. They thus do not necessarily relate to a change in the histological appearance of a disease, but rather to important clinical end points such as average survival time, progression- free survival times, responsiveness to a certain drug, subjective patient- or physician- rated improvements making use established scale systems, tolerability, adverse events. The terms also include responsiveness towards treatment.
  • Descriptor means a measurable parameter on the molecular or cellular level which can be detected in terms of, but not limited to existence, constitution, quantity, localization, co-localization, chemical derivative or other physical property.
  • a descriptor thus reports at least one qualitative and/or quantitative measuring parameter of, but not limited to existence, kinetic variation, clustering, cellular localization or co-localization of at least one specific mRNA, processing or maturation derivatives of at least one specific mRNA, specific DNA-motifs, variants or chemical derivatives of such motifs, such as but not limited to methylation pattern, miRNA motifs, variants or chemical derivatives of such miRNA motifs, proteins or peptides, processing variants or chemical derivatives of such proteins or peptides or any combination of the foregoing.
  • a descriptor may be a protein the over- or under-expression of which can be used to describe a discrete disease-specific state vs. a different discrete disease-specific state or vs. the discrete healthy state. If different proteins, i.e.
  • a set of descriptors may comprise expression data for a first set of proteins, data on post-trans lational modifications of a second set of proteins and data for a group of miRNAs.
  • the measurable parameter of a descriptor is the expression level of a protein and/or gene which may be determined e.g. on the protein and/or RNA level by methods known in the art such as Western Blotting, ELISA, immunoassays, Northern Blotting, array expression analysis etc.
  • Preferred descriptors include genes and gene-related molecules such as mRNAs or proteins.
  • the “qualitative" detection of a descriptor refers preferably to e.g. determining the localization of a descriptor such as a protein, an mRNA or miRNA within e.g. a cell It may also refer to the size and/or the shape of cell.
  • the “quantitative" detection of a descriptor refers preferably to e.g. determining the presence and preferably the amount of a descriptor within a given sample.
  • the quantitative measurement of a descriptor relates to detecting the amount of genes and gene-related molecules such as mRNAs or proteins.
  • “Signature” means a pattern of a set of at least two experimentally detectable and/or quantifiable descriptors with the pattern being a characteristic description for a discrete state.
  • the term “diseases” relate to all types of diseases including hyper-pro liferative diseases. The term reflects the all stages of a disease, e.g. the formation of a disease including initial stages, the development of a disease including the spreading of a disease, the stages of manifestation, the maintenance of a disease, the surveillance of a disease etc.
  • Example of diseases include Parkinson disease, Alzheimer disease, etc..
  • hyper-proliferative diseases relate to all diseases associated with the abnormal growth or multiplication of cells.
  • a hyper-proliferative disease may be a disease that manifests as lesions in a subject.
  • Hyper-proliferative diseases include benign and malignant tumors of all types, but also diseases such as hyperkeratosis and psoriasis.
  • Tumor diseases include cancers such as such as lung cancer (including non small cell lung cancer), kidney cancer, bowel cancer, head and neck cancer, colo(rectal) cancer, glioblastom, breast cancer, prostate cancer, skin cancer, melanoma, non Hodgkin lymphoma and the like.
  • Other cancers include ovarian cancer, hepatocellular carcinoma, acute myeloid leukemia, pheochromocytoma, Burkitt's lymphoma and melanoma.
  • a preferred hyper-proliferative disease is renal cell cancer.
  • cancers considered are as defined according to the International Classification of Diseases in the field of oncology (see
  • Such cancers include epithelial carcinomas such as epithelial neoplasms; squamous cell neoplasms including squamous cell carcinoma; basal cell neoplasms including basal cell carcinoma; transitional cell papillomas and carcinomas; adenomas and adenocarcinomas (glands) including adenoma, adenocarcinoma, linitis plastic, insulinoma, glucagonoma, gastrinoma, vipoma, cholangiocarcinoma, hepatocellular carcinoma, adenoid cystic carcinoma, carcinoid tumor, prolactinoma, oncocytoma, hurthle cell adenoma, renal cell carcinoma, grawitz tumor, multiple endocrine adenomas, endometrioid adenoma; adnexal and skin appendage neoplasms; mucoepider
  • nevi and melanomas including melanocytic nevus, malignant melanoma, melanoma, nodular melanoma, dysplastic nevus, lentigo maligna melanoma, sarcoma and mesenchymal derived cancers, superficial spreading melanoma and acral lentiginous malignant melanoma.
  • sample typically refers to a human or individual that is suspected to suffer from e.g. a hyper-proliferative disease. Such individuals may be designated as patients. Samples may thus be tissue, cells, saliva, blood, serum, etc.
  • cell lines will designate cell lines which are either primary cell lines which were developed from patients' samples or which are typically be considered to be representative for a certain type of hyper-proliferative diseases. It is to be understood that all methods and uses described herein in one embodiment may be performed with at least one step and preferably all steps outside the human or animal body. If it is therefore e.g. mentioned that "a sample is obtained” this means that the sample is preferably provided in a form outside the human or animal body, i.e. as an extracorporeal sample.
  • sample, tissue etc. in the context of the present invention preferably relates to renal cell cancer tissue or breast cancer tissue. It will be first described how signatures can be identified in accordance with the invention. It is to be understood that a signature will be indicative of a discrete disease-specific state.
  • signatures and discrete disease-specific states can be identified by analyzing for the quality and/or quantity of descriptors from at least two different regulatory networks for a multitude of samples from either patients of a hyper- proliferative diseases such as renal cell cancer or cell lines of a hyper-proliferative disease such as renal cell cancer as was described in PCT/EP201 1/057691 and Beleut et al, BMC cancer (2012), 12:310 for renal cell carcinoma.
  • This data is then analyzed for certain patterns by (i) grouping the data for the quality and/or quantity across descriptors and (ii) grouping samples or cell lines in a second step for similarities of the quality and/or quantity of descriptor across all potential descriptors.
  • the present invention describes yet another algorithm based approach for identifying states, signatures and descriptors such as the expression patterns of distinct groups of genes for renal cell cancer, breast cancer or other diseases.
  • This method has led to identification of different sets of descriptors of states A, B and C known from PCT/EP2011/057691 and Beleut et al, BMC cancer (2012), 12:310 and a new state D.
  • This method may, however, also be applied to other tumors such as breast cancer. It is to be understood that the overall group of analyzed descriptors (such as the expression of all genes) does not necessarily have to yield different signatures.
  • a chosen set of descriptors may only yield one signature. This will thus indicate that the disease examined has only one discrete disease-specific state.
  • this assumes that the analysis has been performed with a comprehensive set of sample covering all relevant types of a disease such as renal cell cancer.
  • the overall group descriptors may also yield multiple signatures such as 2, 3, 4, 5 or more signatures.
  • the number of signatures will indicate the number of discrete disease-specific states that can be observed on this level of resolution for a disease. For example, if one analyzes a comprehensive set of samples for renal cell cancer or breast cancer and identifies e.g. four signatures, this means that renal cell cancer or breast cancer can be characterized by four discrete disease-specific states. For each state, one may then select one signature and thus one set of descriptors that allows to determine the respective state.
  • tables 1, 2, 3 and 4 for example describe set of descriptors (genes and proteins), the expression of which can be used to determine whether the renal cell cancer of a particular sample and thus e.g. patient is characterized by state A, B, C and D.
  • tables 5, 6, 7 and 8 for example further describe set of descriptors (genes and proteins), the expression of which can be used to determine whether the breast cancer of a particular sample and thus e.g. patient is characterized by state A, B, C and D.
  • a discrete disease-specific state may be described through multiple signatures depending on what type and combination of descriptors have been used for identifying the signatures.
  • one can identify groups by grouping samples according to the similarity of a parameter which is attributable to a descriptor (such as expression) over a complete set or over a subset of genes or gene-associated molecules, wherein the similarity is preferably measured using a statistical distance measure such as Euclidian distance, Pearson correlation, Spearman correlation, or Manhattan distance.
  • the invention wherever it mentions methods of identifying discrete disease-specific states, signatures etc. always considers that the quality and/or quantity of descriptors has to be tested. This testing may include technical means such as use of e.g. micro-arrays to determine expression of genes. If the invention considers applying such methods by relying on and using data which are indicative of the quality and/or quantity of descriptors and which are deposited in e.g. databases after they have been determined using technical means, these methods will be run on technical devices such as a computer. All methods as they are described herein for identifying discrete disease-specific states, signatures etc. may therefore be performed in a computer-implemented way.
  • signatures and discrete renal cell cancer- or breast cancer-specific states can be used for diagnostic, prognostic, analytical and therapeutic purposes. These aspects will be discussed in parallel for discrete renal cell- and breast cancer-specific states and signatures as if these terms were interchangeable. It has, however, to be born in mind that a discrete renal cell cancer- or breast cancer-specific state can be described through various signatures and depending on the type and combinations of descriptors chosen. If in the following the term signature is used this is thus meant to incorporate all signatures and descriptor types that can be used to describe a single discrete renal cell cancer-or breast cancer-specific state.
  • the invention as mentioned relates to discrete disease-specific states such as discrete renal cell cancer-specific states for use as a diagnostic and/or prognostic marker in classifying samples from patients, which are suspected of being afflicted by a disease, optionally by a hyper-proliferative disease such as renal cell cancer or breast cancer.
  • the invention also relates to discrete disease-specific states such as discrete renal cell cancer-specific states for use as a diagnostic and/or prognostic marker in classifying cell lines of a disease, optionally of a hyper-proliferative disease such as renal cell cancer or breast cancer.
  • the invention further relates to discrete disease- specific states such as discrete renal cell cancer- or breast cancer-specific states for use as a target for development of pharmaceutically active compounds.
  • the invention also relates to signatures for use as a diagnostic and/or prognostic marker in classifying samples from patients, which are suspected of being afflicted by a disease, optionally by hyper-proliferative disease such as renal cell cancer or breast cancer wherein the signature comprises a qualitative and/or quantitative pattern of at least one descriptor and wherein the signature is indicative of a discrete disease-specific state such as a discrete renal cell cancer- or discrete breast cancer- specific state.
  • the invention also relates to signatures for use as a diagnostic and/or prognostic marker in classifying cell lines of a disease, optionally of a hyper-proliferative disease such as renal cell cancer or breast cancer wherein the signature comprises a qualitative and/or quantitative pattern of at least one descriptor and wherein the signature is indicative of a discrete disease-specific state such as a discrete renal cell cancer- or discrete breast cancer-specific state.
  • the invention relates to signatures for use as a read out for a target in the development, identification and/or application of pharmaceutically active compounds, wherein the signature comprises a qualitative and/or quantitative pattern of at least one descriptor and wherein the signature is indicative of a discrete disease-specific state such as a discrete renal cell cancer- or discrete breast cancer-specific state.
  • the target may be the discrete disease specific state which is reflected by the signature.
  • the discrete disease-specific states such as discrete renal cell cancer- or discrete breast cancer-specific states and signatures relating thereto can be used for diagnostic purposes.
  • samples of patients suffering from a disease such as a hyper- proliferative disease, e.g. renal cell cancer or breast cancer may be analyzed for their discrete disease-specific states and classified accordingly.
  • a disease such as a hyper- proliferative disease, e.g. renal cell cancer or breast cancer
  • the importance of discrete disease-specific states for classifying samples and thus for diagnosing patients become clear from the experiments on RCCs as described in PCT/EP201 1/057691 and Beleut et al, BMC cancer (2012), 12:310.
