+

WO2010073248A2 - Signature d'expression génétique pour la classification de tissu provenant d'échantillons tumoraux - Google Patents

Signature d'expression génétique pour la classification de tissu provenant d'échantillons tumoraux Download PDF

Info

Publication number
WO2010073248A2
WO2010073248A2 PCT/IL2009/001212 IL2009001212W WO2010073248A2 WO 2010073248 A2 WO2010073248 A2 WO 2010073248A2 IL 2009001212 W IL2009001212 W IL 2009001212W WO 2010073248 A2 WO2010073248 A2 WO 2010073248A2
Authority
WO
WIPO (PCT)
Prior art keywords
cancer
nucleic acid
origin
acid sequence
sample
Prior art date
Application number
PCT/IL2009/001212
Other languages
English (en)
Other versions
WO2010073248A3 (fr
Inventor
Ranit Aharonov
Nitzan Rosenfeld
Shai Rosenwald
Original Assignee
Rosetta Genomics Ltd.
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 Rosetta Genomics Ltd. filed Critical Rosetta Genomics Ltd.
Priority to CN200980157378XA priority Critical patent/CN102333888B/zh
Publication of WO2010073248A2 publication Critical patent/WO2010073248A2/fr
Publication of WO2010073248A3 publication Critical patent/WO2010073248A3/fr
Priority to IL212979A priority patent/IL212979A/en
Priority to US13/167,489 priority patent/US8802599B2/en
Priority to US13/856,190 priority patent/US9096906B2/en
Priority to US14/320,113 priority patent/US20140315739A1/en
Priority to US14/746,487 priority patent/US20150368724A1/en
Priority to US15/853,258 priority patent/US20180127835A1/en
Priority to US15/909,145 priority patent/US20190032142A1/en
Priority to US16/226,406 priority patent/US20190241966A1/en

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
    • 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
    • C12Q1/6809—Methods for determination or identification of nucleic acids involving differential 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
    • 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
    • G16B25/00—ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression
    • G16B25/10—Gene or protein expression profiling; Expression-ratio estimation or normalisation
    • 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
    • G16B40/10—Signal processing, e.g. from mass spectrometry [MS] or from PCR
    • 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
    • C12Q2600/178—Oligonucleotides characterized by their use miRNA, siRNA or ncRNA
    • 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
    • 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
    • G16B25/00—ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression

