Methods and materials for assessing loss of heterozygosity

Abstract

To provide methods for assessing LOH in a cancer cell or genomic DNA thereof.SOLUTION: Provided are methods and materials involved in assessing samples (e.g. cancer cells) for the presence of a loss of heterozygosity (LOH) signature. For example, provided are methods and materials for determining whether or not a cell (e.g., a cancer cell) contains an LOH signature. Also provided are materials and methods for identifying cells (e.g., cancer cells) having a deficiency in homology directed repair (HDR) as well as materials and methods for identifying cancer patients likely to respond to a particular cancer treatment regimen.SELECTED DRAWING: None

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Patent Application No. 61 / 578,713, filed December 21, 2011, and U.S. Provisional Patent Application No. 61 / 654,402, filed June 1, 2012, the entire contents of which are incorporated herein by reference.

[0002] 1.Technical Field This document relates to methods and materials involved in evaluating a sample (e.g., cancer cells) for the presence of a loss of heterozygosity (LOH) signature. For example, this document provides methods and materials for determining whether a cell (e.g., cancer cell) contains an LOH signature. This document also provides materials and methods for identifying cells (e.g., cancer cells) that have defects in homology directed repair (HDR), as well as materials and methods for identifying cancer patients who are likely to respond to a particular cancer treatment regimen. Throughout this document, HDR deficiency and HRD (homology repair deficiency) are used interchangeably unless otherwise specified. [Background technology]

[0003] 2. Background information Cancer is a major public health problem, with 562,340 cancer deaths in the United States in 2009 alone. American Cancer Society, Cancer Facts & Figures 2009 (available on the American Cancer Society website). One of the most important challenges in cancer treatment is to discover relevant and clinically useful features of a patient's own cancer, and then tailor a treatment plan that best suits the patient's cancer based on these features. Although progress has been made in this field of personalized medicine, there is still a great need for better molecular diagnostic tools to characterize a patient's cancer. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] American Cancer Society, Cancer Facts & Figures 2009 Summary of the Invention

[0005] overview In general, one aspect of the present invention features a method for evaluating LOH in cancer cells or the genomic DNA of said cancer cells.In some embodiments, said method comprises or essentially consists of the steps of: (a) detecting LOH regions in at least one pair of human chromosomes (e.g., any human chromosome pair other than human X / Y sex chromosome pair) from cancer cells in cancer cells or in the genomic DNA obtained from said cancer cells; and (b) determining the number and size (e.g., length) of said LOH regions.In some embodiments, LOH regions are analyzed in a large number of chromosome pairs representative of the whole genome (e.g., sufficient chromosomes are analyzed so that the number and size of LOH regions are expected to be representative of the number and size of LOH regions throughout the genome). In some embodiments, the method further comprises determining the total number of LOH regions (index LOH regions) that are longer than about 1.5 megabases, 5 megabases, 12 megabases, 13 megabases, 14 megabases, 15 megabases, 16 megabases, 17 megabases or more (preferably 14 megabases, 15 megabases, 16 megabases or more, more preferably 15 megabases or more) but shorter than the full length of each chromosome in which the LOH regions are located.Alternatively or additionally, the total length of such index LOH regions is determined.In some particular embodiments, when the total number of index LOH regions or the total length of index LOH regions is equal to or greater than a predetermined reference number, the cancer cell or genomic DNA or the patient having the cancer cell or genomic DNA is identified as having HDR-deficient LOH signature.

[0006] Also provided is an alternative method for evaluating LOH in cancer cell or its genomic DNA, comprising or consisting essentially of the steps of: (a) detecting LOH regions in at least a pair of human chromosomes from cancer cell or genomic DNA obtained from cancer cell, wherein at least a pair of human chromosomes is not a human X / Y sex chromosome pair; and (b) determining the total number of LOH regions and / or the total length of LOH regions in at least a pair of human chromosomes, wherein LOH regions are longer than a first length but shorter than the length of the entire chromosome containing LOH regions, and the first length is about 1.5 megabases or more (or 5 megabases, 10 megabases, 13 megabases, 14 megabases, 15 megabases, 16 megabases or more, preferably 15 megabases or more).In some particular embodiments, when the total number or the total length is equal to a predetermined reference number or is greater than a predetermined reference number, said cancer cell or said genomic DNA or the patient having said cancer cell or said genomic DNA is identified as having HDR-deficient LOH signature.

[0007] In another aspect, the present invention provides a method for predicting the status of BRCA1 gene and BRCA2 gene in cancer cell.The method comprises or consists essentially of the steps of: in cancer cell, determining the total number and / or total length of LOH region in at least a pair of human chromosomes from cancer cell, wherein LOH region is longer than a first length but shorter than the length of the whole chromosome that contains one LOH region, and at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more (or 5 megabases, 10 megabases or more, preferably about 15 megabases or more); and correlating the total number or total length that exceeds the reference number with the increased probability of defect in BRCA1 gene or BRCA2 gene.

[0008] In another aspect, the present invention provides a method for predicting the HDR status in cancer cell.The method comprises or essentially consists of the steps of: in cancer cell, determining the total number and / or total length of LOH region in at least a pair of human chromosomes from cancer cell, wherein LOH region is longer than a first length but shorter than the total length of the chromosome that contains one LOH region, and at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more (or 5 megabases, 10 megabases or more, preferably about 15 megabases or more); and correlating the total number or total length that exceeds the reference number with the increased probability of loss in HDR.