  • Renal cell cancer a Renal cell cancer
  • the present invention provides further evidence that the discrete renal cell cancer-specific states A, B, C and D as reflected by the expression pattern of the descriptors of tables 1, 2, 3 and 4 (see also Experiment 2) are indeed biologically relevant. It was assumed that potential differences, possible representing functional or metabolomic irregularities among states might become evident when best state descriptors for each such state are analyzed by means of bio informatics according to functional, known and predicted protein-protein interactions. To this end STRING
  • the present invention in one aspect thus relates to a method of diagnosing, stratifying and/or screening a hyper-pro liferative disease such as renal cell cancer in at least one patient, which is suspected of being afflicted by said or in at least one cell line of said disease comprising at least the steps of:
  • step b. Allocating a discrete disease-specific state to said sample based on the signature determined in step b.).
  • the sample may be a tumor sample of renal cell cancer.
  • a signature There may be different ways to test for a signature. If the signature is not known yet, one may identify it as described above. If the signature is already known, one can test for it by analyzing the quality and/or quantity of descriptors that were used for identification of the signature. One can also use optimized signatures which allow best differentiation between different states. If for example the signature is based on expression data for a set of given genes or gene-associated molecules such as RNAs or proteins, one can test for a signature by simply determining the expression pattern for this set of molecules. This may be done by standard methods such as by micro- array expression analysis. One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 1, 2, 3 and 4.
  • the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample. If one has identified the signature, one also knows the discrete disease specific state which correlates with this signature. Using such methods one can thus classify patient samples by common molecular mechanisms that lead to the same discrete disease specific molecular states.
  • the invention preferably relates in one embodiment to identifying discrete disease specific states and preferably discrete renal cell cancer-specific states by analyzing a hyper-proliferative disease such as renal cell cancer for signatures being indicative of discrete disease specific states as described above.
  • a hyper-proliferative disease such as renal cell cancer for signatures being indicative of discrete disease specific states as described above.
  • This analysis will be performed for a specific type of hyper-proliferative disease such as e.g. renal cell cancer.
  • the diseases may be identified by common selection criteria such as the organs being affected. However, initially no attention will be given to sub- classifications of these hyper-proliferative diseases, which are based on e.g.
  • discrete disease specific states for a disease like e.g. RCC, lung cancer, breast cancer, or as in the present case renal cell cancer, etc
  • the discrete disease specific state therefore usually allows one to directly predict which sub-type of the disease in question is developing (e.g. state A, B, C or D for renal cell cancer, RCC, lung cancer (see also PCT/EP2011/057691)).
  • subtypes are correlated with e.g. clinically relevant parameters such as survival time.
  • discrete disease specific state preferably allows distinguishing different subtypes of a disease according to a new classification scheme, which links the subtype to clinically or pharmacologically important parameters.
  • discrete disease specific states exist in diseases and can be correlated with subtypes that are characterized not necessarily by their histological properties but by clinically or pharmacologically relevant parameters thus allows deciphering disease through a new code which is based on the discrete disease specific states, substates and levels.
  • the possibility of assigning a discrete disease-specific state to samples allows analyzing the effectiveness of treatments with specific drugs. For example, one can test a patient or a population of patients suffering from a hyper-proliferative disease for (i) their reaction towards treatment with a pharmaceutically active agent and (ii) for their discrete disease specific molecular state.
  • the reaction towards treatment may be measured by e.g. the quality of and quantity of clinical
  • the invention in one aspect thus relates to a method of determining the
  • responsiveness of at least one human or animal individual which is suspected of being afflicted by a hyper-proliferative disease, preferably by renal cell cancer towards a pharmaceutically active agent comprising at least the steps of:
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as renal cell cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 1, 2, 3 and 4. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample. Being able to predict the responsiveness of e.g. patients with a discrete disease specific state towards treatment is helpful in many aspects. For example, if such responsiveness is known, one can pre-select patients for treatment. Identification of signatures and discrete disease specific states can thus serve as companion diagnostics, which allow pre-selecting patients for effective treatment.
  • the invention in one embodiment thus relates to a method of predicting the responsiveness of at least one patient which is suspected of being afflicted by a hyper-proliferative disease, preferably by renal cell cancer towards a
  • pharmaceutically active agent comprising at least the steps of: a. Determining whether a correlation exists between effects on disease symptoms and/or discrete disease-specific states and the initial discrete disease-specific states as a consequence of administration of a pharmaceutically active agent as described above;
  • step d Comparing the discrete disease-specific state of the sample in step c. vs. the discrete disease-specific state for which a correlation has been determined in step a.);
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as renal cell cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 1, 2, 3 and 4. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample.
  • samples from patients can be characterized as to their discrete disease specific states. Further, cell lines of diseases may also display such discrete disease specific states. It is assumed that a pharmaceutically active agent towards which a patient with a discrete disease specific state is responsive may in some instances induce a switch to another discrete disease specific sate (see in this respect PCT/EP201 1/057691 and Beleut et al, BMC cancer (2012), 12:310). This other discrete disease specific state may either be a completely new discrete disease specific state or it may be a discrete disease specific state, which has been found in other patients.
  • a pharmaceutically active agent may induce a switch from a discrete disease specific state which is correlated with low average survival times to a discrete disease specific state which is correlated with a longer average survival time.
  • the discrete disease specific states and signatures relating thereto may be identified as described above.
  • the target on which the pharmaceutically active agent would act is thus the discrete disease specific state.
  • the discrete disease specific states are thus considered to targets of pharmaceutically active agents.
  • the invention in one embodiment therefore relates to a method of determining the effects of a pharmaceutically active compound, comprising at least the steps of: a. Providing a sample of at least one human or animal individual which is suspected of being afflicted by a hyper-pro liferative disease, preferably by renal cell cancer or a cell line of said disease before a pharmaceutically active agent is applied;
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as renal cell cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 1, 2, 3 and 4. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample.
  • the effects that are determined by this method may e.g. allow identification of compounds which may have a positive influence on the disease if e.g. a switch to a discrete disease specific state correlated with a more favorable clinical parameter such as increased survival time is observed.
  • the methods may, however, also allow identification of toxic compounds if these compounds induce a switch to a discrete disease specific state correlated with a less favorable clinical parameter such as decreased survival time.
  • These methods may thus be used as assays in the development, identification and/or screening of potential pharmaceutically active compounds, e.g. to determine the potential effectiveness of a pharmaceutically active compound in a disease such as a hyper-pro liferative disease. These assays may also be used for determining the toxicity of a pharmaceutically active compound.
  • Such discrete state-related assay systems for active and/or toxic drug candidates could be of enormous value to identify new pharmaceuticals.
  • the switch in state monitored by switch in signature marks an interesting screening system as a general "read out" for changing a tumor status. So the "read out” is related to functional efficacy rather than blocking a certain molecular target not necessarily being related to tumor function.
  • Such screening system would simply pick up any compound switching the state irrespective of the molecular target of interaction.
  • Such screening resembles assays interfering with virus propagation in cell cultures rather than screening for inhibitors of a certain viral enzyme just as reverse transcriptase.
  • the present invention in general thus relates to states, signatures and descriptors for use in diagnosing, stratifying, screening, prognosing human or animal individual being suspected of suffering from or suffering from renal cell cancer.
  • the present invention further relates to immunoassays, kits, arrays, and other type of equipment which allows determining the state of human or animal individuals being suspected of suffering from or suffering from renal cell cancer.
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as renal cell cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 1, 2, 3 and 4. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample.
  • the present invention thus also relates to a microarray comprising specifically the sets of descriptors of tables 1, 2, 3 and 4 either alone or in combination.
  • the array comprises preferably at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15 or at least 20 descriptors of table 1, 2, 3, and/or 4.
  • the present inventions also relates to an immunoassay or ELISA kit allowing for determining expression of specifically the sets of descriptors of tables 1, 2, 3 and 4 either alone or in combination.
  • the immunoassay or ELISA kit comprises preferably at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15 or at least 20 descriptors of table 1, 2, 3, and/or 4.
  • the expression patterns of genes 1 to 100 can be used to distinguish between the discrete renal cell cancer specific states A vs. BCD.
  • the expression patterns of genes 101 to 200 can be used to distinguish between the discrete renal cell cancer specific states B vs. ACD.
  • the expression patterns of genes 201 to 300 can be used to distinguish between the discrete renal cell cancer specific states C vs. ABD.
  • the expression patterns of genes 301 to 399 can be used to distinguish between the discrete renal cell cancer specific states D vs. ABC.
  • the implications of these results are set forth. Then, the computer-implemented, algorithm based approaches are explained in further detail.
  • the expression pattern of about 400 genes which are listed in table 1 , 2, 3 and 4 can be used to unambiguously identify the four discrete renal cell cancer specific states, which for sake of nomenclature have been named A, B, C and D herein.
  • genes 1 to 100 (“normal") of table 1 are found to be over- expressed for a sample of a human or animal individual, the individual will be characterized as having the discrete renal cell cancer specific state A. If genes 101 to 185 of table 2 are found to be under-expressed ("invers") and if genes 186 to 200 ("normal) of table 2 are found to be over-expressed for a sample of a human or animal individual, the individual will be characterized as having the discrete renal cell cancer specific state B. If genes 201 to 300 of table 3 are found to be over-expressed ("normal”) for a sample of a human or animal individual, the individual will be characterized as having the discrete renal cell cancer specific state C.
  • genes 301 to 399 of table 4 are found to be under-expressed ("invers") for a sample of a human or animal individual, the individual will be characterized as having the discrete renal cell cancer specific state D.
  • Expression levels may be determined using the Affymetrix gene chips HG-U133A, HG-U133B, HG-U133_Plus_2, etc.
  • the decision as to whether a certain gene in a specific sample is over- or under-expressed will be taken in comparison to a control. This control will be either implemented in the software, or an overall median or other arithmetic mean across measurements is built. By implying a multitude of samples it is also conceivable to calculate a median and/or mean for each gene respectively.
  • a respective gene expression value is monitored as up or down-regulated.
  • Affymetrix gene chip expression analysis one may rely on the "limit value" of tables 1, 2, 3 and 4 for making a decision as to over- or under- expression. The limit value will be put in the respective software, which is used for expression analysis, individually for each gene.
  • the decision as to whether a respective gene is over- or under-expressed is made with respect to a control level which will be specific for the respective detection method and which is determined typically with respect to a value typical for healthy tissue.