Definitions

  • the present invention relates to methods for classification of cancers and the identification of their tissue of origin. Specifically the invention relates to microRNA molecules associated with specific cancers, as well as various nucleic acid molecules relating thereto or derived therefrom.
  • microRNAs are a novel class of non-coding, regulatory RNA genes 1"3 which are involved in oncogenesis 4 and show remarkable tissue-specificity 5"7 . They have emerged as highly tissue-specific biomarkers 2 ' 5 ' 6 postulated to play important roles in encoding developmental decisions of differentiation.
  • Various studies have tied microRNAs to the development of specific malignancies 4 .
  • MicroRNAs are also stable in tissue, stored frozen or as formalin-fixed, paraffin-embedded (FFPE) samples, and in serum. Hundreds of thousands of patients in the U.S. are diagnosed each year with a cancer that has already metastasized, without a clearly identified primary site.
  • FFPE formalin-fixed, paraffin-embedded
  • Oncologists and pathologists are constantly faced with a diagnostic dilemma when trying to identify the primary origin of a patient's metastasis. As metastases need to be treated according to their primary origin, accurate identification of the metastases' primary origin can be critical for determining appropriate treatment.
  • CUP cancer of unknown primary
  • the present invention provides specific nucleic acid sequences for use in the identification, classification and diagnosis of specific cancers and tumor tissue of origin.
  • the nucleic acid sequences can also be used as prognostic markers for prognostic evaluation and determination of appropriate treatment of a subject based on the abundance of the nucleic acid sequences in a biological sample.
  • the present invention further provides a method for accurate identification of tumor tissue origin.
  • microRNA-based classifier for tumor classification.
  • microRNA expression levels were measured in 903 paraffin-embedded samples from 26 different tumor classes, corresponding to 18 distinct tissues and organs, including primary and metastatic tumors.
  • microRNA microarray, of the samples as well as qRT-PCR data, were used to construct a classifier, based on 48 tissue-specific microRNAs, each linked to specific differential-diagnosis roles.
  • the overall sensitivity of the independent blinded test in identifying the tumor tissue of origin is 84%, with 97% specificity. High confidence predictions reach 90% sensitivity with 99% specificity.
  • the findings demonstrate the utility of microRNA as novel biomarkers for the tissue of origin of a metastatic tumor.
  • the classifier has wide biological as well as diagnostic applications.
  • the present invention provides a method of identifying a tissue of origin of a biological sample, the method comprising: obtaining a biological sample from a subject; determining an expression profile of individual nucleic acids for a predetermined set of microRNAs; and classifying the tissue of origin for said sample by a classifier.
  • said classifier is a decision tree model.
  • the present invention provides a method of classifying a tissue of origin of a biological sample, the method comprising: obtaining a biological sample from a subject; determining an expression profile in said sample of nucleic acid sequences selected from the group consisting of SEQ ID NOS: 1-49, or a sequence having at least about 80% identity thereto; and comparing said expression profile to a reference expression profile by using a classifier algorithm; whereby the expression of any of said nucleic acid sequences or combinations thereof allows the identification of the tissue of origin of said sample.
  • said classifier algorithm is a decision tree classifier, logistic regression classifier, linear regression classifier, nearest neighbor classifier (including K nearest neighbors), neural network classifier, Gaussian mixture model (GMM) classifier and Support Vector Machine (SVM) classifier, nearest centroid classifier, random forest classifier or any boosting or bootstrap aggregating (bagging) of those classifiers.
  • said tissue is selected from the group consisting of liver, lung, bladder, prostate, breast, colon, ovary, testis, stomach, thyroid, pancreas, brain, head and neck, kidney, melanocytes, thymus, biliary tract and esophagus.
  • said biological sample is a cancerous sample.
  • the present invention provides a method of classifying a cancer, the method comprising: obtaining a biological sample from a subject; measuring the relative abundance in said sample of nucleic acid sequences selected from the group consisting of SEQ DD NOS: 1-49 or a sequence having at least about 80% identity thereto; and comparing said obtained measurement to reference values representing abundance of said nucleic acid sequences by using a classifier algorithm; whereby the relative abundance of said nucleic acid sequences allows the classification of said cancer.
  • said reference values are predetermined thresholds.
  • said sample is obtained from a subject with a metastatic cancer.
  • said sample is obtained from a subject with cancer of unknown primary (CUP).
  • said sample is obtained from a subject with a primary cancer.
  • said sample is a tumor of unidentified origin, a metastatic tumor or a primary tumor.
  • said cancer is selected from the group consisting of liver cancer, biliary tract cancer, lung cancer, bladder cancer, prostate cancer, breast cancer, colon cancer, ovarian cancer, testicular cancer, stomach cancer, thyroid cancer, pancreas cancer, brain cancer, head and neck cancer, kidney cancer, melanoma, thymus cancer and esophagus cancer.
  • said lung cancer is selected from the group consisting of lung carcinoid, lung small cell carcinoma, lung adenocarcinoma, and lung squamous cell carcinoma.
  • said brain cancer is selected from the group consisting of brain astrocytoma and brain oligodendroglioma.
  • said thyroid cancer is selected from the group consisting of thyroid follicular, thyroid papillary and thyroid medullary cancer.
  • said ovarian cancer is selected from the group consisting of ovarian endometrioid and ovarian serous cancer.
  • said testicular cancer is selected from the group consisting of testicular non-seminoma and testicular seminoma.
  • said esophagus cancer is selected from the group consisting of esophagus adenocarcinoma and esophagus squamous cell carcinoma.
  • said head and neck cancer is selected from the group consisting of larynx carcinoma, pharynx carcinoma and nose carcinoma.
  • said biliary tract cancer is selected from the group consisting of cholangiocarcinoma and gallbladder adenocarcinoma.
  • said biological sample is selected from the group consisting of bodily fluid, a cell line, a tissue sample, a biopsy sample, a needle biopsy sample, a surgically removed sample, and a sample obtained by tissue-sampling procedures.
  • the biological sample is a fine needle aspiration (FNA) sample.
  • said tissue is a fresh, frozen, fixed, wax-embedded or formalin-fixed paraffin-embedded (FFPE) tissue.
  • the classification method of the present invention comprises the use of at least one classifier algorithm, said classifier algorithm is selected from the group consisting of decision tree classifier, logistic regression classifier, linear regression classifier, nearest neighbor classifier (including K nearest neighbors), neural network classifier, Gaussian mixture model (GMM) classifier and Support Vector Machine (SVM) classifier, nearest centroid classifier, random forest classifier or any boosting or bootstrap aggregating (bagging) of those classifiers.
  • the classifier may use a decision tree structure (including binary tree) or a voting
  • the invention further provides a method for classifying a cancer of liver origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 9, 25, 26, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of liver origin.
  • the invention further provides a method for classifying a cancer of testicular origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 26, 41, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of testicular origin.
  • the invention further provides a method for classifying a cancer of testicular seminoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 26, 31, 41, 45, 48 or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of testicular seminoma origin.
  • the invention further provides a method for classifying a cancer of melanoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 15, 17, 26, 41, 46, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of melanoma origin.
  • the invention further provides a method for classifying a cancer of kidney origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 7, 15, 17, 26, 41, 46, 47, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of kidney origin.
  • the invention further provides a method for classifying a cancer of brain origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 7, 15, 17, 26, 41, 46, 47, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of brain origin.
  • the invention further provides a method for classifying a cancer of brain astrocytoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 7, 10, 15, 17, 26, 41, 46, 47, or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of brain astrocytoma origin.
  • the invention further provides a method for classifying a cancer of brain oligodendroglioma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 7, 10, 15, 17, 26, 41, 46, 47, or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of brain oligodendroglioma origin.
  • the invention further provides a method for classifying a cancer of thyroid medullary origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 6, 17-19, 24, 26, 32, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of thyroid medullary origin.
  • the invention further provides a method for classifying a cancer of lung carcinoid origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 6, 17-19, 24, 26, 32, 36, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of lung carcinoid origin.
  • the invention further provides a method for classifying a cancer of lung small cell carcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 6, 17-19, 24, 26, 32, 36, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of lung small cell carcinoma origin.
  • the invention further provides a method for classifying a cancer of colon origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 1, 3, 4, 6, 17-19, 21, 26, 29, 34, 37, 41, 42, 48, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of colon origin.
  • the invention further provides a method for classifying a cancer of stomach origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 1, 3, 4, 6, 17-19, 21, 26, 29, 34, 37, 41, 42, 48, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of stomach origin.
  • the invention further provides a method for classifying a cancer of pancreas origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 1, 3, 6, 17-19, 21, 26, 28, 29, 33, 37, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of pancreas origin.
  • the invention further provides a method for classifying a cancer of biliary tract origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 1, 3, 6, 9, 17-19, 21, 25, 26, 28, 29, 33, 37, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of biliary tract origin.
  • the invention further provides a method for classifying a cancer of prostate origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ED NOS: 3, 6, 17-21, 26, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of prostate origin.
  • the invention further provides a method for classifying a cancer of ovarian origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 5, 6, 11, 17-21, 26, 30, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of ovarian origin.
  • the invention further provides a method for classifying a cancer of ovarian endometrioid origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 2, 3, 5, 6, 11, 17-22, 26, 30, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of ovarian endometrioid origin.
  • the invention further provides a method for classifying a cancer of ovarian serous origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ED NOS: 2, 3, 5, 6, 11, 17-22, 26, 30, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of ovarian serous origin.
  • the invention further provides a method for classifying a cancer of breast origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 5, 6, 11, 17-22, 26, 30, 39, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of breast origin.
  • the invention further provides a method for classifying a cancer of lung adenocarcinoma origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 5, 6, 8, 11, 16 -22, 26, 27, 30, 37, 39, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of lung adenocarcinoma origin.
  • the invention further provides a method for classifying a cancer of papillary thyroid origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 5, 6, 8, 11, 16-22, 26, 27, 29, 30, 37- 39, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of papillary thyroid origin.
  • the invention further provides a method for classifying a cancer of follicular thyroid origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 5, 6, 8, 11, 16-22, 26, 27, 29, 30, 37-39, 41, 42, or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of follicular thyroid origin.
  • the invention further provides a method for classifying a cancer of thymus origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3, 5, 6, 11, 16-22, 26, 27, 29, 30, 35, 39, 41, 42, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of thymus origin.
  • the invention further provides a method for classifying a cancer of bladder origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3-6, 11, 16-22, 26, 27, 29, 30, 35, 39, 41, 42, 44, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of bladder origin.
  • the invention further provides a method for classifying a cancer of lung squamous origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3-6, 11, 16-23, 26, 27, 29, 30, 32, 35, 39, 41, 42, 44, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of lung squamous origin.
  • the invention further provides a method for classifying a cancer of head and neck origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3-6, 11, 14, 16-23, 26, 27, 29, 30, 32, 35, 37, 39, 41, 42, 44, 45, or a sequence having at least about 80% identity thereto in a sample obtained from a subject; wherein the abundance of said nucleic acid sequence is indicative of a cancer of head and neck origin.
  • the invention further provides a method for classifying a cancer of esophagus origin, the method comprising measuring the relative abundance of a nucleic acid sequence selected from the group consisting of SEQ ID NOS: 3-6, 11, 14, 16-23, 26, 27, 29, 30, 32, 35, 37, 39, 41, 42, 44, 45, or a sequence having at least about 80% identity thereto in said sample; wherein the abundance of said nucleic acid sequence is indicative of a cancer of esophagus origin.
  • the nucleic acid sequence expression profile or relative abundance is determined by a method selected from the group consisting of nucleic acid hybridization and nucleic acid amplification.
  • the nucleic acid hybridization is performed using a solid-phase nucleic acid biochip array or in situ hybridization.
  • the nucleic acid amplification method is realtime PCR.
  • the real-time PCR method may comprise forward and reverse primers.
  • the forward primer comprises a sequence selected from the group consisting of SEQ ID NOS: 50-98 and 150.
  • the reverse primer comprises SEQ ID NO: 288.
  • the real-time PCR method further comprises a probe.
  • the probe comprises a sequence selected from the group consisting of a sequence that is complementary to a sequence selected from SEQ ED NOS: 1-49; a fragment thereof and a sequence having at least about 80% identity thereto.
  • the probe comprises a sequence selected from the group consisting of SEQ ID NOS: 99-149 and 151.
  • the present invention provides a kit for cancer classification, said kit comprising a probe comprising a sequence selected from the group consisting of a sequence that is complementary to a sequence selected from SEQ ID NOS: SEQ ID NOS: 1-49; a fragment thereof and a sequence having at least about 80% identity thereto.
  • the probe comprises a sequence selected from the group consisting of SEQ ID NOS: 99-149 and 151.
  • said cancer is selected from the group consisting of liver cancer, biliary tract cancer, lung cancer, bladder cancer, prostate cancer, breast cancer, colon cancer, ovarian cancer, testicular cancer, stomach cancer, thyroid cancer, pancreas cancer, brain cancer, head and neck cancer, kidney cancer, melanoma, thymus cancer and esophagus cancer.