[0009] In another aspect, the present invention provides a method for predicting the response of cancer patients to cancer treatment regimens comprising DNA damage agents, anthracyclines, topoisomerase I inhibitors, radiation, and / or PARP inhibitors.The method comprises or consists essentially of the steps of determining the number and / or total length of LOH regions in at least a pair of human chromosomes from cancer cells of cancer patients, wherein LOH regions are longer than a first length but shorter than the total length of the chromosome that contains one LOH region, and the at least a pair of human chromosomes are not human X / Y sex chromosome pairs, and the first length is about 1.5 megabases or more (or 5 megabases, 10 megabases, or more, preferably about 15 megabases or more); and correlating the total number or total length that exceeds the reference number with the increased probability that the cancer patient can respond to the cancer treatment regimen.In some embodiments, the patient is an untreated patient.

[0010] In another aspect, the present invention relates to a method for predicting the response of cancer patients to a therapeutic regimen, the method comprises or essentially consists of the steps of: determining in a cancer cell derived from a cancer patient the total number and / or total length of LOH regions in at least a pair of human chromosomes from the cancer cell of the cancer patient, the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes are not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more (or 5 megabases, 10 megabases or more, preferably about 15 megabases or more); and correlating the total number or total length that is greater than a reference number with the increased probability that the cancer patient cannot respond to a therapeutic regimen that comprises paclitaxel or docetaxel.

[0011] In another aspect, the present invention relates to a method for treating cancer, which comprises or essentially consists of the steps of: (a) determining the total number and / or total length of LOH regions in at least a pair of human chromosomes from cancer cells from cancer patients or genomic DNA obtained from said cancer cells, wherein LOH regions are longer than a first length but shorter than the total length of the chromosomes containing LOH regions, and at least a pair of human chromosomes are not human X / Y sex chromosome pairs, and the first length is about 1.5 megabases or more (or 5 megabases, 10 megabases or more, preferably about 15 megabases or more); and (b) administering to cancer patients a cancer treatment regimen comprising one or more drugs selected from the group consisting of DNA damage agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors when the total number or total length of LOH regions is greater than a reference number.In some embodiments, the patient is an untreated patient.

[0012] In some embodiments of any one or more of the methods described in the preceding six paragraphs, any one or more of the following may be applied as appropriate. At least 2, 5, 10, or 21 pairs of human chromosomes can be determined as LOH regions. The cancer cell can be an ovarian cancer cell, a breast cancer cell, or an esophageal cancer cell. The first length can be about 6 megabases, 12 megabases, or about 15 megabases, or more. The reference number can be 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, or 20, or more. The at least one pair of human chromosomes can exclude human chromosome 17. The DNA damaging agent may be cisplatin, carboplatin, oxalaplatin, or picoplatin, the anthracycline may be epirubincin or doxorubicin, the topoisomerase I inhibitor may be campothecin, topotecan, or irinotecan, and the PARP inhibitor may be iniparib, olaparib, or velapirib.

[0013] In another aspect, the invention features the use of one or more drugs selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors in the manufacture of a medicament useful for treating cancer in a patient identified as having cancer cells determined to have a total of 5, 8, 9, 10, 12, 15, 17, 20, or more index LOH regions. Index LOH regions can be determined in at least 2, 5, 10, or 21 pairs of human chromosomes. The cancer cells can be ovarian, breast, or esophageal cancer cells. The length of the index LOH region can be about 6, 12, or 15 megabases, or more. The index LOH region can be present in a chromosome other than human chromosome 17. The DNA damaging agent may be a platinum-based chemotherapy drug, the anthracycline may be epirubicin or doxorubicin, the topoisomerase I inhibitor may be campothecin, topotecan, or irinotecan, and the PARP inhibitor may be iniparib, olaparib, or verapirib. In some embodiments, the patient is a treatment-naïve patient.

[0014] In another aspect, the present invention features the use of a plurality of oligonucleotides capable of hybridizing to a plurality of polymorphic regions of human genomic DNA in the manufacture of a diagnostic kit useful for determining the total number or total length of indicator LOH regions in at least one chromosome pair from a human cancer cell obtained from a cancer patient, and useful for detecting (a) an increased likelihood of a defect in the BRCA1 gene or BRCA2 gene in the cancer cell, (b) an increased likelihood of a defect in HDR in the cancer cell, or (c) an increased likelihood that a cancer patient may respond to a cancer treatment regimen that includes a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, or a PARP inhibitor. The indicator LOH regions can be determined in at least 2, 5, 10, or 21 pairs of human chromosomes. The cancer cell may be an ovarian cancer cell, a breast cancer cell, or an esophageal cancer cell. The length of the indicator LOH region may be about 6 megabases, 12 megabases, or 15 megabases, or more. The indicator LOH region may be present in a chromosome other than human chromosome 17.