  • renal cell cancer signatures as they are defined by the expression patterns of the genes of tables 1, 2, 3 and 4 reflect the outcome of a statistical analysis across multiple samples.
  • the reliability of the determination increases if more than one gene is analyzed with respect to its expression.
  • the analysis of the expression of at least 10 genes will usually be sufficient to assign a discrete renal cell cancer-specific state with a reliability of at least about 90%.
  • the analysis of the expression pattern of at least 5 genes of genes 1 to 100 of table 1 will usually allow deciding whether state A is present with a reliability of about 80% or more. This reliability will increase if more genes are analyzed. Thus, the analysis of the expression pattern of at least 10 genes of genes 1 to 100 of table 1 will usually allow deciding whether state A or state BCD is present with a reliability of about 90% or more. The analysis of the expression pattern of at least 15 genes of genes 1 to 100 of table 1 will usually allow deciding whether state A or state BCD is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 1 to 100 of table 1 will usually allow deciding whether state A or state BCD is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 1 to 100 of table 1 will usually allow deciding whether state A or state BCD is present with a reliability of about 99% or more.
  • the analysis of the expression pattern of at least 5 genes of genes 101 to 200 of table 2 will usually allow deciding whether state B is present with a reliability of about 80% or more. This reliability will increase if more genes are analyzed.
  • the analysis of the expression pattern of at least 10 genes of genes 101 to 200 of table 2 will usually allow deciding whether state B or state ACD is present with a reliability of about 90% or more.
  • the analysis of the expression pattern of at least 15 genes of genes 101 to 200 of table 2 will usually allow deciding whether state B or state ACD is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 101 to 200 of table 2 will usually allow deciding whether state B or state ACD is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 101 to 200 of table
  • the analysis of the expression pattern of at least 10 genes of genes 201 to 300 of table 3 will usually allow deciding whether state C or state ABD is present with a reliability of about 90% or more.
  • the analysis of the expression pattern of at least 15 genes of genes 201 to 300 of table 3 will usually allow deciding whether state C or state ABD is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 201 to 300 of table 3 will usually allow deciding whether state C or state ABD is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 201 to 300 of table
  • the analysis of the expression pattern of at least 10 genes of genes 301 to 399 of table 4 will usually allow deciding whether state D or state ABC is present with a reliability of about 90% or more.
  • the analysis of the expression pattern of at least 15 genes of genes 301 to 399 of table 4 will usually allow deciding whether state D or state ABC is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 301 to 399 of table 4 will usually allow deciding whether state D or state ABC is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 301 to 399 of table 4 will usually allow deciding whether state D or state ABC is present with a reliability of about 99% or more.
  • the set of about 4x100 genes of tables 1, 2, 3 and 4 thus serves as a reservoir for the unambiguous characterization of states A, B, C and D.
  • analyzing the expression behavior of e.g. of at least about 2, at least about 3, at least about 4, at least about 5, at least about 6, at least about 7, at least about 8, at least about 9, at least about 10, at least about 11, at least about 12, at least about 13, at least about 14, at least about 15, at least about 16, at least about 17, at least about 18, at least about 19 or at least about 20 genes of genes 1 to 400 one will be able to decide whether a patient suffers from renal cell cancer and (ii) whether the patient suffers from cancer of state A or any of the other states B, C or D.
  • the present invention thus relates to a signature, which can be derived from the expression pattern of at least about 2, at least about 3, at least about 4, at least about 5, of at least about 6, at least about 7, at least about 8, at least about 9, at least about 10, at least about 11, at least about 12, at least about 13, at least about 14, at least about 15, at least about 16, at least about 17, at least about 18, at least about 19 or at least about 20 genes of genes 1 to 400 of tables 1, 2, 3 and 4.
  • This signature will allow to unambiguously decide whether one of four discrete renal cell cancer specific states, namely state A, B, C or D is present.
  • the signature for A is defined by an over-expression of genes 1 to 100 of table 1.
  • the signature for B is defined by an under-expression of genes 101 to 185 and an over-expression of genes 186 to 200 of table 2.
  • the signature for C is defined by an over-expression of genes 201 to 300 of table 3.
  • the signature for D is defined by an under-expression of genes 301 to 399 of table 4. It is to be understood that the survival rates which have been allocated by to states A, B and C in PCT/EP2011/057691 and Beleut et al, BMC cancer (2012), 12:310 equally apply to the states A, B and C as mentioned herein.
  • the signature described herein for state A indicates an RCC type with a high average survival time where about 70 to about 90% such as about 80%> of patients can be expected to live after 60 months.
  • the presence of this signature will be indicative of a discrete disease-specific state in RCC, which is indicative of an intermediate average survival time where about 60 to about 80% such as about 70% of patients can be expected to live after 90 months.
  • intermediate average survival time where about 45 to about 55% such as about 50%> of patients can be expected to live after 60 months.
  • the presence of this signature will be indicative of a discrete disease-specific state in RCC, which is indicative of an intermediate average survival time where about 40 to about 50% such as about 45% of patients can be expected to live after 90 months.
  • the signature described herein for state C indicates an RCC type with a low average survival time where e.g. about 30%> to about 45% such as about 40% of patients can be expected to live after 60 months.
  • the presence of this signature will be indicative of a discrete disease-specific state in RCC, which is indicative of an intermediate average survival time where about 5 to about 30% such as about 10% to 20% of patients can be expected to live after 90 months.
  • the present invention also relates to the above signatures for use as a diagnostic and/or prognostic marker in the context of renal cell cancer. By determining whether the signatures are present, one can take a decision as to whether a patient suffers from renal cell cancer as such and/or will likely develop renal cell cancer as such in the future. Further, one can distinguish between the aggressiveness of renal cell cancer development and adjust therapy accordingly. Further, the present invention relates to the above signatures for use in stratifying test populations for clinical trials for treatment of renal cell cancer. It is to be understood that determining the expression pattern of genes 1 to 400 of tables 1 , 2, 3 and 4 by microarray expression analysis as described is one of various options even though it can be preferred. However, it is also contemplated to perform such expression analysis on the protein level by e.g. ELISA, Immunoassay and/or Western Blotting. It is further to be understood that all methods of expression analysis is preferably conducted on renal cell cancer tissue.
  • the present invention relates to the above signatures for use as a read out of a target for development, identification and/or screening of at least one
  • the present invention also relates to the above signatures for use in stratifying human or animal individuals which are suspected to suffer from ongoing or imminent renal cell cancer development. Stratification allows to group these individuals by their discrete renal cell cancer specific states. Potential pharmaceutically active compounds which are assumed to be effective in renal cell cancer treatment can thus be analyzed in such pre-selected patient groups.
  • the present invention in one embodiment also relates to a method of diagnosing, prognosing, stratifying and/or screening renal cell cancer in at least one human or animal patient, which is suspected of being afflicted by said disease, comprising at least the steps of:
  • the present invention in one embodiment relates to a method of determining the responsiveness of at least one human or animal individual, which is suspected of being afflicted by renal cell cancer, towards a pharmaceutically active agent comprising at least the steps of:
  • the invention relates to a method of predicting the responsiveness of at least one patient which is suspected of being afflicted by renal cell cancer, towards a pharmaceutically active agent comprising at least the steps of:
  • b Testing a sample of a human or animal individual patient which is suspected of being afflicted by renal cell cancer for a signature indicative of a discrete renal cell cancer specific state by determining expression of at least 1 gene, preferably of at least 4 genes, more preferably of at least 5, 6, 7, 8, 9 or 10 genes of genes 1 to 100, 101 to 200, 201 to 300, 301 to 399 of tables 1, 2, 3, and/or 4; c. Allocating a discrete disease-specific state to said sample based on the signature determined in step c);
  • step d Comparing the discrete renal cell cancer-specific state of the sample in step c. vs. the discrete renal cell cancer-specific state for which a correlation has been determined in step a.);
  • One embodiment of the invention relates to a method of determining the effects of a potential pharmaceutically active agent for treatment of renal cell cancer, comprising at least the steps of:
  • the present invention provides further evidence that the discrete breast cancer- specific states A, B, C and D as reflected by the expression pattern of the descriptors of tables 5, 6, 7 and 8 (see also Experiment 4) are indeed biologically relevant. It was assumed that potential differences, possible representing functional or metabolomic irregularities among states might become evident when best state descriptors for each such state are analyzed by means of bio informatics according to functional, known and predicted protein-protein interactions. To this end STRING
  • the present invention in one aspect thus relates to a method of diagnosing, stratifying and/or screening a hyper-proliferative disease such as breast cancer in at least one patient, which is suspected of being afflicted by said or in at least one cell line of said disease comprising at least the steps of:
  • the sample may be a tumor sample of breast cancer.
  • the invention preferably relates in one embodiment to identifying discrete disease specific states and preferably discrete breast cancer-specific states by analyzing a hyper-proliferative disease such as breast cancer for signatures being indicative of discrete disease specific states as described above.
  • a hyper-proliferative disease such as breast cancer for signatures being indicative of discrete disease specific states as described above.
  • This analysis will be performed for a specific type of hyper-proliferative disease such as e.g. breast cancer.
  • the diseases may be identified by common selection criteria such as the organs being affected. However, initially no attention will be given to sub-classifications of these hyper-proliferative diseases, which are based on e.g. histological classification schemes. Once one has identified different discrete disease specific states for a disease like e.g.
  • the disease specific state therefore usually allows one to directly predict which sub-type of the disease in question is developing (e.g. state A, B, C or D for breast cancer, RCC, lung cancer (see also PCT/EP2011/057691 and Beleut et al, BMC cancer (2012), 12:310). These subtypes are correlated with e.g. clinically relevant parameters such as survival time.
  • the term discrete disease specific state preferably allows distinguishing different subtypes of a disease according to a new classification scheme, which links the subtype to clinically or pharmacologically important parameters.
  • the knowledge that discrete disease-specific states exist e.g. in breast cancer can also be used to stratify patient cohorts undergoing clinical trials for new treatments of breast cancer.
  • certain pharmaceutically active agents may act only on specific discrete disease-specific states. If a patient cohort which undergoes a clinical trial with such an active agent consists mainly of individuals with other discrete breast cancer-specific states, any effects of the pharmaceutically active agent on the specific discrete breast cancer-specific state may not be discernible. Such effects may become, however, statistically significant if the patient cohort is grouped according to the discrete breast cancer-specific states.
  • the knowledge on the existence of breast cancer-specific states can be used to stratify test populations undergoing clinical trials according to their discrete breast cancer-specific states.
  • Experiment 5 An illustration of inter alia this aspect of the invention is provided by Experiment 5 which describes that breast cancer patients having a breast cancer-specific state A as it is described hereinafter show prolonged metastasis free survival upon treatment with tamoxifen compared to patients not having this breast cancer-specific state.
  • the knowledge on breast cancer-specific states could be used to stratify patient cohorts for clinical trials involving e.g. future combination therapies including tamoxifen.