  • Figures 1A-1C demonstrate the structure of the binary decision-tree classifier, with 26 nodes (numbered, Table 3) and 27 leaves. Each node is a binary decision between two sets of samples, those to the left and right of the node. A series of binary decisions, starting at node #1 and moving downwards, lead to one of the possible tumor types, which are the "leaves" of the tree. A sample which is classified to the left branch at node #1 continues to node #2, otherwise it continues to node #3. A sample that reaches node #2, is further classified to either the left branch at node #2, and is assigned to the
  • liver class or to the right branch at node #2, and is assigned to the "biliary tract carcinoma” class. Decisions are made at consecutive nodes using microRNA expression levels, until an end-point ("leaf of the tree) is reached, indicating the predicted class for this sample.
  • end-point indicating the predicted class for this sample.
  • clinico-pathological considerations were combined with properties observed in the training set data. Developing a different classifier for e.g. male and female cases or for different tumor sites would inefficiently exploit measured data and would require unwieldy numbers of samples. Instead, exceptions were noted for several special cases: For samples from female patients, testis or prostate origins were excluded from the KNN database, and the right branch was automatically taken in node 3 and node 16 in the decision-tree.
  • liver origin hepatocellular carcinoma and biliary tract carcinomas from within the liver
  • brain origin was excluded and the right branch taken at node 7. Additional information is thus incorporated into the classification decision without loss of generality or need to retrain the classifier.
  • Figure 2 demonstrates binary decisions at node #1 of the decision-tree.
  • Tumors originating from tissues at the left branch at node #1 including tumors from the "liver” class and the “biliary tract” class (liver-cholangio; diamonds) are easily separated from tumors of non- liver and non-biliary tract origins (right branch at node #2; gray squares) using the expression levels of hsa-miR-200c (SEQ ID NO: 26) and hsa-miR-122 (SEQ ID NO: 6) (with one outlier), with a linear classifier (the diagonal line).
  • Figure 3 demonstrates binary decisions at node #5 of the decision-tree.
  • Tumors of epithelial origin left branch at node #5, marked by diamonds
  • tumors of non-epithelial origin right branch at node #5, marked by squares
  • the gray area (with higher levels of hsa-miR-200c) marks the region classified as epithelial (left branch) at this node.
  • Figure 4 demonstrates binary decisions at node #7 of the decision-tree.
  • Tumors originating in the brain are easily separated from tumors of kidney origin (squares) using the expression levels of hsa-miR-124 (SEQ ID NO: 7) and hsa-miR-9* (SEQ ID NO: 47).
  • Figure 5 demonstrates binary decisions at node #10 of the decision-tree.
  • Neuroendocrine tumors originating in the lung are easily separated from tumors of thyroid-medullary origin (squares) using the expression levels of hsa-miR- 200a (SEQ ID NO: 24) and hsa-miR-222 (SEQ ID NO: 32).
  • Figure 6 demonstrates binary decisions at node #12 of the decision-tree. Tumors originating in the gastrointestinal tract (left branch at node #12, marked by diamonds) are easily separated from tumors of non digestive origins (right branch at node #12, marked by squares) using the expression levels of hsa-miR-106a (SEQ ID NO: 3) and hsa-miR- 192 (SEQ ID NO: 21).
  • Figure 7 demonstrates binary decisions at node #16 of the decision-tree.
  • FIG. 8 A shows that the measured levels (normalized C t , inversely proportional to log(abundance)) of hsa-miR- 200c (SEQ ID NO: 26) and hsa-miR-122 (SEQ ID NO: 6) are compared for all training set samples, indicating the left and right branches of node #1 (circles and stars respectively).
  • Figure 8B shows that upon re-examination of the metastatic brain tumor by immunohistochemistry (blinded to the results of the microRNA classifier), this tumor was indeed found to be negative for lung specific markers: the sample was negative for immunohistochemical staining by both CK7 and TTFl, as well as CK20, CEA, CA125, s-100, thyroglobulin, chromogranin, synaptophysin, CD56, GFAP, calcitonin, and anterior pituitary hormones, while staining positive for CAM5.5' and AE1/AE3. This staining pattern was compatible with hepatocellular carcinoma, prompting further staining for HEPAl and alpha fetoprotein.
  • H&E staining (upper panel) showed that the metastasis is composed of sheets of cells with abundant eosinophilic cytoplasm and round to oval nuclei.
  • HEPA-I showed strong and specific immunopositivity (lower panel).
  • the present invention is based in part on the discovery that specific nucleic acid sequences can be used for the identification of the tissue-of-origin of a tumor.
  • the present invention provides a sensitive, specific and accurate method which can be used to distinguish between different tissues and tumor origins.
  • a new microRNA-based classifier was developed for determining tissue origin of tumors based on a surprisingly small number of 48 microRNAs markers. The classifier uses a specific algorithm and allows a clear interpretation of the specific biomarkers. High confidence predictions reach 90% sensitivity and 99% specificity.
  • each node in the classification tree may be used as an independent differential diagnosis tool, for example in the identification of different types of lung cancer.
  • the performance of the classifier using a small number of markers highlights the utility of microRNA as tissue-specific cancer biomarkers, and provides an effective means for facilitating diagnosis of CUP and more generally of identifying tumor origins of metastases. The possibility to distinguish between different tumor origins facilitates providing the patient with the best and most suitable treatment.
  • the present invention provides diagnostic assays and methods, both quantitative and qualitative for detecting, diagnosing, monitoring, staging and prognosticating cancers by comparing the levels of the specific microRNA molecules of the invention. Such levels are preferably measured in at least one of biopsies, tumor samples, fine- needle aspiration (FNA), cells, tissues and/or bodily fluids.
  • the present invention provides methods for diagnosing the presence of a specific cancer by analyzing the levels of said microRNA molecules in biopsies, tumor samples, cells, tissues or bodily fluids.
  • determining the levels of said microRNA in biopsies, tumor samples, cells, tissues or bodily fluid is particularly useful for discriminating between different cancers.
  • All the methods of the present invention may optionally further include measuring levels of other cancer markers.
  • Other cancer markers in addition to said microRNA molecules, useful in the present invention will depend on the cancer being tested and are known to those of skill in the art.
  • Assay techniques that can be used to determine levels of gene expression, such as the nucleic acid sequence of the present invention, in a sample derived from a patient are well known to those of skill in the art.
  • Such assay methods include, but are not limited to, reverse transcriptase PCR (RT-PCR) assays, nucleic acid microarrays and biochip analysis, immunohistochemistry assays, in situ hybridization assays, competitive-binding assays, northern blot analyses and ELISA assays.
  • RT-PCR reverse transcriptase PCR
  • the assay is based on expression level of 48 microRNAs in RNA extracted from FFPE metastatic tumor tissue.
  • the test is a quantitative real time reverse transcriptase polymerase chain reaction (qRT-PCR) test.
  • RNA is first polyadenylated and then reverse transcribed using universal poly(T) adapter to create cDNA.
  • the cDNA is amplified using specific forward primer and universal reverse primer (with a sequence complementary to the 5' tail of the poly(T) adapter), and detected by specific MGB probes (see specific sequences in Table 1).
  • the expression levels are used to infer the sample origin using analysis techniques such as but not limit to decision tree classifier, logistic regression classifier, linear regression classifier, nearest neighbor classifier (including K nearest neighbors), neural network classifier and nearest centroid classifier.
  • the expression levels are used to make binary decisions (at each relevant node) following the pre-defined structure of the binary decision-tree (defined using the training set).
  • the resulting P is compared to a threshold level PTH (which was also determined using the training set), and the classification continues to the left or right branch according to whether P is larger or smaller than PTH for that node. This continues until an end-point ("leaf) of the tree is reached.
  • PTH a threshold level
  • Training the tree algorithm means determining: the tree structure (which nodes there are and what is on each side), which miRs are used in each node and the values of b ⁇ , bl, b2... and PTH. These were determined by a combination of machine learning, optimization algorithm, and trial and error by experts in machine learning and diagnostic algorithms.
  • correlations and/or hierarchical clustering can be used to assess the similarity of the expression level of the nucleic acid sequences of the invention between a specific sample and different exemplars of cancer samples.
  • An arbitrary threshold on the expression level of one or more nucleic acid sequences can be set for assigning a sample or cancer sample to one of two groups.
  • expression levels of one or more nucleic acid sequences of the invention are combined by a method such as logistic regression to define a metric which is then compared to previously measured samples or to a threshold.
  • the threshold for assignment is treated as a parameter, which can be used to quantify the confidence with which samples are assigned to each class.
  • the threshold for assignment can be scaled to favor sensitivity or specificity, depending on the clinical scenario.
  • the correlation value to the reference data generates a continuous score that can be scaled and provides diagnostic information on the likelihood that a sample belongs to a certain class of cancer origin or type. In multivariate analysis, the microRNA signature provides a high level of prognostic information.
  • expression level of the nucleic acids is used to classify a test sample by comparison to a training set of samples.
  • the test sample is compared in turn to each one of the training set samples.
  • Each such pairwise comparison is performed by comparing the expression levels of one or multiple nucleic acids between the test sample and the specific training sample.
  • Each such pairwise comparison generates a combined metric for the multiple nucleic acids, which can be calculated by various numeric methods such as correlation, cosine, Euclidian distance, mean square distance, or other methods known to those skilled in the art.
  • the training samples are then ranked according to this metric, and the samples with the highest values of the metric (or lowest values, according to the type of metric) are identified, indicating those samples that are most similar to the test sample.
  • K this generates a list that includes the K training samples that are most similar to the test sample.
  • Various methods can then be applied to identify from this list the predicted class of the test sample.
  • the test sample is predicted to belong to the class that has the highest number of representative in the list of K most-similar training samples (this method is known as the K Nearest Neighbors method).
  • Other embodiments may provide a list of predictions including all or part of the classes represented in the list, those classes that are represented more than a given minimum number of times, or other voting schemes whereby classes are grouped together.
  • Bindached or “immobilized”, as used herein, to refer to a probe and a solid support means that the binding between the probe and the solid support is sufficient to be stable under conditions of binding, washing, analysis, and removal.
  • the binding may be covalent or non-covalent. Covalent bonds may be formed directly between the probe and the solid support or may be formed by a cross linker or by inclusion of a specific reactive group on either the solid support or the probe or both molecules.
  • Non-covalent binding may be one or more of electrostatic, hydrophilic, and hydrophobic interactions. Included in non-covalent binding is the covalent attachment of a molecule, such as streptavidin, to the support and the non-covalent binding of a biotinylated probe to the streptavidin. Immobilization may also involve a combination of covalent and non-covalent interactions.
  • baseline such as streptavidin
  • Base means the initial cycles of PCR, in which there is little change in fluorescence signal.
  • Biological sample means a sample of biological tissue or fluid that comprises nucleic acids. Such samples include, but are not limited to, tissue or fluid isolated from subjects. Biological samples may also include sections of tissues such as biopsy and autopsy samples, FFPE samples, frozen sections taken for histological purposes, blood, blood fraction, plasma, serum, sputum, stool, tears, mucus, hair, skin, urine, effusions, ascitic fluid, amniotic fluid, saliva, cerebrospinal fluid, cervical secretions, vaginal secretions, endometrial secretions, gastrointestinal secretions, bronchial secretions, cell line, tissue sample, or secretions from the breast.
  • a biological sample may be provided by fine-needle aspiration (FNA).
  • FNA fine-needle aspiration
  • a biological sample may be provided by removing a sample of cells from a subject but can also be accomplished by using previously isolated cells (e.g., isolated by another person, at another time, and/or for another purpose), or by performing the methods described herein in vivo.
  • Archival tissues such as those having treatment or outcome history, may also be used.
  • Biological samples also include explants and primary and/or transformed cell cultures derived from animal or human tissues.
  • cancer is meant to include all types of cancerous growths or oncogenic processes, metastatic tissues or malignantly transformed cells, tissues, or organs, irrespective of histopathologic type or stage of invasiveness.
  • cancers include, but are not limited, to solid tumors and leukemias, including: apudoma, choristoma, branchioma, malignant carcinoid syndrome, carcinoid heart disease, carcinoma (e.g., Walker, basal cell, basosquamous, Brown-Pearce, ductal, Ehrlich tumor, non-small cell lung (e.g., lung squamous cell carcinoma, lung adenocarcinoma and lung undifferentiated large cell carcinoma), oat cell, papillary, bronchiolar, bronchogenic, squamous cell, and transitional cell), histiocytic disorders, leukemia (e.g., B cell, mixed cell, null cell, T cell, T-cell chronic, HTLV-II
  • classification refers to a procedure and/or algorithm in which individual items are placed into groups or classes based on quantitative information on one or more characteristics inherent in the items (referred to as traits, variables, characters, features, etc.) and based on a statistical model and/or a training set of previously labeled items.
  • a "classification tree” is a decision tree that places categorical variables into classes.
  • complement “Complement” or “complementary” is used herein to refer to a nucleic acid may mean Watson-Crick (e.g., A-T/U and C-G) or Hoogsteen base pairing between nucleotides or nucleotide analogs of nucleic acid molecules.
  • a full complement or fully complementary means 100% complementary base pairing between nucleotides or nucleotide analogs of nucleic acid molecules.
  • the complementary sequence has a reverse orientation (5 ' -3 ').
  • Ct signals represent the first cycle of PCR where amplification crosses a threshold (cycle threshold) of fluorescence. Accordingly, low values of Ct represent high abundance or expression levels of the microRNA. hi some embodiments the PCR Ct signal is normalized such that the normalized
  • a "data processing routine” refers to a process that can be embodied in software that determines the biological significance of acquired data (i.e., the ultimate results of an assay or analysis). For example, the data processing routine can make determination of tissue of origin based upon the data collected, hi the systems and methods herein, the data processing routine can also control the data collection routine based upon the results determined.
  • the data processing routine and the data collection routines can be integrated and provide feedback to operate the data acquisition, and hence provide assay-based judging methods. data set
  • data set refers to numerical values obtained from the analysis. These numerical values associated with analysis may be values such as peak height and area under the curve. data structure
  • data structure refers to a combination of two or more data sets, applying one or more mathematical manipulations to one or more data sets to o obtain one or more new data sets, or manipulating two or more data sets into a form that provides a visual illustration of the data in a new way.
  • An example of a data structure prepared from manipulation of two or more data sets would be a hierarchical cluster.