[0015] In another aspect, the present invention features a system for determining the LOH status of cancer cells in a cancer patient. The system comprises or essentially consists of: (a) a sample analyzer configured to generate a plurality of signals for genomic DNA of at least a pair of human chromosomes from the cancer cell; and (b) a computer subsystem programmed to calculate the number or length of indicator LOH regions in at least a pair of human chromosomes based on the plurality of signals. The computer subsystem can be programmed to compare the number or length of indicator LOH regions with a reference number to determine (a) the likelihood of defects in BRCA1 and / or BRCA2 genes in the cancer cell, (b) the likelihood of defects in HDR in the cancer cell, or (c) the likelihood that the cancer patient may respond to a cancer treatment regimen that includes a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, or a PARP inhibitor. The system may comprise an output module configured to display the likelihood of (a), (b), or (c). The system may comprise an output module configured to display a recommendation for the use of the cancer treatment regimen. The index LOH region can be determined in at least 2, 5, 10, or 21 pairs of human chromosomes. The cancer cells can be ovarian, breast, or esophageal cancer cells. The index LOH region can be about 6, 12, or 15 megabases long, or longer. The index LOH region can be present in a chromosome other than human chromosome 17. The DNA damaging agent can be a platinum-based chemotherapy drug, the anthracycline can be epirubicin or doxorubicin, the topoisomerase I inhibitor can be campothecin, topotecan, or irinotecan, and the PARP inhibitor can be iniparib, olaparib, or verapirib.

[0016] In another aspect, the present invention provides a computer program product embodied in a computer readable medium, which, when executed on a computer, provides instructions for detecting the presence or absence of any LOH regions along one or more human chromosomes other than human X and Y sex chromosomes, where the length of the LOH region is about 1.5 megabases or more (or 5 megabases, 10 megabases, or more, preferably 15 megabases or more), but less than the length of the entire chromosome containing the LOH region; and instructions for determining the total number or total length of LOH regions in one or more chromosome pairs. The computer program product may comprise other instructions. The indicator LOH regions can be determined in at least 2 pairs, 5 pairs, 10 pairs, or 21 pairs of human chromosomes. The cancer cells may be ovarian cancer cells, breast cancer cells, or esophageal cancer cells. The length of the indicator LOH region may be about 6 megabases, 12 megabases, or 15 megabases, or more. The indicator LOH region may be present in a chromosome other than human chromosome 17. The DNA damaging agent may be a platinum-based chemotherapy drug, the anthracycline may be epirubicin or doxorubicin, the topoisomerase I inhibitor may be campothecin, topotecan, or irinotecan, and the PARP inhibitor may be iniparib, olaparib, or velapirib.

[0017] In another aspect, the present invention provides a diagnostic kit.The kit comprises or essentially consists of at least 500 oligonucleotides that can hybridize to multiple polymorphic regions of human genomic DNA; and the computer program product provided herein.The computer program product can be incorporated in a computer readable medium, and when executed on a computer, it provides instructions for detecting the presence or absence of any LOH region along one or more human chromosomes other than human X and Y sex chromosomes, and the length of the LOH region is about 1.5 megabases or more (or 5 megabases or 10 megabases or more, preferably about 15 megabases or more), but less than the length of the entire chromosome that contains the LOH region; and instructions for determining the total number and / or total length of the LOH region in one or more chromosome pairs.

[0018] In another aspect, this document features a method for evaluating the presence of LOH signature in cancer cells of a patient.The method comprises or consists essentially of the steps of: (a) detecting the presence of LOH regions in at least one pair of human chromosomes from cancer cells of a cancer patient that are greater than a reference number, the LOH regions being longer than a first length but shorter than the length of the entire chromosome that contains the LOH regions, the at least one pair of human chromosomes being not a human X / Y sex chromosome pair, and the first length being about 1.5 megabases or more; and (b) identifying the patient as having cancer cells with LOH signature.

[0019] In another aspect, this document features a method for evaluating a patient's cancer cells for the presence of HDR-deficient condition.The method comprises or consists essentially of the steps of: (a) detecting the presence of LOH regions in at least a pair of human chromosomes from a cancer patient's cancer cells that are greater than a reference number, wherein the LOH regions are greater than a first length but less than the length of the entire chromosome containing the LOH regions, and the at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) identifying the patient as having cancer cells with HDR-deficient condition.

[0020] In another aspect, this document features a method for evaluating the presence of genetic mutations in genes from HDR pathways in cancer cells of a patient.The method comprises or consists essentially of the steps of: (a) detecting the presence of LOH regions in at least a pair of human chromosomes from cancer cells of a cancer patient that are greater than a reference number, wherein the LOH regions are greater than a first length but less than the length of the entire chromosome that contains the LOH regions, and the at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) identifying the patient as having cancer cells with genetic mutations.

[0021] In another aspect, this document features a method for determining whether a patient is likely to respond to a cancer treatment regimen, comprising administering a drug selected from the group consisting of radiation or a DNA damaging agent, anthracycline, a topoisomerase I inhibitor, and a PARP inhibitor.The method comprises or consists essentially of: (a) detecting the presence of LOH regions in at least one pair of human chromosomes from a cancer cell of a cancer patient that are greater than a reference number, wherein the LOH regions are greater than a first length but less than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) identifying the patient as likely to respond to the cancer treatment regimen.

[0022] In another aspect, this document features a method for evaluating a patient, the method includes or consists essentially of: (a) determining that the patient contains cancer cells with LOH signature, and the presence of LOH regions in at least one pair of human chromosomes from the cancer patient's cancer cells more than a reference number indicates that the cancer cells contain LOH signature, the LOH region is longer than a first length but shorter than the length of the entire chromosome containing the LOH region, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) diagnosing the patient as having cancer cells with LOH signature.

[0023] In another aspect, this document features a method for evaluating a patient, the method comprising or consisting essentially of: (a) determining that the patient contains cancer cells having an HDR deficient state, where the presence of more LOH regions than a reference number in at least one pair of human chromosomes from the cancer patient's cancer cells indicates that the cancer cells have an HDR deficient state, the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more, and (b) diagnosing the patient as having cancer cells having an HDR deficient state.