  • This knowledge may also be used for diagnostic purpose, namely to identify patients diagnosed with breast cancer which would be responsive to tamoxifen treatment.
  • the classification of samples be it of patients or cell lines for hyper-proliferative diseases such as breast cancer, for their discrete disease specific states has further implications. Given that discrete disease specific states seem to reflect decisive stages of the underlying molecular disease mechanisms, they can be linked to relevant clinical and pharmacological parameters such as average survival times or responsiveness to drugs.
  • the possibility of assigning a discrete disease-specific state to samples allows analyzing the effectiveness of treatments with specific drugs. For example, one can test a patient or a population of patients suffering from a hyper-proliferative disease for (i) their reaction towards treatment with a pharmaceutically active agent and (ii) for their discrete disease specific molecular state.
  • the reaction towards treatment may be measured by e.g. the quality of and quantity of clinical
  • responsiveness of at least one human or animal individual which is suspected of being afflicted by a hyper-proliferative disease, preferably by breast cancer towards pharmaceutically active agent comprising at least the steps of:
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as breast cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 5, 6, 7 and 8. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample.
  • An example of for this embodiment of the invention is the treatment of breast cancer patients with tamoxifen as described hereinafter which shows that breast cancer patients with breast cancer-specific state A show a prolonged distant metastasis free survival upon treatment with tamoxifen.
  • the invention in one embodiment thus relates to a method of predicting the responsiveness of at least one patient which is suspected of being afflicted by a hyper-proliferative disease, preferably by breast cancer towards a pharmaceutically active agent comprising at least the steps of:
  • step d Comparing the discrete disease-specific state of the sample in step c. vs. the discrete disease-specific state for which a correlation has been determined in step a.);
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as breast cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 5, 6, 7 and 8. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample.
  • An example of for this embodiment of the invention is the treatment of breast cancer patients with tamoxifen as described hereinafter which shows that breast cancer patients with breast cancer-specific state A show a prolonged distant metastasis free survival upon treatment with tamoxifen.
  • a pharmaceutically active agent may induce a switch from a discrete disease specific state which is correlated with low average survival times to a discrete disease specific state which is correlated with a longer average survival time.
  • the discrete disease specific states and signatures relating thereto may be identified as described above.
  • the target on which the pharmaceutically active agent would act is thus the discrete disease specific state.
  • the discrete disease specific states are thus considered to targets of pharmaceutically active agents.
  • the invention in one embodiment therefore relates to a method of determining the effects of a pharmaceutically active compound, comprising at least the steps of: a. Providing a sample of at least one human or animal individual which is suspected of being afflicted by a hyper-pro liferative disease, preferably by breast cancer or a cell line of said disease before a pharmaceutically active agent is applied;
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as breast cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 5, 6, 7, and 8. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample.
  • An example of for this embodiment of the invention is the treatment of breast cancer patients with tamoxifen as described hereinafter which shows that breast cancer patients with breast cancer-specific state A show a prolonged distant metastasis free survival upon treatment with tamoxifen.
  • the effects that are determined by this method may e.g. allow identification of compounds which may have a positive influence on the disease if e.g.
  • the methods may, however, also allow identification of toxic compounds if these compounds induce a switch to a discrete disease specific state correlated with a less favorable clinical parameter such as decreased survival time.
  • These methods may thus be used as assays in the development, identification and/or screening of potential pharmaceutically active compounds, e.g. to determine the potential effectiveness of a pharmaceutically active compound in a disease such as a hyper-pro liferative disease.
  • These assays may also be used for determining the toxicity of a pharmaceutically active compound.
  • Such discrete state-related assay systems for active and/or toxic drug candidates could be of enormous value to identify new pharmaceuticals.
  • the switch in state monitored by switch in signature marks an interesting screening system as a general "read out" for changing a tumor status. So the "read out” is related to functional efficacy rather than blocking a certain molecular target not necessarily being related to tumor function.
  • Such screening system would simply pick up any compound switching the state irrespective of the molecular target of interaction.
  • the present invention in general thus relates to states, signatures and descriptors for use in diagnosing, stratifying, screening, prognosing human or animal individual being suspected of suffering from or suffering from breast cancer.
  • the present invention further relates to immunoassays, kits, arrays, and other type of equipment which allows determining the state of human or animal individuals being suspected of suffering from or suffering from breast cancer.
  • the signature may be tested for as described above.
  • the sample may be a tumor sample such as breast cancer.
  • One way of determining a signature is to test for the expression pattern of the descriptor sets of tables 5, 6, 7 and 8. If the descriptor sets show an expression profile as described below, one can allocate a signature and thus state A, B, C or D to the respective sample.
  • the present invention thus also relates to a microarray comprising specifically the sets of descriptors of tables 5, 6, 7, and 8 either alone or in combination.
  • the array comprises preferably at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15 or at least 20 descriptors of table 1, 2, 3, and/or 4.
  • the present inventions also relates to an immunoassay or ELISA kit allowing for determining expression of specifically the sets of descriptors of tables 5, 6, 7, and 8 either alone or in combination.
  • the immunoassay or ELISA kit comprises preferably at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15 or at least 20 descriptors of table 5, 6, 7, and/or 4.
  • the expression patterns of genes 1 to 100 can be used to distinguish between the discrete breast cancer specific states A vs. BCD.
  • the expression patterns of genes 101 to 200 can be used to distinguish between the discrete breast cancer specific states B vs. ACD.
  • the expression patterns of genes 201 to 300 can be used to distinguish between the discrete breast cancer specific states C vs. ABD.
  • the expression patterns of genes 301 to 399 can be used to distinguish between the discrete breast cancer specific states D vs. ABC.
  • the expression pattern of about 400 genes which are listed in table 5, 6, 7, and 8 can be used to unambiguously identify the four discrete breast cancer specific states, which for sake of nomenclature have been named A, B, C and D herein.
  • genes 1 to 100 of Table 5 are found to be over-expressed for a sample of a human or animal individual, the individual will be characterized as having the discrete breast cancer specific state A. If genes 101 to 200 of table 6 are found to be under-expressed ("invers") for a sample of a human or animal individual, the individual will be characterized as having the discrete breast cancer specific state B.
  • genes 201 to 292 of table 7 are found to be under-expressed ("invers") and if genes 293 to 300 of table 7 are found to be over-expressed (“normal") for a sample of a human or animal individual, the individual will be characterized as having the discrete breast cancer specific state C. If genes 301 to 399 of table 8 are found to be under-expressed ("invers") for a sample of a human or animal individual, the individual will be characterized as having the discrete breast cancer specific state D. Expression levels may be determined using the Affymetrix gene chips HG-U133A, HG-U133B, HG-U133_Plus_2, etc.
  • the decision as to whether a respective gene is over- or under-expressed is made with respect to a control level which will be specific for the respective detection method and which is determined typically with respect to a value typical for healthy tissue.
  • breast cancer signatures as they are defined by the expression patterns of the genes of tables 5, 6, 7, and 8 reflect the outcome of a statistical analysis across multiple samples.
  • the reliability of the determination increases if more than one gene is analyzed with respect to its expression.
  • the analysis of the expression of at least 10 genes will usually be sufficient to assign a discrete breast cancer-specific state with a reliability of at least about 90%..
  • the analysis of the expression pattern of at least 5 genes of genes 1 to 100 of table 5 will usually allow deciding whether state A is present with a reliability of about 80% or more. This reliability will increase if more genes are analyzed. Thus, the analysis of the expression pattern of at least 10 genes of genes 1 to 100 of table 5 will usually allow deciding whether state A or state BCD is present with a reliability of about 90% or more. The analysis of the expression pattern of at least 15 genes of genes 1 to 100 of table 5 will usually allow deciding whether state A or state BCD is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 1 to 100 of table 5 will usually allow deciding whether state A or state BCD is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 1 to 100 of table 5 will usually allow deciding whether state A or state BCD is present with a reliability of about 99% or more.
  • the analysis of the expression pattern of at least 5 genes of genes 101 to 200 of table 6 will usually allow deciding whether state B is present with a reliability of about 80% or more. This reliability will increase if more genes are analyzed.
  • the analysis of the expression pattern of at least 10 genes of genes 101 to 200 of table 6 will usually allow deciding whether state B or state ACD is present with a reliability of about 90% or more.
  • the analysis of the expression pattern of at least 15 genes of genes 101 to 200 of table 6 will usually allow deciding whether state B or state ACD is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 101 to 200 of table 6 will usually allow deciding whether state B or state ACD is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 101 to 200 of table
  • the analysis of the expression pattern of at least 10 genes of genes 201 to 300 of table 7 will usually allow deciding whether state C or state ABD is present with a reliability of about 90% or more.
  • the analysis of the expression pattern of at least 15 genes of genes 201 to 300 of table 7 will usually allow deciding whether state C or state ABD is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 201 to 300 of table 7 will usually allow deciding whether state C or state ABD is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 201 to 300 of table
  • the analysis of the expression pattern of at least 10 genes of genes 301 to 399 of table 8 will usually allow deciding whether state D or state ABC is present with a reliability of about 90% or more.
  • the analysis of the expression pattern of at least 15 genes of genes 301 to 399 of table 8 will usually allow deciding whether state D or state ABC is present with a reliability of about 95% or more.
  • the analysis of the expression pattern of at least 20 genes of genes 301 to 399 of table 8 will usually allow deciding whether state D or state ABC is present with a reliability of about 98% or more and the analysis of the expression pattern of at least 25 genes of genes 301 to 399 of table 8 will usually allow deciding whether state D or state ABC is present with a reliability of about 99% or more.
  • the set of about 4x100 genes of tables 5, 6, 7 and 8 thus serves as a reservoir for the unambiguous characterization of states A, B, C and D.
  • the present invention thus relates to a signature, which can be derived from the expression pattern of at least about 2, at least about 3, at least about 4, at least about 5, of at least about 6, at least about 7, at least about 8, at least about 9, at least about 10, at least about 11, at least about 12, at least about 13, at least about 14, at least about 15, at least about 16, at least about 17, at least about 18, at least about 19 or at least about 20 genes of genes 1 to 400 of tables 5, 6, 7 and 8.
  • This signature will allow to unambiguously decide whether one of four discrete breast cancer specific states, namely state A, B, C or D is present.
  • the signature for A is defined by an over-expression of genes 1 to 100 of table 5.
  • the signature for B is defined by an under-expression of genes 101 to 200 of table 6.
  • the signature for C is defined by an under-expression of genes 201 to 292 and an over-expression of genes 293 to 300 of Table 7.
  • the signature for D is defined by an under-expression of genes 301 to 399 of Table 8.
  • the present invention also relates to the above signatures for use as a diagnostic and/or prognostic marker in the context of breast cancer. By determining whether the signatures are present, one can take a decision as to whether a patient suffers from breast cancer as such and/or will likely develop breast cancer as such in the future. Further, one can distinguish between the aggressiveness of breast cancer