  • Detection means detecting the presence of a component in a sample. Detection also means detecting the absence of a component. Detection also means determining the level of a component, either quantitatively or qualitatively. differential expression
  • differential expression means qualitative or quantitative differences in the temporal and/or spatial gene expression patterns within and among cells and tissue.
  • a differentially expressed gene may qualitatively have its expression altered, including an activation or inactivation, in, e.g., normal versus diseased tissue. Genes may be turned on or turned off in a particular state, relative to another state, thus permitting comparison of two or more states.
  • a qualitatively regulated gene may exhibit an expression pattern within a state or cell type which may be detectable by standard techniques. Some genes may be expressed in one state or cell type, but not in both.
  • the difference in expression may be quantitative, e.g., in that expression is modulated, up-regulated, resulting in an increased amount of transcript, or down-regulated, resulting in a decreased amount of transcript.
  • the degree to which expression differs needs only to be large enough to quantify via standard characterization techniques such as expression arrays, quantitative reverse transcriptase PCR, northern blot analysis, real-time PCR, in situ hybridization and RNase protection.
  • expression profile is used broadly to include a genomic expression profile, e.g., an expression profile of microRNAs. Profiles may be generated by any convenient means for determining a level of a nucleic acid sequence, e.g., quantitative hybridization of microRNA, labeled microRNA, amplified microRNA, cDNA, etc., quantitative PCR, ELISA for quantitation, and the like, and allow the analysis of differential gene expression between two samples.
  • a subject or patient tumor sample e.g., cells or collections thereof, e.g., tissues, is assayed. Samples are collected by any convenient method, as known in the art.
  • Nucleic acid sequences of interest are nucleic acid sequences that are found to be predictive, including the nucleic acid sequences provided above, where the expression profile may include expression data for 5, 10, 20, 25, 50, 100 or more of the nucleic acid sequences, including all of the listed nucleic acid sequences.
  • expression profile means measuring the relative abundance of the nucleic acid sequences in the measured samples. expression ratio
  • “Expression ratio” refers to relative expression levels of two or more nucleic acids as determined by detecting the relative expression levels of the corresponding nucleic acids in a biological sample.
  • Fram is used herein to indicate a non- full-length part of a nucleic acid. Thus, a fragment is itself also a nucleic acid.
  • Gene may be a natural (e.g., genomic) or synthetic gene comprising transcriptional and/or translational regulatory sequences and/or a coding region and/or non-translated sequences (e.g., introns, 5'- and 3 '-untranslated sequences).
  • the coding region of a gene may be a nucleotide sequence coding for an amino acid sequence or a functional RNA, such as tRNA, rRNA, catalytic RNA, siRNA, miRNA or antisense RNA.
  • a gene may also be an mRNA or cDNA corresponding to the coding regions (e.g., exons and miRNA) optionally comprising 5'- or 3 '-untranslated sequences linked thereto.
  • a gene may also be an amplified nucleic acid molecule produced in vitro, comprising all or a part of the coding region and/or 5'- or 3 '-untranslated sequences linked thereto.
  • Groove binder/minor groove binder may be used interchangeably and refer to small molecules that fit into the minor groove of double-stranded DNA, typically in a sequence-specific manner.
  • Minor groove binders may be long, flat molecules that can adopt a crescent-like shape and thus fit snugly into the minor groove of a double helix, often displacing water.
  • Minor groove binding molecules may typically comprise several aromatic rings connected by bonds with torsional freedom such as furan, benzene, or pyrrole rings.
  • Minor groove binders may be antibiotics such as netropsin, distamycin, berenil, pentamidine and other aromatic diamidines, Hoechst 33258, SN 6999, aureolic anti-tumor drugs such as chromomycin and mithramycin, CC- 1065, dihydrocyclopyrroloindole tripeptide (DPI 3 ), l,2-dihydro-(3H)-pyrrolo[3,2-e]indole-7- carboxylate (CDPI 3 ), and related compounds and analogues, including those described in Nucleic Acids in Chemistry and Biology, 2nd ed., Blackburn and Gait, eds., Oxford University Press, 1996, and PCT Published Application No.
  • antibiotics such as netropsin, distamycin, berenil, pentamidine and other aromatic diamidines, Hoechst 33258, SN 6999, aureolic anti-tumor drugs such as chromo
  • a minor groove binder may be a component of a primer, a probe, a hybridization tag complement, or combinations thereof. Minor groove binders may increase the T m of the primer or a probe to which they are attached, allowing such primers or probes to effectively hybridize at higher temperatures.
  • Host cell may be a naturally occurring cell or a transformed cell that may contain a vector and may support replication of the vector.
  • Host cells may be cultured cells, explants, cells in vivo, and the like.
  • Host cells may be prokaryotic cells, such as E. coli, or eukaryotic cells, such as yeast, insect, amphibian, or mammalian cells, such as CHO and HeLa cells. identity
  • nucleic acids or polypeptide sequences mean that the sequences have a specified percentage of residues that are the same over a specified region. The percentage may be calculated by optimally aligning the two sequences, comparing the two sequences over the specified region, determining the number of positions at which the identical residue occurs in both sequences to yield the number of matched positions, dividing the number of matched positions by the total number of positions in the specified region, and multiplying the result by 100 to yield the percentage of sequence identity.
  • “In situ detection”, as used herein, means the detection of expression or expression levels in the original site, hereby meaning in a tissue sample such as biopsy.
  • k-nearest neighbor refers to a classification method that classifies a point by calculating the distances between the point and points in the training data set. It then assigns the point to the class that is most common among its k-nearest neighbors (where k is an integer).
  • Label means a composition detectable by spectroscopic, photochemical, biochemical, immunochemical, chemical, or other physical means.
  • useful labels include 32 P, fluorescent dyes, electron-dense reagents, enzymes (e.g., as commonly used in an ELISA), biotin, digoxigenin, or haptens and other entities which can be made detectable.
  • a label may be incorporated into nucleic acids and proteins at any position. logistic regression
  • Logistic regression is part of a category of statistical models called generalized linear models. Logistic regression can allow one to predict a discrete outcome, such as 1 group membership, from a set of variables that may be continuous, discrete, dichotomous, or a mix of any of these. The dependent or response variable can be dichotomous, for example, one of two possible types of cancer. Logistic regression models the natural log of the odds ratio, i.e., the ratio of the probability of belonging to the first group (P) over the probability of belonging to the second group (J-P), as a linear combination of the different expression levels (in log-space).
  • the logistic regression output can be used as a classifier by prescribing that a case or sample will be classified into the first type if P is greater than 0.5 or 50%.
  • the calculated probability P can be used as a variable in other contexts, such as a ID or 2D threshold classifier.
  • 1D/2D threshold classifier may mean an algorithm for classifying a case or sample such as a cancer sample into one of two possible types such as two types of cancer.
  • ID threshold classifier the decision is based on one variable and one predetermined threshold value; the sample is assigned to one class if the variable exceeds the threshold and to the other class if the variable is less than the threshold.
  • a 2D threshold classifier is an algorithm for classifying into one of two types based on the values of two variables.
  • a threshold may be calculated as a function (usually a continuous or even a monotonic function) of the first variable; the decision is then reached by comparing the second variable to the calculated threshold, similar to the ID threshold classifier. metastasis
  • Metalastasis means the process by which cancer spreads from the place at which it first arose as a primary tumor to other locations in the body.
  • the metastatic progression of a primary tumor reflects multiple stages, including dissociation from neighboring primary tumor cells, survival in the circulation, and growth in a secondary location. node
  • a “node” is a decision point in a classification (i.e., decision) tree. Also, a point in a neural net that combines input from other nodes and produces an output through application of an activation function.
  • a “leaf” is a node not further split, the terminal grouping in a classification or decision tree.
  • the depiction of a single strand also defines the sequence of the complementary strand.
  • a nucleic acid also encompasses the complementary strand of a depicted single strand.
  • Many variants of a nucleic acid may be used for the same purpose as a given nucleic acid.
  • a nucleic acid also encompasses substantially identical nucleic acids and complements thereof.
  • a single strand provides a probe that may hybridize to a target sequence under stringent hybridization conditions.
  • a nucleic acid also encompasses a probe that hybridizes under stringent hybridization conditions.
  • Nucleic acids may be single-stranded or double-stranded, or may contain portions of both double-stranded and single-stranded sequences.
  • the nucleic acid may be DNA, both genomic and cDNA, RNA, or a hybrid, where the nucleic acid may contain combinations of deoxyribo- and ribo-nucleotides, and combinations of bases including uracil, adenine, thymine, cytosine, guanine, inosine, xanthine hypoxanthine, isocytosine and isoguanine.
  • Nucleic acids may be obtained by chemical synthesis methods or by recombinant methods.
  • a nucleic acid will generally contain phosphodiester bonds, although nucleic acid analogs may be included that may have at least one different linkage, e.g., phosphoramidate, phosphorothioate, phosphorodithioate, or O-methylphosphoroamidite linkages and peptide nucleic acid backbones and linkages.
  • Other analog nucleic acids include those with positive backbones, non-ionic backbones and non-ribose backbones, including those described in U.S. Pat. Nos. 5,235,033 and 5,034,506, which are incorporated herein by reference.
  • Nucleic acids containing one or more non-naturally occurring or modified nucleotides are also included within one definition of nucleic acids.
  • the modified nucleotide analog may be located for example at the 5 '-end and/or the 3'-end of the nucleic acid molecule.
  • Representative examples of nucleotide analogs may be selected from sugar- or backbone-modified ribonucleotides.
  • nucleobase-modif ⁇ ed ribonucleotides i.e., ribonucleotides, containing a non-naturally occurring nucleobase instead of a naturally occurring nucleobase such as uridine or cytidine modified at the 5-position, e.g., 5-(2-amino) propyl uridine, 5-bromo uridine; adenosine and guanosine modified at the 8-position, e.g., 8-bromo guanosine; deaza nucleotides, e.g., 7-deaza-adenosine; O- and N-alkylated nucleotides, e.g., N6- methyl adenosine are suitable.
  • a non-naturally occurring nucleobase such as uridine or cytidine modified at the 5-position, e.g., 5-(2-amino) propyl uridine, 5-bromo uridine
  • the 2'-OH-group may be replaced by a group selected from H, OR, R, halo, SH, SR, NH 2 , NHR, NR 2 or CN, wherein R is C1-C6 alkyl, alkenyl or alkynyl and halo is F, Cl, Br or I.
  • Modified nucleotides also include nucleotides conjugated with cholesterol through, e.g., a hydroxyprolinol linkage as described in Krutzfeldt et al, Nature 2005; 438:685-689, Soutschek et al, Nature 2004; 432:173-178, and U.S. Patent Publication No. 20050107325, which are incorporated herein by reference.
  • Modifications of the ribose-phosphate backbone may be done for a variety of reasons, e.g., to increase the stability and half-life of such molecules in physiological environments, to enhance diffusion across cell membranes, or as probes on a biochip.
  • the backbone modification may also enhance resistance to degradation, such as in the harsh endocytic environment of cells.
  • the backbone modification may also reduce nucleic acid clearance by hepatocytes, such as in the liver and kidney. Mixtures of naturally occurring nucleic acids and analogs may be made; alternatively, mixtures of different nucleic acid analogs, and mixtures of naturally occurring nucleic acids and analogs may be made.
  • Probe means an oligonucleotide capable of binding to a target nucleic acid of complementary sequence through one or more types of chemical bonds, usually through complementary base pairing, usually through hydrogen bond formation. Probes may bind target sequences lacking complete complementarity with the probe sequence depending upon the stringency of the hybridization conditions. There may be any number of base pair mismatches which will interfere with hybridization between the target sequence and the single-stranded nucleic acids described herein. However, if the number of mutations is so great that no hybridization can occur under even the least stringent of hybridization conditions, the sequence is not a complementary target sequence.
  • a probe may be single-stranded or partially single- and partially double- stranded. The strandedness of the probe is dictated by the structure, composition, and properties of the target sequence. Probes may be directly labeled or indirectly labeled such as with biotin to which a streptavidin complex may later bind. reference value
  • the term "reference value” or “reference expression profile” refers to a criterion expression value to which measured values are compared in order to determine the detection of a specific cancer.
  • the reference value may be based on the abundance of the nucleic acids, or may be based on a combined metric score thereof. In preferred embodiments the reference value is determined from statistical analysis of studies that compare microRNA expression with known clinical outcomes. sensitivity
  • Sensitivity may mean a statistical measure of how well a binary classification test correctly identifies a condition, for example, how frequently it correctly classifies a cancer into the correct type out of two possible types.
  • the sensitivity for class A is the proportion of cases that are determined to belong to class "A” by the test out of the cases that are in class "A”, as determined by some absolute or gold standard.
  • Specificity may mean a statistical measure of how well a binary classification test correctly identifies a condition, for example, how frequently it correctly classifies a cancer into the correct type out of two possible types.
  • the sensitivity for class A is the proportion of cases that are determined to belong to class "not A” by the test out of the cases that are in class "not A”, as determined by some absolute or gold standard.
  • stringent hybridization conditions mean conditions under which a first nucleic acid sequence (e.g., probe) will hybridize to a second nucleic acid sequence (e.g., target), such as in a complex mixture of nucleic acids. Stringent conditions are sequence-dependent and will be different in different circumstances. Stringent conditions may be selected to be about 5-10°C lower than the thermal melting point (T m ) for the specific sequence at a defined ionic strength pH.
  • the T m may be the temperature (under defined ionic strength, pH, and nucleic concentration) at which 50% of the probes complementary to the target hybridize to the target sequence at equilibrium (as the target sequences are present in excess, at T m , 50% of the probes are occupied at equilibrium).
  • Stringent conditions may be those in which the salt concentration is less than about 1.0 M sodium ion, such as about 0.01-1.0 M sodium ion concentration (or other salts) at pH 7.0 to 8.3 and the temperature is at least about 30°C for short probes (e.g., about 10-50 nucleotides) and at least about 60°C for long probes (e.g., greater than about 50 nucleotides).
  • Stringent conditions may also be achieved with the addition of destabilizing agents such as formamide.
  • a positive signal may be at least 2 to 10 times background hybridization.
  • Exemplary stringent hybridization conditions include the following: 50% formamide, 5x SSC, and 1% SDS, incubating at 42°C, or, 5x SSC, 1% SDS, incubating at 65°C, with wash in 0.2x SSC, and O.l% SDS at 65°C.