[0024] In another aspect, this document features a method for evaluating a patient, the method comprising or consisting essentially of: (a) determining that the patient contains cancer cells having a genetic mutation in a gene from a HDR pathway, wherein the presence of more than a reference number of LOH regions in at least one pair of human chromosomes from the cancer patient's cancer cells indicates that the cancer cells have the genetic mutation, the LOH region is longer than a first length but shorter than the length of the entire chromosome containing the LOH region, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more, and (b) diagnosing the patient as having cancer cells having the genetic mutation.

[0025] In another aspect, this document features a method for evaluating a patient for the likelihood of responding to a cancer treatment regimen, comprising administering a drug selected from the group consisting of radiation or DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors.The method comprises or essentially consists of: (a) determining that the patient comprises cancer cells having a LOH signature, wherein the presence of LOH regions in at least one pair of human chromosomes from the cancer patient's cancer cells more than a reference number indicates that the cancer cells have a LOH signature, the LOH region is longer than a first length but shorter than the length of the entire chromosome containing the LOH region, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) diagnosing the patient as likely to respond to the cancer treatment regimen based at least in part on the presence of the LOH signature.

[0026] In another aspect, this document features a method for performing a diagnostic analysis of cancer cells of a patient, the method comprising or consisting essentially of: (a) detecting the presence of LOH regions in at least a pair of human chromosomes from cancer cells that are greater than a reference number, the LOH regions being greater than a first length but less than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes being not a human X / Y sex chromosome pair, and the first length being about 1.5 megabases or greater; and (b) identifying the patient as having cancer cells with the LOH signature.

[0027] In another aspect, this document features a method for performing a diagnostic analysis of cancer cells of a patient, the method comprising or consisting essentially of: (a) detecting the presence of LOH regions in at least a pair of human chromosomes from cancer cells that are greater than a reference number, the LOH regions being greater than a first length but less than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes being not a human X / Y sex chromosome pair, and the first length being about 1.5 megabases or greater; and (b) identifying the patient as having cancer cells with an HDR-deficient condition.

[0028] In another aspect, this document features a method for performing diagnostic analysis of cancer cells of a patient, the method comprising or consisting essentially of: (a) detecting the presence of LOH regions in at least a pair of human chromosomes from a cancer cell that are greater than a reference number, the LOH regions being greater than a first length but less than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes being not a human X / Y sex chromosome pair, and the first length being about 1.5 megabases or greater; and (b) identifying the patient as having cancer cells that have genetic mutations in genes from HDR pathways.

[0029] In another aspect, this document features a method for performing diagnostic analysis of a patient's cancer cells to determine whether the cancer patient is likely to respond to a cancer treatment regimen, comprising administering a drug selected from the group consisting of radiation or a DNA damaging agent, anthracycline, a topoisomerase I inhibitor, and a PARP inhibitor. The method comprises or consists essentially of: (a) detecting the presence of more than a reference number of LOH regions in at least a pair of human chromosomes from the cancer cells, where the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, where the at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and where the first length is about 1.5 megabases or more; and (b) identifying the patient as likely to respond to the cancer treatment regimen.

[0030] In another aspect, this document features a method for diagnosing a patient as having cancer cells with LOH signature.The method comprises or essentially consists of the steps of: (a) determining that a patient has cancer cells with LOH signature, and the presence of LOH regions in at least one pair of human chromosomes from the cancer patient's cancer cells more than a reference number indicates that the cancer cells have LOH signature, the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) diagnosing the patient as having cancer cells with LOH signature.

[0031] In another aspect, this document features a method for diagnosing a patient as having cancer cells with HDR-deficient condition, the method comprises or consists essentially of the steps of: (a) determining that the patient has cancer cells with HDR-deficient condition, wherein the presence of more LOH regions than a reference number in at least one pair of human chromosomes from the cancer patient's cancer cells indicates that the cancer cells have HDR-deficient condition, the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more, and (b) diagnosing the patient as having cancer cells with HDR-deficient condition.

[0032] In another aspect, this document features a method for diagnosing a patient as having cancer cells with genetic mutations in genes from HDR pathways.The method includes or consists essentially of the steps of: (a) determining that the patient has cancer cells with genetic mutations, wherein the presence of LOH regions in at least one pair of human chromosomes from the cancer patient's cancer cells more than a reference number indicates that the cancer cells have the genetic mutation, the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) diagnosing the patient as having cancer cells with genetic mutations.

[0033] In another aspect, this document features a method for diagnosing a patient as a candidate for a cancer treatment regimen, comprising administering radiation or a drug selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors.The method comprises or essentially consists of: (a) determining that the patient comprises cancer cells having an LOH signature, wherein the presence of more than a reference number of LOH regions in at least one pair of human chromosomes from the cancer patient's cancer cells indicates that the cancer cells have an LOH signature, the LOH region is longer than a first length but shorter than the length of the entire chromosome containing the LOH region, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) diagnosing the patient as likely to respond to the cancer treatment regimen based at least in part on the presence of the LOH signature.

[0034] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice of the present invention, suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. In addition, the materials, methods, and examples are illustrative only and are not intended to be limiting.