  • the present invention relates to the above signatures for use in stratifying test populations for clinical trials for treatment of breast cancer.
  • determining the expression pattern of genes 1 to 400 of tables 5, 6, 7 and 8 by microarray expression analysis as described is one of various options even though it can be preferred. However, it is also contemplated to perform such expression analysis on the protein level by e.g. ELISA, Immunoassay and/or Western Blotting. It is further to be understood that all methods of expression analysis is preferably conducted on breast cancer tissue. Further, the present invention relates to the above signatures for use as a read out of a target for development, identification and/or screening of at least one
  • the present invention also relates to the above signatures for use in stratifying human or animal individuals which are suspected to suffer from ongoing or imminent breast cancer development. Stratification allows to group these individuals by their discrete breast cancer specific states. Potential pharmaceutically active compounds which are assumed to be effective in breast cancer treatment can thus be analyzed in such pre- selected patient groups.
  • the present invention in one embodiment also relates to a method of diagnosing, prognosing, stratifying and/or screening breast cancer in at least one human or animal patient, which is suspected of being afflicted by said disease, comprising at least the steps of:
  • the present invention in one embodiment relates to a method of determining the responsiveness of at least one human or animal individual, which is suspected of being afflicted by breast cancer, towards a pharmaceutically active agent comprising at least the steps of:
  • the invention relates to a method of predicting the responsiveness of at least one patient which is suspected of being afflicted by breast cancer, towards a pharmaceutically active agent comprising at least the steps of:
  • step c Allocating a discrete disease-specific state to said sample based on the signature determined in step c);
  • step d Comparing the discrete breast cancer-specific state of the sample in step c. vs. the discrete breast cancer-specific state for which a correlation has been determined in step a.);
  • An example of for this embodiment of the invention is predicting the responsiveness of breast cancer patients depending on their breast-cancer specific states A, B, C, and D toward treatment with tamoxifen.
  • the examples as presented hereinafter show that a breast cancer patient with the breast cancer-specific state A, but not on a state other than A, will react positively towards treatment with tamoxifen as can be taken from the prolonged distant metastasis free survival time.
  • One embodiment of the invention relates to a method of determining the effects of a potential pharmaceutically active agent for treatment of breast cancer, comprising at least the steps of:
  • testing said sample for a signature indicative of a discrete breast cancer specific state by determining expression of at least 1 , preferably of at least 4 genes, more preferably of at least 5, 6,7, 8, 9 or 10 genes of genes 1 to 100, 101 to 200, 201 to 300, 301 to 399 of tables 5, 6, 7, and/or 8;
  • the present invention also relates to a computer implemented method, a computer or other technical device which is suitable to perform the above steps and methods or those described in Experiment 1 and Figure 3.
  • the latter computer-implemented methods will allow identifying states, signatures and descriptors for a disease.
  • the invention further relates to the use of such computer-implemented methods, computers, technical devices etc. for classifying renal cell cancer samples, tissues etc. Such classification may enable the above-mentioned uses of diagnosis, stratification etc.
  • Examples of such other tests include nCounter Gene Expression assays from Nanostring (Seattle, WA, U.S. A), alternative expression analysis by sequencing (ALEXA-seq, www.alexaplatform.org), Serial Analysis of Gene Expression (SAGE), Northern Blotting, and more.
  • Example 6 it is described how one can obtain a set of descriptors, which are predictive for a renal-cancer specific state and which are measurable e.g. by PCR, and optionally by qPCR, by selecting such predictive descriptors from the group of descriptors such as expressed genes that are obtainable by unsupervised two-way hierarchical clustering for microarray expression data (e.g., using Affymetrix HG- U133 A, G-U133 B, HG-U133 Plus 2.0, Agilent, Nimblegen and their derivatives such as Illumina) as described in PCT/EP201 1/057691 and Beleut et al., BMC cancer (2012), 12:310 as well as herein.
  • a set of descriptors which are predictive for a renal-cancer specific state and which are measurable e.g. by PCR, and optionally by qPCR
  • this approach can be used for determining set of predictive descriptors for other hyper proliferative diseases such as breast cancer, ovarian cancer, colorectal cancer, lung cancer, prostate cancer, brain cancer, hepato cellular carcinoma, acute myeloma, pheochromocytoma, Burkitt's lymphoma, myeloma or other types of diseases such as Parkinson's disease once it has been shown by e.g. unsupervised two-way hierarchical clustering for microarray expression data as described in PCT/EP201 1/057691 and Beleut et al., BMC cancer (2012), 12:310 as well as herein that disease-specific states exist for these types of diseases.
  • the approach starts from the group of descriptors, such as expressed genes that are obtainable by unsupervised two-way hierarchical clustering for microarray expression data as described in PCT/EP2011/057691 and Beleut et al, BMC cancer (2012), 12:310 as well as herein.
  • probe array consists of a number of oligonucleotide probe cells and each probe cell contains a unique oligonucleotide probe.
  • Probes are tiled in probe pairs as a Perfect Match (PM) and a Mismatch (MM).
  • PM andMM are the same, except for a change to the Watson-Crick complement in the middle of the MM probe sequence.
  • a probe set consists of a series of probe pairs and represents an expressed transcript. The sample should bind stronger to PM than to MM, so one assumes that for a given probe set for example 25% or more of the PM values should be higher than the MM values to be deemed valid.
  • a next step one selects at least the 2 genes, preferably at least 10 out of the remaining genes with the most positive, and the at least 2 genes, preferably at least 10 out of the remaining genes with the most negative correlation for a disease- specific state such as renal cancer-specific states A, B, or C. Further it is possible to add at least 2, preferably at least 10 genes that are randomly selected and at least 2, preferably at least 10 genes showing the least variation across all the states.
  • the resulting set of genes can then be tested e.g. in qPCR experiments with genes grouped in pairs since qPCR readers are usually designed to measure just pairs of genes.
  • correlations between the measured qPCR expression values and the states were calculated in a first step.
  • Genes for further analysis thus are chosen according to their correlation between mRNA expression microarray and qPCR tests. All genes with a qPCR/microarray correlation inferior to a threshold value between 0 and 1, preferably 0.35 may be excluded from the model leaving a reduced set of genes.
  • This set of genes should comprise typically at least 20 genes to then allow identification of sets of predictive descriptors for each disease-specific state.
  • 22 genes had values above the selected threshold of 0.35, which are A2M, ANGPTL4, AP2M1, BDH1, CD99, COBLL1, DOCK9, EPAS1, F5, H3F3B, IFITM3, LAPTM3B, LDB2, LPCAT3, MAPRE1, NDUFA4, PGBD5, RGS5, SERBP1, SERINC3, TSG101, UFSP2.
  • X E PASI denotes the qPCR value measured for patient (j) and gene EPAS1.
  • means and standard deviations for all patients which are not allocated to the particular state are calculated, that is, nonA, nonB, nonC are calculated.
  • nonA the mean over all patients not being allocated to state A (that is, allocated to either B or C, or not allocated at all) and gene EPAS1 is calculated according to
  • X E PASI denotes the qPCR value measured for patient (j) and gene EPAS1.
  • means and standard deviations for all patients which are not allocated to the particular state are calculated, that is, nonA, nonB, nonC are calculated.
  • nonB the mean over all patients not being allocated to state B (that is, allocated to either A or C, or not allocated at all) and gene EPAS1 is calculated according to n (all petients
  • X E PASI denotes the qPCR value measured for patient (j) and gene EPAS1.
  • means and standard deviations for all patients which are not allocated to the particular state are calculated, that is, nonA, nonB, nonC are calculated.
  • nonB the mean over all patients not being allocated to state B (that is, allocated to either A or B, or not allocated at all) and gene EPAS1 is calculated according to
  • MAPRE1 25 0.8 24.9 0.7 24.7 0.6 25.5 0.8 25.2 0.8 24.5 0.5
  • the posterior probability that the measured qPCR value x for EPAS 1 denotes state A is calculated according to
  • the values are calcul for state A, according to
  • the state of a sample is allocated by determining the maximum value of the individual subset probabilities. For example to determine whether a sample is state A or not, the state SA is calculated by
  • the values are calculated, for example for state A, according to
  • the state of a sample is allocated by determining the maximum value of the individual subset probabilities. For example to determine whether a sample is state B or not, the state SB is calculated by
  • the values are calculat for example for state A, according to
  • the state of a sample is allocated by determining the maximum value of the individual subset probabilities. For example to determine whether a sample is state C or not, the state SC is calculated by
  • 3 ⁇ 4BC max ⁇ PA,subset ⁇ > ⁇ , subset B,Pc, subset c)
  • the subsets of genes for the individual states are selected from the qPCR set of genes according to the following iterative procedure: 1. the subset of predictive genes is empty in the beginning of the procedure
  • step 3 select a number of genes with the best accuracy of step 2 and add them to the set of predictive genes, and take them out of the set.
  • the number of genes selected is at least one, two, or three.
  • step 4 select a number of genes with the best accuracy of step 4 and add them to the set of predictive genes, and take them out of the set. Preferably the number of genes selected is at least one, two, or three. 5. repeat steps 4) and 5) until a desired accuracy is obtained.
  • the prediction accuracy will increase with the number of genes tested. However, cost, effort, and time spent for a test will increase as well with the number of tested genes and will eventually exceed practical limits. It is possible to set a desired or required predication accuracy such as e.g. at least about 60%, at least about 65%o, at least about 70%>, at least about 75%, at least about 80%>, at least about 85%), or at least about 90%>, at least about 91%>, at least about 92%, at least about 93%), at least about 94%>, at least about 95%, at least about 96%>, at least about 975, at least about 98%>, or at least about 99% and repeat the selection procedure until the subset of genes selected provides this accuracy.
  • Table 12 shows the accuracy obtained with an increasing subset of genes, for predicting a single state (a) and to predict the state of a sample directly from a single set of measurements (b)
  • a set of at least 2 descriptors e.g. genes will typically allow to predict a disease-specific state with an accuracy of at least 65%.
  • the accuracy can be improved if more descriptors, e.g. genes are included into the SP list for each state.
  • a set of at least 4 descriptors may typically allow to predict a disease-specific state with an accuracy of at least 75%.
  • a set of at least 10 descriptors, e.g. genes may typically allow to predict a disease-specific state with an accuracy of at least 80%.
  • a set of at least 15 descriptors, e.g. genes will typically allow to predict a disease-specific state with an accuracy of at least, 90% or at least 95%.
  • Sn for the state n is calculated according to:
  • the invention in one aspect relates to a method of diagnosing, prognosing, classifying, stratifying and/or screening a disease in a human or animal patient, which is suspected of being afflicted by said disease, comprising at least the steps of:
  • determining the expression of a set of at least two predictive descriptors e.g. genes in a sample of said patient for a disease-specific state
  • said set of predictive descriptors is selected from a group of descriptors which is indicative of disease-specific states and which is identifiable by unsupervised two-way hierarchical clustering of gene expression data for samples of said disease from different patients.
  • the control sample may be a sample, preferably an extracorporeal sample from a healthy subject.
  • the group of descriptors, from which the sets of predictive descriptors are selected and which is indicative of disease-specific states, is identifiable by unsupervised two- way hierarchical clustering of gene expression data for samples of said disease from different patients as described in PCT/EP201 1/057691 and Beleut et al, BMC cancer (2012), 12:310.
  • the approach may thus rely on identifying disease-specific states, e.g. in renal cell cancer or breast cancer by performing an unsupervised two-way hierarchical clustering approach with TIGR MeV (Saeed et al., Methods Enzymol.
  • the sets of at least 2 predictive descriptors which is preferably analyzed by qPCR analysis, is selected from a group of descriptors, which is indicative of disease- specific states and which is identifiable by unsupervised two way hierarchical clustering of gene expression data for samples of said disease from different patients, by a process comprising at least the steps of:
  • a second list or subset of predictive descripotrs e.g. genes which is empty in the beginning of the following procedure, g. computing the accuracy of predicting the desired state disease-specific state for all single descriptors, e.g. genes separetly, of the set by employing the above naive Bayes model,
  • step g select a number of descriptors, e.g. genes with the best accuracy of step g) and add them to the second list or subset of predictive descriptors, e.g. genes, and take them out of the first list,
  • step h select a number of descriptors, e.g. genes with the best accuracy of step h) and add them to the second list or subset of predictive descriptors, e.g. genes, and take them out of the first list
  • steps g) and j) repeat steps g) and j) until a desired accuracy is obtained until said second list contains at least 2 descriptors for each disease-specific state, or until the prediction accuracy reaches a predefined threshold.
  • the number of descriptors, e.g. genes selected in steps g) and j) is at least one, two, or three.
  • descriptors of steps a. to d. are combined in a first list
  • the number of descriptors, e.g. genes to be combined in said first list should be at least 40, preferably at least 50, at least 60, at least 70, at least 80, at least 90, or at least 100.
  • a set of predictive descriptors for a given disease-specific state which is measurable by e.g. PCR and optionally qPCR, should comprise at least 2, at least 4, at least 8, at least 10, at least 12, at least 14, at least 16, or at least 20 descriptors, e.g. genes.
  • the present invention further relates to the above-mentioned method of diagnosing, prognosing, classifying, stratifying and/or screening a disease in a human or animal patient, which is suspected of being afflicted by said disease, by determining a measurable quantity, e.g. expression of a set of disease-specific state-predictive descriptors, e.g. genes, wherein expression of said set of at least 2 descriptors per disease-specific state is determined by qPCR analysis and wherein, based on the qPCR results, assignment of a disease-state for a sample of patient is calculated according to:
  • hyper proliferative diseases include renal cell cancer, breast cancer, ovarian cancer, colorectal cancer, lung cancer, prostate cancer, brain cancer, hepato cellular carcinoma, acute myeloma, pheochromocytoma, Burkitt's lymphoma or myeloma.
  • Particularly preferred hyper proliferative diseases are renal cell cancer and breast cancer.
  • a particularly preferred embodiment relates to the use of sets of predictive descriptors for diagnosing, prognosing, classifying, stratifying and/or screening renal cell cancer in a human or animal patient, which is suspected of being afflicted by said disease, the descriptor being selected from the genes of Table 10.
  • the descriptor being selected from the genes of Table 10.
  • EPAS1, LAPTM4B, DOCK9, BDHl, AP2M1, LPCAT3, State A of renal cell cancer can be predicted with an accuracy of 85%.
  • DOCK9, CD99, BDHl, PGBDB5, NDUF4A, LPCAT3, State B of renal cell cancer can be predicted with an accuracy of 93%.
  • LPCAT3, LAPTM4B, RGS5, SERINC3, F5, COBLL1, State C of renal cell cancer can be predicted with an accuracy of 76% (see also Table 11).
  • the present invention thus further relates to a combination of predictive descriptors for diagnosing, prognosing, classifying, stratifying and/or screening renal cell cancer in a human or animal patient, which is suspected of being afflicted by said disease, said descriptors being selected from Table 10.
  • the present invention relates to a method of identifying sets of predictive descriptors, e.g. genes for a disease-specific state in a sample of a patient suffering from said disease which are suitable diagnosing, prognosing, classifying, stratifying and/or screening a disease in a human or animal patient, comprising at least the steps of:
  • descriptors e.g. genes genes with the best accuracy of step h
  • second list or subset of predictive descriptors e.g. genes
  • take them out of the first list compute the accuracy of predicting the desired disease-specific state for the combination of all descriptors, e.g. genes in the second list or subset in combination with each single remaining descriptor, e.g. gene of the first list
  • step j select a number of descriptors, e.g. genes with the best accuracy of step j) and add them to the second list or subset of predictive descriptors, e.g. genes, and take them out of the first list, repeat steps i) and k) until a desired accuracy is obtained until said second list contains at least 2 descriptors, e.g. genes for each disease- specific state, or until the prediction accuracy reaches a predefined threshold.
  • a number of descriptors e.g. genes with the best accuracy of step j
  • the second list or subset of predictive descriptors e.g. genes
  • the number of descriptors, e.g. genes selected in steps g) and j) is at least one, two, or three.
  • the sets of predictive descriptors in a sample of a patient may be analyzed by qPCR analysis.
  • the sets of predictive descriptors e.g. genes may be identified for a hyper proliferative disease.
  • a hyper proliferative disease is selected from the group comprising include renal cell cancer, breast cancer, ovarian cancer, colorectal cancer, lung cancer, prostate cancer, brain cancer, hepato cellular carcinoma, acute myeloma, pheochromocytoma, Burkitt's lymphoma or myeloma. Even more preferavly.
  • said hyper proliferative disease is renal cell cancer or breast cancer.
  • a set of predictive descriptors for a disease-specific state id identified comprising at least two, four, six, eight, 10, 12, 14, 16, 18, or 20 predictive descriptors for a discrete disease-specific state
  • the sets of descriptors may be identified in an extracorporeal sample of a patient.
  • the present invention also relates to combinations of predictive descriptors for diagnosing, prognosing, classifying, stratifying and/or screening renal cell cancer in a human or animal patient, which is suspected of being afflicted by said disease, being identifiable by the above method.
  • a two-step algorithm is applied: 1. In a first step the samples for the other tumor type such as colorectal cancer are classified according to the states found in the RCC, leading to a tissue-specific classification of the samples "Ts-C".
  • tissue-specific gene set Ts-G
  • each RCC sample can be characterized by a list of 100 values for the gene expression levels. This ordered list can be regarded as a vector in a 100-dimensional Hilbert space H. Once could select more or less than 100 values. However, this number is a reasonable size that provides reliable results.
  • the set of genes, which build the base of this space are optimized such that an optimal clustering (with a maximal distance within this vector space given the metric in this space) of the states is given by this base.
  • Figure 1 shows this process of representation of the states in a higher- dimensional space, together with the identification of centers for the states (denoted by crosses in the sketch).
  • a further vector is defined by the centre of all samples, which could not have been covered by the previous mapping. This forth group is denoted by "D”.
  • the set of genes RCC-G can be taken to define a similar vector space for the tumor specific samples of another disease or cancer such as colorectal cancer.
  • the data for these other tumor specific samples such as colorectal cancer may come e.g. from microarray expression analysis.
  • the cancer related data may be publicly available expression data, for example from Affymterix gene chip data, preferably for whole genome expression data. Not only this base is defined, moreover with the already calculated centers of the states one can now generate a tissue specific classification (Ts-C) by
  • tissue-specific classification is defined, illustrated in figure 2.
  • Ts-C tumor-specific classification
  • Each of the genes within this set is able to discriminate between one specific state and the other states, e.g. between "A” on the one hand, and "B", “C” and “D” on the other hand.
  • the tumor specific gene set relates to colorectal cancer.
  • the flowchart of the algorithm is sketched in Figure 3.
  • the samples for the other tumor type such as breast cancer are classified according to the states found in the RCC, leading to a tissue-specific classification of the samples "Ts-C".
  • tissue-specific gene set Ts-G
  • each RCC sample can be characterized by a list of 100 values for the gene expression levels. This ordered list can be regarded as a vector in a 100-dimensional Hilbert space H. Once could select more or less than 100 values. However, this number is a reasonable size that provides reliable results.
  • the set of genes, which build the base of this space are optimized such that an optimal clustering (with a maximal distance within this vector space given the metric in this space) of the states is given by this base. This defines the optimal set of genes RCC-G.
  • Figure 1 shows this process of representation of the states in a higher- dimensional space, together with the identification of centers for the states (denoted by crosses in the sketch).
  • a further vector is defined by the centre of all samples, which could not have been covered by the previous mapping. This forth group is denoted by "D”.
  • the set of genes RCC-G can be taken to define a similar vector space for the tumor specific samples of another disease or cancer such as breast cancer.
  • the data for these other tumor specific samples such as breast cancer may come e.g. from microarray expression analysis.
  • the breast cancer related data were publicly available expression data, for example from Affymterix gene chip data, preferably for whole genome expression data.
  • Ts-C tissue specific classification
  • Ts-C tumor-specific classification
  • Each of the genes within this set is able to discriminate between one specific state and the other states, e.g. between "A” on the one hand, and "B", “C” and “D” on the other hand.
  • Ts-G tumor specific gene set
  • the tumor specific gene set relates to breast cancer.
  • the flowchart of the algorithm is sketched in Figure 3.
  • Tamoxifen is approved by the U.S. Food and Drug Administration to treat women diagnosed with estrogen-receptor(ER)-positive, early and late stage breast cancer after primary intervention (chemotherapy, radiation, surgery) to reduce the risk of recurrence of the cancer.
  • ER-positive breast cancers show a heterogeneous range of response rates suggesting a complex biology of these tumours.
  • DMFS disant metastasis free survival
  • breast cancer-specific state A is associated with a markedly prolonged time to disease progression.
  • Patients with other states or belonging to group E (“non-State A") show a clinical course that is indistinguishable from patients that did not receive medication as demonstrated with cohort II which had not been treated with Tamoxifen (Fig. 2B).
  • Data for the above patients can be retrieved by typing the patient identifiers into the GEO accession no. field at http://www.ncbi.nlm.nih.gov/geo/.
  • Affymetrix a probe array consists of a number of oligonucleotide probe cells and each probe cell contains a unique oligonucleotide probe. Probes are tiled in probe pairs as a Perfect Match (PM) and a Mismatch (MM). The sequence for PM andMM are the same, except for a change to the Watson-Crick complement in the middle of the MM probe sequence.
  • a probe set consists of a series of probe pairs and represents an expressed transcript.
  • the sample should bind stronger to PM than to MM, so one assumes that for a given probe set for example 25% or more of the PM values should be higher than the MM values to be deemed valid.
  • the expression values of the remaining genes were then correlated with the patient states A, B, C.
  • the list of genes describing states A, B, and C is shown in Table 9.
  • a next step one selected at least the 2 genes, preferably at least 10 out of the remaining genes with the most positive, and the at least 2 genes, preferably at least 10 with the most negative correlation for a disease-specific state such as renal cancer-specific states A, B, or C. Further it is possible to add at least 2, preferably at least 10 genes that are randomly selected and at least 2, preferably at least 10 genes showing the least variation across all the states.
  • the values are cal for state A, according to
  • the state of a sample is allocated by determining the maximum value of the individual subset probabilities. For example to determine whether a sample is state A or not, the state SA is calculated by
  • 3 ⁇ 4BC max ⁇ PA,subset ⁇ > ⁇ , subset B,Pc, subset c)
  • step 3 select a number of genes with the best accuracy of step 2 and add them to the set of predictive genes, and take them out of the set.
  • the number of genes selected is at least one, two, or three.
  • step 4 select a number of genes with the best accuracy of step 4 and add them to the set of predictive genes, and take them out of the set.
  • the number of genes selected is at least one, two, or three.
  • Table 12 shows the accuracy obtained with an increasing subset of genes, for predicting a single state (a) and to predict the state of a sample directly from a single set of measurements (b)
  • No refers to gene numbers as mentioned herein.
  • ProbeSetID refers to the identification number on the Affymetrix gene chip HT HG-U133A.
  • State refers the respective renal cell cancer specific states.
  • the term “Mode” defines whether a gene has to be over- or under-expressed for state A, B, C or D.
  • “Invers” indicates under-expression and “normal” indicates over-expression relative to the value "limit value”, which describes the value which used as control to decide on over-expression or under-expression.
  • the term “Fit” describes the reliability of the limit value with a value of 0.5 indicating maximum reliability. The limit value will be put in the respective software, which is used for expression analysis, individually for each gene.
  • SEQ ID No. refers to SEQ ID No. of the sequence listing. Table 5