  • substantially complementary means that a first sequence is at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 97%, 98% or 99% identical to the complement of a second sequence over a region of 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100 or more nucleotides, or that the two sequences hybridize under stringent hybridization conditions. substantially identical
  • substantially identical means that a first and a second sequence are at least 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 97%, 98% or 99% identical over a region of 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100 or more nucleotides or amino acids, or with respect to nucleic acids, if the first sequence is substantially complementary to the complement of the second sequence.
  • the term "subject” refers to a mammal, including both human and other mammals.
  • the methods of the present invention are preferably applied to human subjects. target nucleic acid
  • Target nucleic acid means a nucleic acid or variant thereof that may be bound by another nucleic acid.
  • a target nucleic acid may be a DNA sequence.
  • the target nucleic acid may be RNA.
  • the target nucleic acid may comprise a mRNA, tRNA, shRNA, siRNA or Piwi-interacting RNA, or a pri-miRNA, pre-miRNA, miRNA, or anti-miRNA.
  • the target nucleic acid may comprise a target miRNA binding site or a variant thereof.
  • One or more probes may bind the target nucleic acid.
  • the target binding site may comprise 5-100 or 10-60 nucleotides.
  • the target binding site may comprise a total of 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15,16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30- 40, 40-50, 50-60, 61, 62 or 63 nucleotides.
  • the target site sequence may comprise at least 5 nucleotides of the sequence of a target miRNA binding site disclosed in U.S. Patent Application Nos. 11/384,049, 11/418,870 or 11/429,720, the contents of which are incorporated herein. threshold
  • tissue sample As used herein, a tissue sample is tissue obtained from a tissue biopsy using methods well known to those of ordinary skill in the related medical arts.
  • Tumor refers to all neoplastic cell growth and proliferation, whether malignant or benign, and all pre-cancerous and cancerous cells and tissues. variant
  • nucleic acid means (i) a portion of a referenced nucleotide sequence; (ii) the complement of a referenced nucleotide sequence or portion thereof; (iii) a nucleic acid that is substantially identical to a referenced nucleic acid or the complement thereof; or (iv) a nucleic acid that hybridizes under stringent conditions to the referenced nucleic acid, complement thereof, or a sequence substantially identical thereto.
  • wild-type sequence refers to a coding, a non-coding or an interface sequence which is an allelic form of sequence that performs the natural or normal function for that sequence. Wild-type sequences include multiple allelic forms of a cognate sequence, for example, multiple alleles of a wild type sequence may encode silent or conservative changes to the protein sequence that a coding sequence encodes.
  • the present invention employs miRNAs for the identification, classification and diagnosis of specific cancers and the identification of their tissues of origin. 1. microRNA processing
  • a gene coding for microRNA may be transcribed leading to production of a miRNA primary transcript known as the pri-miRNA.
  • the pri-miRNA may comprise a hairpin with a stem and loop structure.
  • the stem of the hairpin may comprise mismatched bases.
  • the pri-miRNA may comprise several hairpins in a polycistronic structure.
  • the hairpin structure of the pri-miRNA may be recognized by Drosha, which is an RNase III endonuclease. Drosha may recognize terminal loops in the pri-miRNA and cleave approximately two helical turns into the stem to produce a 60-70 nt precursor known as the pre-miRNA. Drosha may cleave the pri-miRNA with a staggered cut typical of RNase III endonucleases yielding a pre-miRNA stem loop with a 5' phosphate and ⁇ 2 nucleotide 3' overhang. Approximately one helical turn of stem (-10 nucleotides) extending beyond the Drosha cleavage site may be essential for efficient processing.
  • Drosha is an RNase III endonuclease.
  • Drosha may recognize terminal loops in the pri-miRNA and cleave approximately two helical turns into the stem to produce a 60-70 nt precursor known as the pre-mi
  • the pre-miRNA may then be actively transported from the nucleus to the cytoplasm by Ran- GTP and the export receptor Ex-portin-5.
  • the pre-miRNA may be recognized by Dicer, which is also an RNase III endonuclease. Dicer may recognize the double-stranded stem of the pre-miRNA. Dicer may also cut off the terminal loop two helical turns away from the base of the stem loop, leaving an additional 5' phosphate and a ⁇ 2 nucleotide 3' overhang.
  • the resulting siRNA- like duplex which may comprise mismatches, comprises the mature miRNA and a similar-sized fragment known as the miRNA*.
  • the miRNA and miRNA* may be derived from opposing arms of the pri-miRNA and pre-miRNA. MiRNA* sequences may be found in libraries of cloned miRNAs, but typically at lower frequency than the miRNAs.
  • RISC RNA-induced silencing complex
  • RISC RNA-induced silencing complex
  • Various proteins can form the RISC, which can lead to variability in specificity for miRNA/miRNA* duplexes, binding site of the target gene, activity of miRNA (repress or activate), and which strand of the miRNA/miRNA* duplex is loaded in to the RISC.
  • the miRNA strand of the miRNA:miRNA* duplex is loaded into the RISC, the miRNA* may be removed and degraded.
  • the strand of the miRNA:miRNA* duplex that is loaded into the RISC may be the strand whose 5' end is less tightly paired, hi cases where both ends of the miRNA:miRNA* have roughly equivalent 5' pairing, both miRNA and miRNA* may have gene silencing activity.
  • the RISC may identify target nucleic acids based on high levels of complementarity between the miRNA and the mRNA, especially by nucleotides 2-7 of the miRNA. Only one case has been reported in animals where the interaction between the miRNA and its target was along the entire length of the miRNA.
  • miR-196 mediates the cleavage of the Hox B8 mRNA (Yekta et al. Science 2004; 304:594-596). Otherwise, such interactions are known only in plants (Bartel & Bartel 2003; 132:709-717).
  • miRNAs may direct the RISC to down-regulate gene expression by either of two mechanisms: mRNA cleavage or translational repression.
  • the miRNA may specify cleavage of the mRNA if the mRNA has a certain degree of complementarity to the miRNA. When a miRNA guides cleavage, the cut may be between the nucleotides pairing to residues 10 and 11 of the miRNA. Alternatively, the miRNA may repress translation if the miRNA does not have the requisite degree of complementarity to the miRNA. Translational repression may be more prevalent in animals since animals may have a lower degree of complementarity between the miRNA and binding site.
  • any pair of miRNA and miRNA* there may be variability in the 5' and 3' ends of any pair of miRNA and miRNA*. This variability may be due to variability in the enzymatic processing of Drosha and Dicer with respect to the site of cleavage. Variability at the 5' and 3' ends of miRNA and miRNA* may also be due to mismatches in the stem structures of the pri-miRNA and pre-miRNA. The mismatches of the stem strands may lead to a population of different hairpin structures. Variability in the stem structures may also lead to variability in the products of cleavage by Drosha and Dicer. 2. Nucleic Acids
  • Nucleic acids are provided herein.
  • the nucleic acids comprise the sequences of SEQ ID NOS: 1-288 or variants thereof.
  • the variant may be a complement of the referenced nucleotide sequence.
  • the variant may also be a nucleotide sequence that is substantially identical to the referenced nucleotide sequence or the complement thereof.
  • the variant may also be a nucleotide sequence which hybridizes under stringent conditions to the referenced nucleotide sequence, complements thereof, or nucleotide sequences substantially identical thereto.
  • the nucleic acid may have a length of from about 10 to about 250 nucleotides.
  • the nucleic acid may have a length of at least 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, 125, 150, 175, 200 or 250 nucleotides.
  • the nucleic acid may be synthesized or expressed in a cell (in vitro or in vivo) using a synthetic gene described herein.
  • the nucleic acid may be synthesized as a single-strand molecule and hybridized to a substantially complementary nucleic acid to form a duplex.
  • the nucleic acid may be introduced to a cell, tissue or organ in a single- or double-stranded form or capable of being expressed by a synthetic gene using methods well known to those skilled in the art, including as described in U.S. Patent No. 6,506,559, which is incorporated herein by reference.
  • Table 1 SEQ ID NOS of miRs, forward primers and MGB probes
  • the nucleic acid may further comprise one or more of the following: a peptide, a protein, a RNA-DNA hybrid, an antibody, an antibody fragment, a Fab fragment, and an aptamer.
  • the nucleic acid may comprise a sequence of a pri-miRNA or a variant thereof.
  • the pri-miRNA sequence may comprise from 45-30,000, 50-25,000, 100-20,000, 1,000- 1,500 or 80-100 nucleotides.
  • the sequence of the pri-miRNA may comprise a pre- miRNA, miRNA and miRNA*, as set forth herein, and variants thereof.
  • the sequence of the pri-miRNA may comprise any of the sequences of SEQ ID NOS: 1-49 or variants thereof.
  • the pri-miRNA may comprise a hairpin structure.
  • the hairpin may comprise a first and a second nucleic acid sequence that are substantially complimentary.
  • the first and second nucleic acid sequence may be from 37-50 nucleotides.
  • the first and second nucleic acid sequence may be separated by a third sequence of from 8-12 nucleotides.
  • the hairpin structure may have a free energy of less than -25 Kcal/mole, as calculated by the Vienna algorithm with default parameters, as described in Hofacker et al. (Monatshefte f. Chemie 1994; 125:167-188), the contents of which are incorporated herein by reference.
  • the hairpin may comprise a terminal loop of 4-20, 8-12 or 10 nucleotides.
  • the pri-miRNA may comprise at least 19% adenosine nucleotides, at least 16% cytosine nucleotides, at least 23% thymine nucleotides and at least 19% guanine nucleotides.
  • the nucleic acid may also comprise a sequence of a pre-miRNA or a variant thereof.
  • the pre-miRNA sequence may comprise from 45-90, 60-80 or 60-70 nucleotides.
  • the sequence of the pre-miRNA may comprise a miRNA and a miRNA* as set forth herein.
  • the sequence of the pre-miRNA may also be that of a pri-miRNA excluding from 0-160 nucleotides from the 5' and 3' ends of the pri-miRNA.
  • the sequence of the pre-miRNA may comprise the sequence of SEQ ED NOS: 1-49 or variants thereof. 6.
  • the nucleic acid may also comprise a sequence of a miRNA (including miRNA*) or a variant thereof.
  • the miRNA sequence may comprise from 13-33, 18-24 or 21-23 nucleotides.
  • the miRNA may also comprise a total of at least 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39 or 40 nucleotides.
  • the sequence of the miRNA may be the first 13-33 nucleotides of the pre-miRNA.
  • the sequence of the miRNA may also be the last 13-33 nucleotides of the pre-miRNA.
  • the sequence of the miRNA may comprise the sequence of SEQ ID NOS: 1-49 or variants thereof.
  • Probes A probe comprising a nucleic acid described herein is also provided. Probes may be used for screening and diagnostic methods, as outlined below.
  • the probe may be attached or immobilized to a solid substrate, such as a biochip.
  • the probe may have a length of from 8 to 500, 10 to 100 or 20 to 60 nucleotides.
  • the probe may also have a length of at least 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 45, 50, 60, 70, 80, 90, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280 or 300 nucleotides.
  • the probe may further comprise a linker sequence of from 10-60 nucleotides.
  • the probe may comprise a nucleic acid that is complementary to a sequence selected from the group consisting of SEQ ID NOS: 1-49 or variants thereof.
  • the probe may comprise a sequence selected from the group consisting of SEQ H) NOS: 99-149 and 151.
  • Biochip A biochip is also provided.
  • the biochip may comprise a solid substrate comprising an attached probe or plurality of probes described herein.
  • the probes may be capable of hybridizing to a target sequence under stringent hybridization conditions.
  • the probes may be attached at spatially defined addresses on the substrate. More than one probe per target sequence may be used, with either overlapping probes or probes to different sections of a particular target sequence.
  • the probes may be capable of hybridizing to target sequences associated with a single disorder appreciated by those in the art.
  • the probes may either be synthesized first, with subsequent attachment to the biochip, or may be directly synthesized on the biochip.
  • the solid substrate may be a material that may be modified to contain discrete individual sites appropriate for the attachment or association of the probes and is amenable to at least one detection method.
  • substrates include glass and modified or functionalized glass, plastics (including acrylics, polystyrene and copolymers of styrene and other materials, polypropylene, polyethylene, polybutylene, polyurethanes, TeflonJ, etc.), polysaccharides, nylon or nitrocellulose, resins, silica or silica-based materials including silicon and modified silicon, carbon, metals, inorganic glasses and plastics.
  • the substrates may allow optical detection without appreciably fluorescing.
  • the substrate may be planar, although other configurations of substrates may be used as well. For example, probes may be placed on the inside surface of a tube, for flow-through sample analysis to minimize sample volume.
  • the substrate may be flexible, such as flexible foam, including closed cell foams made of particular plastics.
  • the biochip and the probe may be derivatized with chemical functional groups for subsequent attachment of the two.
  • the biochip may be derivatized with a chemical functional group including, but not limited to, amino groups, carboxyl groups, oxo groups or thiol groups.
  • the probes may be attached using functional groups on the probes either directly or indirectly using a linker.
  • the probes may be attached to the solid support by either the 5' terminus, 3' terminus, or via an internal nucleotide.
  • the probe may also be attached to the solid support non-covalently.
  • biotinylated oligonucleotides can be made, which may bind to surfaces covalently coated with streptavidin, resulting in attachment.
  • probes may be synthesized on the surface using techniques such as photopolymerization and photolithography.
  • diagnosis refers to classifying pathology, or a symptom, determining a severity of the pathology (grade or stage), monitoring pathology progression, forecasting an outcome of pathology and/or prospects of recovery.
  • the phrase "subject in need thereof refers to an animal or human subject who is known to have cancer, at risk of having cancer (e.g., a genetically predisposed subject, a subject with medical and/or family history of cancer, a subject who has been exposed to carcinogens, occupational hazard, environmental hazard) and/or a subject who exhibits suspicious clinical signs of cancer (e.g., blood in the stool or melena, unexplained pain, sweating, unexplained fever, unexplained loss of weight up to anorexia, changes in bowel habits (constipation and/or diarrhea), tenesmus (sense of incomplete defecation, for rectal cancer specifically), anemia and/or general weakness).
  • cancer e.g., a genetically predisposed subject, a subject with medical and/or family history of cancer, a subject who has been exposed to carcinogens, occupational hazard, environmental hazard
  • a subject who exhibits suspicious clinical signs of cancer e.g.,
  • the subject in need thereof can be a healthy human subject undergoing a routine well-being check up.
  • Analyzing presence of malignant or pre-malignant cells can be effected in vivo or ex vivo, whereby a biological sample (e.g., biopsy) is retrieved.
  • a biological sample e.g., biopsy
  • Such biopsy samples comprise cells and may be an incisional or excisional biopsy. Alternatively, the cells may be retrieved from a complete resection.
  • treatment regimen refers to a treatment plan that specifies the type of treatment, dosage, schedule and/or duration of a treatment provided to a subject in need thereof (e.g., a subject diagnosed with a pathology).
  • the selected treatment regimen can be an aggressive one which is expected to result in the best clinical outcome (e.g., complete cure of the pathology) or a more moderate one which may relieve symptoms of the pathology yet results in incomplete cure of the pathology. It will be appreciated that in certain cases the treatment regimen may be associated with some discomfort to the subject or adverse side effects (e.g., damage to healthy cells or tissue).