[0035] [The present invention 1001] A method for assessing LOH in a cancer cell or genomic DNA of said cancer cell, comprising the steps of: (a) detecting in a cancer cell or in genomic DNA obtained from the cancer cell, a region of LOH in at least one pair of human chromosomes from the cancer cell, wherein the at least one pair of human chromosomes is not a human X / Y sex chromosome pair; and (b) determining the total number of LOH regions in the at least one pair of human chromosomes, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, and wherein the first length is about 1.5 megabases or more. [The present invention 1002] A method for predicting the status of BRCA1 and BRCA2 genes in cancer cells, comprising the steps of: determining, in said cancer cell, a total number of LOH regions in at least a pair of human chromosomes from said cancer cell, wherein said LOH regions are longer than a first length but shorter than the length of the entire chromosome containing said LOH regions, said at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and said first length is about 1.5 megabases or more; and Correlating said total number above a reference number with an increased likelihood of a defect in the BRCA1 or BRCA2 gene. [The present invention 1003] A method for predicting HDR status in cancer cells, comprising the steps of: determining, in said cancer cell, a total number of LOH regions in at least a pair of human chromosomes from said cancer cell, wherein said LOH regions are longer than a first length but shorter than the length of the entire chromosome containing said LOH regions, said at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and said first length is about 1.5 megabases or more; and Correlating said total number above a reference number with an increased likelihood of a defect in the HDR. [The present invention 1004] 1. A method for predicting a cancer patient's response to a cancer treatment regimen comprising a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, and / or a PARP inhibitor, comprising the steps of: determining, in cancer cells derived from the cancer patient, the number of LOH regions in at least one pair of human chromosomes from the cancer cells of the cancer patient, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and Correlating a total number above the reference number with an increased likelihood that the cancer patient will respond to the cancer treatment regimen. [The present invention 1005] 1. A method for predicting a cancer patient's response to a treatment regimen, comprising the steps of: determining, in cancer cells derived from the cancer patient, a total number of LOH regions in at least one pair of human chromosomes from the cancer cells of the cancer patient, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and Correlating said total number above a reference number with an increased likelihood that said cancer patient will not respond to a treatment regimen including paclitaxel or docetaxel. [The present invention 1006] 1. A method of treating cancer, comprising the steps of: (a) determining in a cancer cell derived from a cancer patient or in genomic DNA obtained from the cancer cell a total number of LOH regions in at least a pair of human chromosomes from the cancer cell, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) administering to the cancer patient a cancer treatment regimen comprising one or more drugs selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors if the total number of LOH regions exceeds the reference number. [The present invention 1007] The use of one or more drugs selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors to manufacture a medicament useful for treating cancer in a patient identified as having cancer cells determined to have a total of five or more index LOH regions. [The present invention 1008] (a) a sample analyzer configured to generate a plurality of signals for genomic DNA of at least one pair of human chromosomes from a cancer cell; and (b) a computer subsystem programmed to calculate the number of indicator LOH regions in the at least one pair of human chromosomes based on the plurality of signals. A system for determining the LOH status of cancer cells of a cancer patient, comprising: [The present invention 1009] the computer subsystem compares the number of index LOH regions to a reference number; (a) a possible defect in the BRCA1 and / or BRCA2 genes in the cancer cells; (b) a possible defect in HDR in the cancer cell; or (c) the likelihood that the cancer patient will respond to a cancer treatment regimen that includes a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, or a PARP inhibitor. The system of the present invention 1008 is programmed to determine [The present invention 1010] When running on a computer, detecting the presence or absence of any LOH regions along one or more of the human chromosomes other than the human X and Y sex chromosomes, wherein the length of the LOH region is about 1.5 megabases or more, but less than the length of the entire chromosome containing the LOH region; and determining the total number of said LOH regions in one or more chromosome pairs A computer program product embodied in a computer-readable medium, the computer program product performing steps including: [The present invention 1011] at least 500 oligonucleotides capable of hybridizing to multiple polymorphic regions of human genomic DNA; and Computer program product of the present invention 1010 A diagnostic kit comprising: [The present invention 1012] It is useful for determining the total number of index LOH regions in at least one chromosome pair from a human cancer cell obtained from a cancer patient, and (a) an increased likelihood of a defect in the BRCA1 or BRCA2 gene in the cancer cells; (b) an increased likelihood of a defect in HDR in the cancer cell; or (c) increasing the likelihood that the cancer patient will respond to a cancer treatment regimen that includes a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, or a PARP inhibitor. The use of a plurality of oligonucleotides capable of hybridizing to a plurality of polymorphic regions of human genomic DNA to prepare a diagnostic kit which is useful for detecting [The present invention 1013] The method according to any one of claims 1001 to 1006, wherein said LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes. [The present invention 1014] The method according to any one of claims 1001 to 1006, wherein said cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [The present invention 1015] The method according to any one of claims 1001 to 1006, wherein the total number of LOH regions is 9, 15, 20 or more. [The present invention 1016] The method of any one of claims 1001 to 1006, wherein said first length is about 6 megabases, 12 megabases, or 15 megabases or more. [The present invention 1017] The method of any one of claims 1001 to 1006, wherein the number of references is 6, 7, 8, 9, 10, 11, 12, or 13 or more. [The present invention 1018] The use of the present invention 1007 or 1012, wherein the indicator LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes. [The present invention 1019] The use of invention 1007 or 1012, wherein said cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [The present invention 1020] The use of the present invention 1007 or 1012, wherein the total number of the index LOH regions is 9, 15, 20 or more. [The present invention 1021] The use of the present invention 1007 or 1012, wherein the length of the index LOH region is about 6 megabases, 12 megabases, or 15 megabases or more. [The present invention 1022] The system of 1008 or 1009, wherein the indicator LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes. [The present invention 1023] The system of any one of claims 1008 to 1009, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [The present invention 1024] The system of the present invention 1008 or 1009, wherein the total number of the index LOH regions is 9, 15, 20, or more. [The present invention 1025] The system of the present invention 1008 or 1009, wherein the length of the index LOH region is about 6 megabases, 12 megabases, or 15 megabases, or more. [The present invention 1026] The computer program product of the present invention 1010, wherein the index LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes. [The present invention 1027] The computer program product of the present invention 1010, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [The present invention 1028] The computer program product of the present invention 