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Medical Informatics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Chemical & Material Sciences (AREA)
  • Biotechnology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Evolutionary Biology (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Analytical Chemistry (AREA)
  • Genetics & Genomics (AREA)
  • Databases & Information Systems (AREA)
  • Organic Chemistry (AREA)
  • Public Health (AREA)
  • Evolutionary Computation (AREA)
  • Epidemiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioethics (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • Zoology (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • Wood Science & Technology (AREA)
  • Oncology (AREA)
  • Microbiology (AREA)
  • Hospice & Palliative Care (AREA)
  • Biochemistry (AREA)
  • General Engineering & Computer Science (AREA)
  • Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

La présente invention concerne l'utilisation d'états discrets et de signatures pour classer des échantillons de cancer, de préférence des échantillons de cancer à cellules rénales.
PCT/EP2012/072578 2011-11-15 2012-11-14 États discrets destinés à être utilisés en tant que marqueurs biologiques pour des cancers, tel que le cancer à cellules rénales Ceased WO2013072346A2 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
EP11189178 2011-11-15
EP11189206 2011-11-15
EP11189206.3 2011-11-15
EP11189178.4 2011-11-15

Publications (2)

Publication Number Publication Date
WO2013072346A2 true WO2013072346A2 (fr) 2013-05-23
WO2013072346A9 WO2013072346A9 (fr) 2013-11-21