  • the type of treatment can include a surgical intervention (e.g., removal of lesion, diseased cells, tissue, or organ), a cell replacement therapy, an administration of a therapeutic drug (e.g., receptor agonists, antagonists, hormones, chemotherapy agents) in a local or a systemic mode, an exposure to radiation therapy using an external source (e.g., external beam) and/or an internal source (e.g., brachytherapy) and/or any combination thereof.
  • a surgical intervention e.g., removal of lesion, diseased cells, tissue, or organ
  • a cell replacement therapy e.g., an administration of a therapeutic drug (e.g., receptor agonists, antagonists, hormones, chemotherapy agents) in a local or a systemic mode
  • an exposure to radiation therapy using an external source e.g., external beam
  • an internal source e.g., brachytherapy
  • the dosage, schedule and duration of treatment can vary, depending on the severity of pathology and the selected type of treatment, and those
  • a method of diagnosis comprises detecting an expression level of a specific cancer-associated nucleic acid in a biological sample.
  • the sample may be derived from a patient. Diagnosis of a specific cancer state in a patient may allow for prognosis and selection of therapeutic strategy. Further, the developmental stage of cells may be classified by determining temporarily expressed specific cancer- associated nucleic acids.
  • In situ hybridization of labeled probes to tissue arrays may be performed.
  • the skilled artisan can make a diagnosis, a prognosis, or a prediction based on the findings. It is further understood that the nucleic acid sequence which indicate the diagnosis may differ from those which indicate the prognosis and molecular profiling of the condition of the cells may lead to distinctions between responsive or refractory conditions or may be predictive of outcomes.
  • Kits A kit is also provided and may comprise a nucleic acid described herein together with any or all of the following: assay reagents, buffers, probes and/or primers, and sterile saline or another pharmaceutically acceptable emulsion and suspension base, hi addition, the kits may include instructional materials containing directions (e.g., protocols) for the practice of the methods described herein.
  • the kit may further comprise a software package for data analysis of expression profiles.
  • the kit may be a kit for the amplification, detection, identification or quantification of a target nucleic acid sequence.
  • the kit may comprise a poly (T) primer, a forward primer, a reverse primer, and a probe.
  • kits may comprise, in suitable container means, an enzyme for labeling the miRNA by incorporating labeled nucleotide or unlabeled nucleotides that are subsequently labeled. It may also include one or more buffers, such as reaction buffer, labeling buffer, washing buffer, or a hybridization buffer, compounds for preparing the miRNA probes, components for in situ hybridization and components for isolating miRNA.
  • Other kits of the invention may include components for making a nucleic acid array comprising miRNA, and thus may include, for example, a solid support.
  • FFPE formalin-fixed paraffin-embedded
  • Tumor samples were obtained from several sources. Institutional review approvals were obtained for all samples in accordance with each institute's institutional review board or
  • RNA extraction included primary tumors and metastases of defined origins, according to clinical records. Tumor content was at least 50% for >95% of samples, as determined by a pathologist based on hematoxylin-eosin (H&E) stained slides. 204 of the 903 samples were used only in the validation phase, as an independent blinded test set. The reference diagnosis of these samples from the original clinical record was confirmed by an additional review of pathological specimens. 2.
  • miR array platform Custom microarrays (Agilent Technologies, Santa Clara, CA) were produced by printing DNA oligonucleotide probes to more than 900 human microRNAs. Each probe, printed in triplicate, carried up to 22 -nucleotide (nt) linker at the 3' end of the microRNA's complement sequence, in addition to an amine group used to couple the probes to coated glass slides. Twenty ⁇ M of each probe were dissolved in 2X SSC + 0.0035% SDS and spotted in triplicate on Schott Nexterion® Slide E-coated microarray slides using a Genomic Solutions® BioRobotics MicroGrid II according the MicroGrid manufacturer's directions.
  • nt nucleotide
  • Fifty-four negative control probes were designed using the sense sequences of different microRNAs. Two groups of positive control probes were designed to hybridize to miR array: (i) synthetic small RNAs were spiked to the RNA before labeling to verify the labeling efficiency; and (ii) probes for abundant small RNA (e.g., small nuclear RNAs (U43, U49, U24, Z30, U6, U48, U44), 5.8s and 5s ribosomal RNA) are spotted on the array to verify RNA quality. The slides were blocked in a solution containing 50 mM ethanolamine, 1 M Tris (pH9.0) and 0.1% SDS for 20 min at 50 0 C, then thoroughly rinsed with water and spun dry. 4. Cy-dye labeling of miRNA for miR array
  • RNA-linker p-rCrU-Cy/dye (Dharmacon)
  • Dharmacon an RNA-linker
  • the labeling reaction contained total RNA, spikes (0.1-20 fmoles), 300 ng RNA- linker-dye, 15% DMSO, Ix ligase buffer and 20 units of T4 RNA ligase (NEB), and proceeded at 4°C for 1 h, followed by 1 h at 37°C.
  • Array signal calculation and normalization Triplicate spots were combined to produce one signal for each probe by taking the logarithmic mean of reliable spots. All data were log-transformed (natural base) and the analysis was performed in log-space.
  • the aim of a logistic regression model is to use several features, such as expression levels of several microRNAs, to assign a probability of belonging to one of two possible groups, such as two branches of a node in a binary decision-tree.
  • Logistic regression models the natural log of the odds ratio, i.e., the ratio of the probability of belonging to the first group, for example, the left branch in a node of a binary decision- tree (P) over the probability of belonging to the second group, for example, the right branch in such a node (1-P), as a linear combination of the different expression levels (in log-space).
  • P binary decision- tree
  • the logistic regression assumes that:
  • the regression error on each sample is the difference between the assigned probability P and the true "probability" of this sample, i.e., 1 if this sample is in the left branch group and 0 otherwise.
  • the training and optimization of the logistic regression model calculates the parameters P and the p-values (for each microRNA by the WaId statistic and for the overall model by the ⁇ 2 (chi-square) difference), maximizing the likelihood of the data given the model and minimizing the total regression error
  • the probability output of the logistic model is here converted to a binary decision p p ⁇ p by comparing " to a threshold, denoted by m , i.e., if TM then the sample belongs to the left branch ("first group") and vice versa.
  • a threshold denoted by m
  • TM if TM then the sample belongs to the left branch
  • first group a threshold
  • m i.e., if TM then the sample belongs to the left branch
  • a probability threshold of 0.5 leads to a minimization of the sum of the regression errors.
  • a modification which p adjusts the probability threshold ( m ) was used in order to minimize the overall number of mistakes at each node (Table 3).
  • the original data contain the expression levels of multiple microRNAs for each sample, i.e., multiple of data features.
  • training the classifier for each node only a small subset of these features was selected and used for optimizing a logistic regression model. In the initial training this was done using a forward stepwise scheme. The features were sorted in order of decreasing log-likelihoods, and the logistic model was started off and optimized with the first feature. The second feature was then added, and the model re-optimized. The regression error of the two models was compared: if the addition of the feature did not provide a significant advantage (a ⁇ 2 difference less than 7.88, p-value of 0.005), the new feature was discarded. Otherwise, the added feature was kept.
  • Adding a new feature may make a previous feature redundant (e.g., if they are very highly correlated). To check for this, the process iteratively checks if the feature with lowest likelihood can be discarded (without losing ⁇ 2 difference as above). After ensuring that the current set of features is compact in this sense, the process continues to test the next feature in the sorted list, until features are exhausted. No limitation on the number of feature was inserted into the algorithm, but in most cases 2-3 features were selected.
  • the stepwise logistic regression method was used on subsets of the training set samples by re-sampling the training set with repetition ("bootstrap"), so that each of the
  • KNN K-nearest-neighbors
  • the KNN algorithm calculates the distance (Pearson correlation) of any sample to all samples in the training set, and classifies the sample by the majority vote of the k samples which are most similar (k being a parameter of the classifier).
  • RNA (1 ⁇ g) is subjected to polyadenylation reaction as described before (Gilad et al, PLoS ONE 2008; 3:e3148). Briefly, RNA is incubated in the presence of poly (A) polymerase (PAP) (Takara-2180A), MnC12, and ATP for 1 h at 37°C. Reverse transcription is performed on the total RNA.
  • An oligodT primer harboring a consensus sequence (complementary to the reverse primer, oligodT starch, an N nucleotide (a mixture of all A, C, and G) and V nucleotide (mixture of four nucleotides) was used for the reverse transcription reaction.
  • the primer was first annealed to the poly A-RNA and then subjected to a reverse transcription reaction of Superscript II RT (Invitrogen).
  • the cDNA was then amplified by a real-time PCR reaction, using a microRNA-specific forward primer, TaqMan probe and universal reverse primer that is complementary to the 3' sequence of the oligo dT tail.
  • the reactions were incubated for 10 min at 95 °C, followed by 42 cycles of 95 0 C for 15 s and 60°C for 1 min.
  • qRT-PCR was performed using probes for the 104 candidate microRNAs, of which 5 were tested with two different forward primers, and for U6 snoRNA.
  • the training samples were kept with average C t below 36 and at least 30 microRNAs detected (C t ⁇ 38). Each sample was normalized by subtracting from the C t of each microRNA the average C t of all microRNAs of the sample, and adding back a scaling constant (the average C t over the entire sample set). Feature selection and classifier training were using the scaled C t as the input signal. The feature selection resulted in a set of 48 microRNAs.
  • the decision-tree (Fig. 1) used logistic regression on combinations of two-to-three microRNAs in each node to make binary decisions.
  • KNN was based on comparing the expression of all 48 microRNAs in each sample to all other samples in the training database.
  • the decision-tree and KNN each return a predicted tissue of origin and histological type where applicable.
  • the classifier returns the two different predictions or a single consensus prediction if the predictions concur.
  • the tissue of origin is returned as a consensus prediction with no histological type indicated.
  • the negative control was a no-RNA sample that served to detect potential contaminations, and should not give any signal in the PCR reaction.
  • the extracted RNA, together with a positive control sample underwent cDNA preparation and 48 microRNAs were measured by qRT-PCR in duplicates in one 96-well plate per sample.
  • the positive control was a specific RNA sample that should meet defined C t ranges in the assay. Quality assessment of each well was based on the fluorescence amplification curve, using thresholds on the maximal fluorescence and the linear slope as a function of the measured Q. For each microRNA, C t m ⁇ R was calculated by taking the average C t of the two repeats.
  • Quality assessment for each sample was based on the number and identity of expressed microRNAs (C t ⁇ 38) and the average C t of the measured microRNAs.
  • C t m ⁇ R values for each sample were normalized by rescaling as described above. The rescaled values were used as input to the classifier that was trained using qRT-PCR data (as described above).
  • An alternative assay was constructed, which does not identify bladder as an origin, i.e., differentiates between 25 classes representing 17 tissue origins.
  • a validation set of 255 new FFPE tumor samples was used to assess the performance of the assay, representing 26 different tumor origins or "classes" (see Table 2 for a summary of samples). About half of the samples in the set were metastatic tumors to different sites (e.g., lung, bone, brain and liver). Tumor percentage was at least 50% for all samples in the set.
  • a tumor classifier was built using the microRNA expression levels by applying a binary tree classification scheme (Fig. 1).
  • This framework is set up to utilize the specificity of microRNAs in tissue differentiation and embryogenesis: different microRNAs are involved in various stages of tissue specification, and are used by the algorithm at different decision points or "nodes".
  • the tree breaks up the complex multi- tissue classification problem into a set of simpler binary decisions.
  • classes which branch out earlier in the tree are not considered, reducing interference from irrelevant samples and further simplifying the decision.
  • the decision at each node can then be accomplished using only a small number of microRNA biomarkers, which have well-defined roles in the classification (Table 3).
  • the structure of the binary tree was based on a hierarchy of tissue development and morpho logical similarity 18 , which was modified by prominent features of the microRNA expression patterns.
  • the expression patterns of microRNAs indicated a significant difference between liver- cholangio tumors and tumors of non-liver origin, and these are therefore separated at node #1 (Fig. 2) into separate branches (Fig. 1).
  • the test performance was assessed using an independent set of 204 validation samples. These archival samples included primary as well as metastatic tumor samples, preserved as FFPE blocks, whose original clinical diagnosis ("reference diagnosis") was one of the origins on which the classifier was trained. The samples were processed by personnel who were blinded to the original reference diagnosis for these samples, and classifications were automatically generated by dedicated software. 16 of the 204 samples (8%) failed QA criteria. For 188 samples (92%), including 87 metastatic tumor samples (46% of the samples), the test was completed successfully and produced tissue- of-origin predictions. For 159 of these samples (84%), the reference diagnosis for tissue of origin was predicted by at least one of the two classifiers (Table 4).
  • the two classifiers agreed, generating a consensus prediction for a single tissue- of-origin.
  • the sensitivity was 90% (111/124 of the classifications agreed with the reference diagnosis), and it exceeded 90% for most tissue-types.
  • Specificity (negative agreement) in this group ranged from 94% to 100%.
  • FFPE sections from 73 of the validation samples were processed independently and blindly in a second laboratory. Data and classifications for these samples were compared between the two laboratories.
  • the mean correlation for the qRT-PCR signals was 0.979 (4 samples had correlation coefficients between 0.91 and 0.95, all other correlations were greater than 0.95).
  • the two labs disagreed on only 4 samples. For another 8, they had one of two answers in common and for the remaining 61, classifications matched perfectly between the two laboratories, demonstrating the precision of the test.
  • microRNAs in Table 3 For some of the microRNAs in Table 3, other variant microRNAs having a similar seed sequence (identical nucleotides 2-8) are known in the human genome (see
  • microRNAs in Table 3 For some of the microRNAs in Table 3, other microRNAs that are known in the human genome are located in close proximity on the genome (genomic cluster) (see Table 6), and are co-transcibed with the indicated miRs. These microRNAs from nearly the same genomic location may be substituted for the indicated miRs.
  • microRNAs with similar sequence For some of the microRNAs in Table 3, other microRNAs that are known in the human genome have similar sequences (less than 6 mismatches in the sequence) (see Table 7), and may therefore also be captured by probes with the same design. These microRNAs with similar overall sequence may be substituted for the indicated miRs. Table 7: microRNAs with similar sequence