1010, wherein the total number of the index LOH regions is 9, 15, 20, or more. [The present invention 1029] The computer program product of the present invention 1010, wherein the length of the index LOH region is about 6 megabases, 12 megabases, or 15 megabases or more. [The present invention 1030] The method of any of claims 1001 to 1006, wherein the at least one pair of human chromosomes is not human chromosome 17. [The present invention 1031] The use of 1007 or 1012 of the present invention, wherein the indicator LOH region is not present in human chromosome 17. [The present invention 1032] The system of the present invention 1008 or 1009, wherein the indicator LOH region is not present in human chromosome 17. [The present invention 1033] The computer program product of the present invention 1010, wherein the indicator LOH region is not present in human chromosome 17. [The present invention 1034] The method of any one of claims 1004 to 1006, wherein the DNA damaging agent is cisplatin, carboplatin, oxalaplatin, or picoplatin, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or velapirib. [The present invention 1035] The use of invention 1012, wherein the DNA damaging agent is a platinum-based chemotherapy drug, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or velapirib. [The present invention 1036] The system of the present invention 1009, wherein the DNA damaging agent is a platinum-based chemotherapy drug, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or velapirib. [The present invention 1037] The computer program product of the present invention 1010, wherein the DNA damaging agent is a platinum-based chemotherapy drug, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or velapirib. [The present invention 1038] (a) detecting, in cancer cells or genomic DNA obtained from the cancer cells, LOH regions in a representative number of human chromosome pairs from the cancer cells; and (b) determining the number and size of the LOH regions. A method comprising: [The present invention 1039] The method of any one of claims 1 to 38, wherein said representative number of human chromosome pairs is representative of the entire genome. [The present invention 1040] The method of any one of claims 10 to 15, further comprising the step of correlating an increase in the number of LOH regions of a particular size with an increased likelihood of deletion in HDR. [The present invention 1041] The method of the present invention 1040, wherein the specified size is greater than about 1.5 megabases, 2 megabases, 2.5 megabases, 3 megabases, 4 megabases, 5 megabases, 6 megabases, 7 megabases, 8 megabases, 9 megabases, 10 megabases, 11 megabases, 12 megabases, 13 megabases, 14 megabases, 15 megabases, 16 megabases, 17 megabases, 18 megabases, 19 megabases, 20 megabases, 25 megabases, 30 megabases, 35 megabases, 40 megabases, 45 megabases, 50 megabases, 75 megabases, or 100 megabases, and is less than the length of the entire chromosome containing the LOH region. [The present invention 1042] The method of any one of claims 1040 to 1041, wherein 6, 7, 8, 9, 10, 11, 12, or 13 or more LOH regions of said particular size are correlated with an increased likelihood of deletion in HDR. [The present invention 1043] A method for determining prognosis in a patient, comprising the steps of: (a) determining whether the patient contains cancer cells having an LOH signature, wherein the presence of more LOH regions than a reference number in at least one pair of human chromosomes from a cancer cell of the cancer patient indicates that the cancer cell has the LOH signature, wherein the LOH region is longer than a first length but shorter than the length of the entire chromosome containing the LOH region, wherein the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and wherein the first length is about 1.5 megabases or more; and (b)(1) determining that the patient has a relatively good prognosis based, at least in part, on the presence of the LOH signature; or (b)(2) determining that the patient has a relatively poor prognosis based, at least in part, on the absence of the LOH signature. [The present invention 1044] 1. A composition comprising a therapeutic agent selected from the group consisting of a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, and a PARP inhibitor for use in treating a disease, a cancer selected from the group consisting of breast cancer, ovarian cancer, liver cancer, esophageal cancer, lung cancer, head and neck cancer, prostate cancer, colon cancer, rectal cancer, colorectal cancer, and pancreatic cancer, in a patient, wherein the patient has more LOH regions than a reference number in at least a pair of human chromosomes from the patient's cancer cells, the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes are not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more. [The present invention 1045] The composition of the present invention 1044, wherein the LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes. [The present invention 1046] The composition of the present invention 1044, wherein the total number of LOH regions is 9, 15, 20 or more. [The present invention 1047] The composition of the present invention 1044, wherein said first length is about 6 megabases, 12 megabases, or 15 megabases or more. [The present invention 1048] The composition of the present invention 1044, wherein said reference number is 6, 7, 8, 9, 10, 11, 12, or 13 or more. [The present invention 1049] 1. A method of treating cancer in a patient, comprising the steps of: determining in a sample derived from said patient a number of LOH regions in at least one pair of human chromosomes from a cancer cell of said cancer patient, said LOH regions being longer than a first length but shorter than the length of the entire chromosome containing said LOH regions, indicating said cancer cell has an LOH signature, said at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and said first length is about 1.5 megabases or more; providing a test value derived from the number of LOH regions; Comparing the test value to one or more reference values ​​(e.g., mean, median, tercile, quartile, quintile, etc.) derived from the number of LOH regions in a reference population; and administering an anti-cancer drug to the patient or recommending or prescribing or initiating a treatment regimen including chemotherapy and / or synthetic lethal agents based at least in part on the comparing step revealing that the test value exceeds at least one of the reference values ​​(e.g., at least 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, or 10-fold the reference value; at least 1-, 2-, 3-, 4-, 5-, 6-, 7-, 8-, 9-fold, or 10-fold greater than the reference value); or recommending, prescribing or initiating a treatment regimen that does not include chemotherapy and / or synthetic lethal agents based at least in part on the comparing step revealing that the test value does not exceed at least one of the reference values ​​(e.g., is less than ½, ⅓, ¼, ⅕, ⅙, ⅙, ⅙, or ⅙ of the reference value; is less than 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 standard deviations of the reference value). [The present invention 1050] The method of claim 1049, wherein said LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes. [The present invention 1051] The method of claim 1049, wherein the total number of LOH regions is 9, 15, 20, or more. [The present invention 1052] The method of any one of claims 10 to 15, wherein said first length is about 6 megabases, 12 megabases, or 15 megabases or more. [The present invention 1053] 1049. The method of claim 1049, wherein the reference number is 6, 7, 8, 9, 10, 11, 12, or 13 or more. [The present invention 1054] 1049. The method of claim 1049, wherein said chemotherapy is selected from the group consisting of DNA damaging agents, anthracyclines, and topoisomerase I inhibitors, and / or said synthetic lethal agent is a PARP inhibitor. [The present invention 1055] 1049. The method of claim 1049, wherein the DNA damaging agent is cisplatin, carboplatin, oxalaplatin, or picoplatin, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, and / or the PARP inhibitor is iniparib, olaparib, or velapirib. The details of one or more embodiments of the invention are set forth in the following description and accompanying drawings. The materials, methods, and examples are illustrative only and not intended to be limiting. Other features, objects, and advantages of the invention will be apparent from the description and drawings, and from the claims. [Brief description of the drawings]