Family

ID=47148837

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2012/072578 Ceased WO2013072346A2 (fr) 2011-11-15 2012-11-14 États discrets destinés à être utilisés en tant que marqueurs biologiques pour des cancers, tel que le cancer à cellules rénales

Country Status (1)

Country Link
WO (1) WO2013072346A2 (fr)

Non-Patent Citations (7)

* Cited by examiner, † Cited by third party
Title
BELEUT ET AL., BMC CANCER, vol. 12, 2012, pages 3 10
BELEUT ET AL., BMC CANCER, vol. 12, 2012, pages 310
HANAHAN; WEINBERG, THE HALLMARKS OF CANCER, 2000
NATURE, vol. 490, 2012, pages 61 - 70
SAEED ET AL., METHODS ENZYMOL., vol. 411, 2006, pages 134 - 193
TUSHER ET AL., PROC NATL ACAD SCI USA, vol. 98, no. 9, 2001, pages 5116 - 5121
WANG Y ET AL., LANCET, 2005, pages 671 - 9

Also Published As

Publication number Publication date
WO2013072346A9 (fr) 2013-11-21

Similar Documents

Publication Publication Date Title
Sanchez-Carbayo et al. Defining molecular profiles of poor outcome in patients with invasive bladder cancer using oligonucleotide microarrays
Singh et al. Gene expression correlates of clinical prostate cancer behavior
Ye et al. Predicting hepatitis B virus–positive metastatic hepatocellular carcinomas using gene expression profiling and supervised machine learning
CN103733065B (zh) 用于癌症的分子诊断试验
US20080032299A1 (en) Methods for prognosis and treatment of solid tumors
JP2007049991A (ja) 乳癌の骨への再発の予測
WO2010063121A1 (fr) Procédés pour l’identification de biomarqueur et biomarqueur pour le cancer du poumon à grandes cellules
EP2417545A1 (fr) Procédé pour le diagnostic in vitro d'une maladie complexe
EP2304631A1 (fr) Algorithmes de prediction de resultat pour des patientes atteintes de cancer du sein traite par chimiotherapie avec atteinte ganglionnaire
WO2006119593A1 (fr) Pronostic algorithmique du cancer base sur les genes gene-based algorithmic cancer prognosis
CN101313306A (zh) 用于鉴别鼻咽癌中的预后型亚类的基因表达谱
US9721067B2 (en) Accelerated progression relapse test
US20150038359A1 (en) Method of predicting outcome in cancer patients
SG174333A1 (en) Identification of biologically and clinically essential genes and gene pairs, and methods employing the identified genes and gene pairs
JP2008520251A (ja) 固形腫瘍の予後および処置のための方法およびシステム
US20130157887A1 (en) Discrete states for use as biomarkers
Liu et al. Bioinformatics analysis with graph-based clustering to detect gastric cancer-related pathways
Thomassen et al. Prediction of metastasis from low‐malignant breast cancer by gene expression profiling
EP1797429A2 (fr) Procedes et kits pour la prevision d'un succes therapeutique et d'une survie exempte de rechute en therapie du cancer
Gevaert et al. Prediction of cancer outcome using DNA microarray technology: past, present and future
WO2013072346A2 (fr) États discrets destinés à être utilisés en tant que marqueurs biologiques pour des cancers, tel que le cancer à cellules rénales
US11976330B2 (en) MiRNA signature expression in cancer
Tiwari Microarrays and cancer diagnosis.
AU2011251964A1 (en) Discrete states for use as biomarkers
WO2014009798A1 (fr) Profilage d'expression génique à l'aide de 5 gènes pour prédire le pronostic dans le cancer du sein

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12784022

Country of ref document: EP

Kind code of ref document: A2

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12784022

Country of ref document: EP

Kind code of ref document: A2