Landscapes

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

Abstract

La présente invention concerne un procédé pour la classification de cancers et de tissus d'origine par l'analyse de profils d'expression de micro-ARN spécifiques et de molécules d'acide nucléique qui y sont associées. La classification selon un cadre d'expression basée sur un arbre des micro-ARN permet l'optimisation de traitement, et la détermination de thérapie spécifique.
PCT/IL2009/001212 2007-03-27 2009-12-23 Signature d'expression génétique pour la classification de tissu provenant d'échantillons tumoraux WO2010073248A2 (fr)

Priority Applications (9)

Application Number Priority Date Filing Date Title
CN200980157378XA CN102333888B (zh) 2008-12-24 2009-12-23 用于肿瘤样本起源组织分类的基因表达签名
IL212979A IL212979A (en) 2008-12-24 2011-05-18 Gene expression signature to classify the source tissue of cancer tumor samples
US13/167,489 US8802599B2 (en) 2007-03-27 2011-06-23 Gene expression signature for classification of tissue of origin of tumor samples
US13/856,190 US9096906B2 (en) 2007-03-27 2013-04-03 Gene expression signature for classification of tissue of origin of tumor samples
US14/320,113 US20140315739A1 (en) 2007-03-27 2014-06-30 Gene expression signature for classification of tissue of origin of tumor samples
US14/746,487 US20150368724A1 (en) 2007-03-27 2015-06-22 Methods and materials for classification of tissue of origin of tumor samples
US15/853,258 US20180127835A1 (en) 2007-03-27 2017-12-22 Gene expression signature for classification of tissue of origin of tumor samples
US15/909,145 US20190032142A1 (en) 2007-03-27 2018-03-01 Methods and materials for classification of tissue of origin of tumor samples
US16/226,406 US20190241966A1 (en) 2007-03-27 2018-12-19 Gene Expression Signature for Classification of Tissue of Origin of Tumor Samples

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US14064208P 2008-12-24 2008-12-24
US61/140,642 2008-12-24

Related Parent Applications (4)

Application Number Title Priority Date Filing Date
US12/532,940 Continuation-In-Part US20100178653A1 (en) 2007-03-27 2008-03-20 Gene expression signature for classification of cancers
PCT/IL2008/000396 Continuation-In-Part WO2008117278A2 (fr) 2007-03-27 2008-03-20 Signature d'une expression génique permettant la classification des cancers
US53294009A Continuation-In-Part 2007-03-27 2009-09-24
PCT/IL2011/000849 Continuation-In-Part WO2012070037A2 (fr) 2007-03-27 2011-11-01 Procédés et matériaux pour la classification de tissus originaires d'échantillons tumoraux

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US13/167,489 Continuation-In-Part US8802599B2 (en) 2007-03-27 2011-06-23 Gene expression signature for classification of tissue of origin of tumor samples

Publications (2)

Publication Number Publication Date
WO2010073248A2 true WO2010073248A2 (fr) 2010-07-01
WO2010073248A3 WO2010073248A3 (fr) 2010-09-16

Family

ID=42288202

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IL2009/001212 WO2010073248A2 (fr) 2007-03-27 2009-12-23 Signature d'expression génétique pour la classification de tissu provenant d'échantillons tumoraux

Country Status (2)

Country Link
CN (1) CN102333888B (fr)
WO (1) WO2010073248A2 (fr)

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011154008A1 (fr) * 2010-06-11 2011-12-15 Rigshospitalet Classification de micro-arn de néoplasie folliculaire de la thyroïde
WO2012010584A1 (fr) * 2010-07-20 2012-01-26 Febit Holding Gmbh Ensembles de miarn complexes en tant que nouveaux biomarqueurs pour le cancer gastrique
WO2012070037A2 (fr) 2010-11-22 2012-05-31 Rosetta Genomics Ltd. Procédés et matériaux pour la classification de tissus originaires d'échantillons tumoraux
WO2012089630A1 (fr) 2010-12-30 2012-07-05 Fondazione Istituto Firc Di Oncologia Molecolare (Ifom) Méthode d'identification d'individus asymptomatiques à haut risque ayant un cancer des poumons à un stade précoce grâce à la détection de mirnas dans les fluides biologiques
WO2012145129A3 (fr) * 2011-04-18 2013-02-07 Cornell University Sous-typage moléculaire, pronostic et traitement du cancer de la prostate
US8541170B2 (en) 2008-11-17 2013-09-24 Veracyte, Inc. Methods and compositions of molecular profiling for disease diagnostics
EP2653558A1 (fr) * 2012-04-18 2013-10-23 Roche Diagniostics GmbH Procédé de détection de cibles d'acide nucléique à l'aide d'un classificateur statistique
JP2013544098A (ja) * 2010-11-17 2013-12-12 アシュラジェン インコーポレイテッド 良性甲状腺新生物と悪性甲状腺新生物を区別するためのバイオマーカーとしてのｍｉＲＮＡ
US8669057B2 (en) 2009-05-07 2014-03-11 Veracyte, Inc. Methods and compositions for diagnosis of thyroid conditions
US8831327B2 (en) 2011-08-30 2014-09-09 General Electric Company Systems and methods for tissue classification using attributes of a biomarker enhanced tissue network (BETN)
US20140309123A1 (en) * 2011-03-28 2014-10-16 Rosetta Genomics Ltd. Methods for lung cancer classification
US9096906B2 (en) 2007-03-27 2015-08-04 Rosetta Genomics Ltd. Gene expression signature for classification of tissue of origin of tumor samples
JP6011945B2 (ja) * 2011-05-20 2016-10-25 公一 中城 マイクロｒｎａ又はその発現系を含む組成物
US9495515B1 (en) 2009-12-09 2016-11-15 Veracyte, Inc. Algorithms for disease diagnostics
EP3369818A1 (fr) * 2011-12-22 2018-09-05 InteRNA Technologies B.V. Miarn permettant le traitement du cancer de la tête et du cou
US10114924B2 (en) 2008-11-17 2018-10-30 Veracyte, Inc. Methods for processing or analyzing sample of thyroid tissue
US10422009B2 (en) 2009-03-04 2019-09-24 Genomedx Biosciences Inc. Compositions and methods for classifying thyroid nodule disease
US10446272B2 (en) 2009-12-09 2019-10-15 Veracyte, Inc. Methods and compositions for classification of samples
US10865452B2 (en) 2008-05-28 2020-12-15 Decipher Biosciences, Inc. Systems and methods for expression-based discrimination of distinct clinical disease states in prostate cancer
US11035005B2 (en) 2012-08-16 2021-06-15 Decipher Biosciences, Inc. Cancer diagnostics using biomarkers
US11078542B2 (en) 2017-05-12 2021-08-03 Decipher Biosciences, Inc. Genetic signatures to predict prostate cancer metastasis and identify tumor aggressiveness
US11208697B2 (en) 2017-01-20 2021-12-28 Decipher Biosciences, Inc. Molecular subtyping, prognosis, and treatment of bladder cancer
US11217329B1 (en) 2017-06-23 2022-01-04 Veracyte, Inc. Methods and systems for determining biological sample integrity
US11414708B2 (en) 2016-08-24 2022-08-16 Decipher Biosciences, Inc. Use of genomic signatures to predict responsiveness of patients with prostate cancer to post-operative radiation therapy
US11639527B2 (en) 2014-11-05 2023-05-02 Veracyte, Inc. Methods for nucleic acid sequencing
US11873532B2 (en) 2017-03-09 2024-01-16 Decipher Biosciences, Inc. Subtyping prostate cancer to predict response to hormone therapy
US11976329B2 (en) 2013-03-15 2024-05-07 Veracyte, Inc. Methods and systems for detecting usual interstitial pneumonia
US12270080B2 (en) 2010-11-19 2025-04-08 The Regents Of The University Of Michigan NcRNA and uses thereof