[0036] [Figure 1] Figure 1 is a graph plotting allele dosage, as determined using SNP arrays, of breast cancer cells from breast cancer patients along chromosome 1. Arrows indicate transitions from regions of heterozygosity to regions of LOH. [Diagram 2] Figure 2 shows a plot of allelic dosage in breast cancer cells from the same breast cancer patient along chromosome 1 as in Figure 1, as determined using high-throughput sequencing. Arrows indicate transitions from regions of heterozygosity to regions of LOH. [Diagram 3] 1 is a flow chart of an exemplary process for evaluating the genome of a cell (e.g., a cancer cell) for an LOH signature. [Figure 4] FIG. 1 illustrates an example of a computing device and a mobile computing device that can be used to implement the techniques described herein. [Diagram 5]1 is a graph plotting the length distribution of LOH regions detected in ovarian cancer cells from 62 human patients. The corrected length refers to the percentage of the chromosome arm covered by the LOH region. [Figure 6] 1 is a graph plotting the number of LOH regions longer than 15 Mb and shorter than an entire chromosome for a training set of ovarian cancer cell samples with intact or deleted BRCA1 and BRCA2 genes. The size of the circle is proportional to the number of samples with such a number of LOH regions. [Figure 7] 1 is a graph plotting the number of LOH regions longer than 15 Mb and shorter than an entire chromosome for training and validation sets of ovarian cancer cell samples with intact or deleted BRCA1 and BRCA2 genes. The size of the circle is proportional to the number of samples with such a number of LOH regions. [Figure 8] 1 is a graph plotting the number of LOH regions longer than 15 Mb and shorter than the entire chromosome for ovarian cancer cell samples with BRCA somatic mutations, BRCA germline mutations, BRCA1 low expression, or intact BRCA (BRCA normal) ovarian cancer cell samples. The size of the circle is proportional to the number of samples with such number of LOH regions. [Figure 9] 1 is a table showing the percentage of BRCA-deficient, HDR-deficient / BRCA-intact, and HDR-intact ovarian cancer samples. [Figure 10] Figure 1 shows a graph plotting the number of LOH regions longer than 15 Mb and shorter than an entire chromosome across cancer cell lines for the indicated cancers. The size of the circle is proportional to the number of samples with such a number of LOH regions. [Figure 11] 1 is a graph plotting the number of LOH regions longer than 15 Mb and shorter than the entire chromosome in lung cancer samples. [Figure 12]1 is a graph plotting the percentage of indicated cancers or cancer cell lines that have HDR deficiency. [Figure 13] Graphs are included plotting IC50 values ​​(Log10 (IC50) for camptothecin and the average of Log10 (IC50) values ​​for platinum compounds (oxaliplatin, cisplatin, and carboplatin) or anthracyclines (doxorubicin and epirubicin) upon exposure to 29 breast cancer cell lines with the indicated number of LOH regions longer than 15 Mb and shorter than an entire chromosome, or IC50 values ​​(Log10 (IC50) for paclitaxel) upon exposure to 27 ovarian cancer cell lines with the indicated number of LOH regions longer than 15 Mb and shorter than an entire chromosome. The threshold is indicated at 9 by the dashed line. [Figure 14] A labeled version of the graph from Figure 13 plotting the average Log10(IC50) values ​​of platinum compounds (oxaliplatin, cisplatin, and carboplatin) upon exposure to 29 breast cancer cell lines with the indicated number of LOH regions longer than 15 Mb and shorter than an entire chromosome. [Figure 15] 1 is a flow chart of an exemplary computational process for identifying loci and regions of LOH. [Figure 16] The percentage of the length of the LOH regions, and the length of these regions corrected for the length of the paired chromosome arms are shown. The maximum corrected value in this figure is equal to 2, which corresponds to LOH across the entire chromosome. [Figure 17a] The HRD scores in tumor samples are shown. Blue circles: BRCA1 or BRCA2-deficient samples. Red circles: BRCA1 and BRCA2-intact samples. The sum of the areas of the blue and red circles is the same. The area of ​​each circle is proportional to the number of samples with the corresponding number of LOH regions. HRD scores of the first cohort (46 of 152 samples were BRCA1 or BRCA2-deficient). [Figure 17b]HRD scores in tumor samples are shown. Blue circles: BRCA1 or BRCA2-deficient samples. Red circles: BRCA1 and BRCA2-intact samples. The sum of the areas of the blue and red circles is the same. The area of ​​each circle is proportional to the number of samples with the corresponding number of LOH regions. HRD