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP7431760B2 (ja) * 2018-06-30 2024-02-15 ２０／２０ ジェネシステムズ，インク 癌分類子モデル、機械学習システム、および使用方法
CN110706749B (zh) * 2019-09-10 2022-06-10 至本医疗科技（上海）有限公司 一种基于组织器官分化层次关系的癌症类型预测系统和方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050059005A1 (en) * 2001-09-28 2005-03-17 Thomas Tuschl Microrna molecules
US20060084056A1 (en) * 2002-02-13 2006-04-20 Nadia Harbeck Methods for selecting treatment regimens and predicting outcomes in cancer patients
US20070065844A1 (en) * 2005-06-08 2007-03-22 Massachusetts Institute Of Technology Solution-based methods for RNA expression profiling
US20070083945A1 (en) * 2000-03-10 2007-04-12 Byrum Joseph R Nucleic acid molecules and other molecules associated with plants
US20080076674A1 (en) * 2006-07-06 2008-03-27 Thomas Litman Novel oligonucleotide compositions and probe sequences useful for detection and analysis of non coding RNAs associated with cancer
US20080305962A1 (en) * 2005-07-29 2008-12-11 Ralph Markus Wirtz Methods and Kits for the Prediction of Therapeutic Success, Recurrence Free and Overall Survival in Cancer Therapies

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070083945A1 (en) * 2000-03-10 2007-04-12 Byrum Joseph R Nucleic acid molecules and other molecules associated with plants
US20050059005A1 (en) * 2001-09-28 2005-03-17 Thomas Tuschl Microrna molecules
US20060084056A1 (en) * 2002-02-13 2006-04-20 Nadia Harbeck Methods for selecting treatment regimens and predicting outcomes in cancer patients
US20070065844A1 (en) * 2005-06-08 2007-03-22 Massachusetts Institute Of Technology Solution-based methods for RNA expression profiling
US20080305962A1 (en) * 2005-07-29 2008-12-11 Ralph Markus Wirtz Methods and Kits for the Prediction of Therapeutic Success, Recurrence Free and Overall Survival in Cancer Therapies
US20080076674A1 (en) * 2006-07-06 2008-03-27 Thomas Litman Novel oligonucleotide compositions and probe sequences useful for detection and analysis of non coding RNAs associated with cancer

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
SCHULTZ ET AL.: 'MicroRNA let-7b targets important cell cycle molecules in malignant melanoma cells and interferes with anchorageindependent growth.' CELL RES. vol. 18, no. 5, May 2008, pages 549 - 57 *

Cited By (42)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9096906B2 (en) 2007-03-27 2015-08-04 Rosetta Genomics Ltd. Gene expression signature for classification of tissue of origin of tumor samples
US10865452B2 (en) 2008-05-28 2020-12-15 Decipher Biosciences, Inc. Systems and methods for expression-based discrimination of distinct clinical disease states in prostate cancer
US10672504B2 (en) 2008-11-17 2020-06-02 Veracyte, Inc. Algorithms for disease diagnostics
US10236078B2 (en) 2008-11-17 2019-03-19 Veracyte, Inc. Methods for processing or analyzing a sample of thyroid tissue
US10114924B2 (en) 2008-11-17 2018-10-30 Veracyte, Inc. Methods for processing or analyzing sample of thyroid tissue
US8541170B2 (en) 2008-11-17 2013-09-24 Veracyte, Inc. Methods and compositions of molecular profiling for disease diagnostics
US10422009B2 (en) 2009-03-04 2019-09-24 Genomedx Biosciences Inc. Compositions and methods for classifying thyroid nodule disease
US10934587B2 (en) 2009-05-07 2021-03-02 Veracyte, Inc. Methods and compositions for diagnosis of thyroid conditions
US8669057B2 (en) 2009-05-07 2014-03-11 Veracyte, Inc. Methods and compositions for diagnosis of thyroid conditions
US12110554B2 (en) 2009-05-07 2024-10-08 Veracyte, Inc. Methods for classification of tissue samples as positive or negative for cancer
US9495515B1 (en) 2009-12-09 2016-11-15 Veracyte, Inc. Algorithms for disease diagnostics
US10731223B2 (en) 2009-12-09 2020-08-04 Veracyte, Inc. Algorithms for disease diagnostics
US10446272B2 (en) 2009-12-09 2019-10-15 Veracyte, Inc. Methods and compositions for classification of samples
US9856537B2 (en) 2009-12-09 2018-01-02 Veracyte, Inc. Algorithms for disease diagnostics
WO2011154008A1 (fr) * 2010-06-11 2011-12-15 Rigshospitalet Classification de micro-arn de néoplasie folliculaire de la thyroïde
WO2012010584A1 (fr) * 2010-07-20 2012-01-26 Febit Holding Gmbh Ensembles de miarn complexes en tant que nouveaux biomarqueurs pour le cancer gastrique
US10150999B2 (en) 2010-11-17 2018-12-11 Interpace Diagnostics, Llc miRNAs as biomarkers for distinguishing benign from malignant thyroid neoplasms
US11118231B2 (en) 2010-11-17 2021-09-14 Interpace Diagnostics, Llc miRNAs as biomarkers for distinguishing benign from malignant thyroid neoplasms
EP2772550B1 (fr) 2010-11-17 2017-03-29 Interpace Diagnostics, LLC Micro-ARN comme biomarqueurs pour différencier des néoplasmes de thyroïde bénins et malins
JP2013544098A (ja) * 2010-11-17 2013-12-12 アシュラジェン インコーポレイテッド 良性甲状腺新生物と悪性甲状腺新生物を区別するためのバイオマーカーとしてのｍｉＲＮＡ
US12270080B2 (en) 2010-11-19 2025-04-08 The Regents Of The University Of Michigan NcRNA and uses thereof
EP2643479A2 (fr) * 2010-11-22 2013-10-02 Rosetta Genomics Ltd Procédés et matériaux pour la classification de tissus originaires d'échantillons tumoraux
EP2643479A4 (fr) * 2010-11-22 2014-08-27 Rosetta Genomics Ltd Procédés et matériaux pour la classification de tissus originaires d'échantillons tumoraux
WO2012070037A2 (fr) 2010-11-22 2012-05-31 Rosetta Genomics Ltd. Procédés et matériaux pour la classification de tissus originaires d'échantillons tumoraux
WO2012089630A1 (fr) 2010-12-30 2012-07-05 Fondazione Istituto Firc Di Oncologia Molecolare (Ifom) Méthode d'identification d'individus asymptomatiques à haut risque ayant un cancer des poumons à un stade précoce grâce à la détection de mirnas dans les fluides biologiques
US9914972B2 (en) * 2011-03-28 2018-03-13 Rosetta Genomics Ltd. Methods for lung cancer classification
US20140309123A1 (en) * 2011-03-28 2014-10-16 Rosetta Genomics Ltd. Methods for lung cancer classification
EP2505663A1 (fr) 2011-03-30 2012-10-03 IFOM Fondazione Istituto Firc di Oncologia Molecolare Procédé pour identifier des individus asymptomatiques à haut risque touchés par un cancer du poumon à l'état précoce au moyen de la détection d'ARNmi dans les liquides corporels
WO2012145129A3 (fr) * 2011-04-18 2013-02-07 Cornell University Sous-typage moléculaire, pronostic et traitement du cancer de la prostate
JP6011945B2 (ja) * 2011-05-20 2016-10-25 公一 中城 マイクロｒｎａ又はその発現系を含む組成物
US8831327B2 (en) 2011-08-30 2014-09-09 General Electric Company Systems and methods for tissue classification using attributes of a biomarker enhanced tissue network (BETN)
EP3369818A1 (fr) * 2011-12-22 2018-09-05 InteRNA Technologies B.V. Miarn permettant le traitement du cancer de la tête et du cou
EP2653558A1 (fr) * 2012-04-18 2013-10-23 Roche Diagniostics GmbH Procédé de détection de cibles d'acide nucléique à l'aide d'un classificateur statistique
US9399794B2 (en) 2012-04-18 2016-07-26 Roche Molecular Systems, Inc. Method of detecting nucleic acid targets using a statistical classifier
US11035005B2 (en) 2012-08-16 2021-06-15 Decipher Biosciences, Inc. Cancer diagnostics using biomarkers
US11976329B2 (en) 2013-03-15 2024-05-07 Veracyte, Inc. Methods and systems for detecting usual interstitial pneumonia
US11639527B2 (en) 2014-11-05 2023-05-02 Veracyte, Inc. Methods for nucleic acid sequencing
US11414708B2 (en) 2016-08-24 2022-08-16 Decipher Biosciences, Inc. Use of genomic signatures to predict responsiveness of patients with prostate cancer to post-operative radiation therapy
US11208697B2 (en) 2017-01-20 2021-12-28 Decipher Biosciences, Inc. Molecular subtyping, prognosis, and treatment of bladder cancer
US11873532B2 (en) 2017-03-09 2024-01-16 Decipher Biosciences, Inc. Subtyping prostate cancer to predict response to hormone therapy
US11078542B2 (en) 2017-05-12 2021-08-03 Decipher Biosciences, Inc. Genetic signatures to predict prostate cancer metastasis and identify tumor aggressiveness
US11217329B1 (en) 2017-06-23 2022-01-04 Veracyte, Inc. Methods and systems for determining biological sample integrity

Also Published As

Publication number Publication date
WO2010073248A3 (fr) 2010-09-16
CN102333888B (zh) 2013-07-10
CN102333888A (zh) 2012-01-25

Similar Documents

Publication Publication Date Title
US20190241966A1 (en) Gene Expression Signature for Classification of Tissue of Origin of Tumor Samples
WO2010073248A2 (fr) Signature d'expression génétique pour la classification de tissu provenant d'échantillons tumoraux
US9803247B2 (en) MicroRNAs expression signature for determination of tumors origin
WO2008117278A2 (fr) Signature d'une expression génique permettant la classification des cancers
US9096906B2 (en) Gene expression signature for classification of tissue of origin of tumor samples
US9133522B2 (en) Compositions and methods for the diagnosis and prognosis of mesothelioma
WO2009153775A2 (fr) Procédés permettant de différencier différents types de cancers du poumon
EP2643479A2 (fr) Procédés et matériaux pour la classification de tissus originaires d'échantillons tumoraux
US9914972B2 (en) Methods for lung cancer classification
US9068232B2 (en) Gene expression signature for classification of kidney tumors
US9834821B2 (en) Diagnosis and prognosis of various types of cancers
WO2010004562A2 (fr) Procédés et compositions permettant de détecter un cancer colorectal
WO2009066291A2 (fr) Signature d'expression de micro-arn pour la détermination de l'origine de tumeurs
US9340823B2 (en) Gene expression signature for classification of kidney tumors
WO2011039757A2 (fr) Compositions et méthodes de pronostic du cancer du rein

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200980157378.X

Country of ref document: CN

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

Ref document number: 09834226

Country of ref document: EP

Kind code of ref document: A2

WWE Wipo information: entry into national phase

Ref document number: 212979

Country of ref document: IL

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09834226

Country of ref document: EP

Kind code of ref document: A2

点击 这是indexloc提供的php浏览器服务，不要输入任何密码和下载