scores of the second cohort (19 of 53 samples were BRCA1 or BRCA2-deficient). [Figure 17c] HRD scores in tumor samples are shown. Blue circles: BRCA1 or BRCA2-deficient samples. Red circles: BRCA1 and BRCA2-intact samples. The sum of the areas of the blue and red circles is the same. The area of ​​each individual circle is proportional to the number of samples with the corresponding number of LOH regions. HRD scores of the third cohort (146 of 435 samples were BRCA1 or BRCA2-deficient). [Figure 17d] Figure 1 shows the HRD scores in tumor samples. Blue circles: BRCA1 or BRCA2-deficient samples. Red circles: BRCA1 and BRCA2-intact samples. The sum of the areas of the blue and red circles is the same. The area of ​​each individual circle is proportional to the number of samples with the corresponding number of LOH regions. HRD scores of the combined data from all three cohorts. Row A: 224 samples with either a missing BRCA1 or a missing BRCA2 or a missing RAD51C gene; B: 84 BRCA1 mutants; C: 43 BRCA2 mutants; D: 82 samples with BRCA1 low expression or BRCA1 methylation; E: 13 RAD51C methylation samples. Red circles: 416 samples with BRCA1, BRCA2, and RAD51C intact genes. [Figure 18a]Comparison of HRD scores in cancer cell lines. Red circles: cell lines with intact BRCA1 or BRCA2. A: 30 intact non-ovarian cell lines; B: 22 intact ovarian cell lines. Green circles: 6 carriers of BRCA1 heterozygous mutations or BRCA2 heterozygous mutations. Purple circles: 2 carriers of homozygous mutations with reversion mutations in either BRCA1 or BRCA2. Blue circles: 7 carriers with homozygous mutations in either BRCA1 or BRCA2 or with BRCA1 methylation. The sum of the areas of the green, red, blue, and purple circles is the same. The area of ​​each circle is proportional to the number of samples with the corresponding number of LOH regions. [Figure 18b] Kaplan-Meier plot of postoperative overall survival by HRD score divided by median. These data were generated using 507 samples from the TCGA dataset for which copy number data and survival information were available. Median overall survival for samples with high and low HRD scores was 1499 (95% CI = (1355-1769)) days and 1163 (95% CI = (1081-1354)) days, respectively. [Figure 19] Correlation between LOH score and HR loss calculated for different LOH region length cutoffs for the first cohort is shown. The corresponding log10(p-value) is on the y-axis. The significance of the relationship between cutoffs for the size of the LOH region and the correlation between LOH score and HR loss was examined. The figure shows that the LOH length cutoff can easily be 11-21 Mb. In some preferred embodiments, a cutoff of 15 Mb, which is approximately in the middle of the interval, can be used because it was found to be highly sensitive to statistical noise present in the data. [Figure 20]Comparison of LOH scores in three groups of BRCA1 and BRCA2 null samples across combined data from all three cohorts: Row A: 49 BRCA1 germline mutation carriers; B: 25 BRCA1 somatic mutation carriers; C: 82 samples with BRCA1 methylation or BRCA1 low expression; D: 27 BRCA2 germline mutation carriers; E: 9 BRCA2 somatic mutation carriers. [Figure 21] A comparison of LOH scores for BRCA1-deficient, BRCA2-deficient, and RAD51C-deficient samples is shown. Blue circles correspond to BRCA1-deficient samples, red circles correspond to BRCA2-deficient samples, and green circles correspond to RAD51C-deficient samples. The sum of the areas of the red, blue, and green circles is the same. The area of ​​each circle is proportional to the number of samples with the corresponding number of LOH regions. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0037] Detailed Description This document provides methods and materials involved in evaluating a sample (e.g., cancer cells) for the presence of LOH signatures. For example, this document provides methods and materials for determining whether a cell (e.g., a human cancer cell) contains an LOH signature (e.g., an HDR-deficient LOH signature).

[0038] Generally, by comparing sequences present at the same locus on each chromosome (each autosome in males), it is possible to determine whether a particular locus is homozygous or heterozygous in a cell's genome. Typically, an individual receives one copy from his biological father and one copy from his biological mother, so polymorphic loci in the human genome are generally heterozygous within an individual. In some cases, a polymorphic locus or set of polymorphic loci within an individual will be homozygous as a result of inheriting identical copies from two biological parents.

[0039] Loss of heterozygosity (LOH) can result from several mechanisms. For example, in some cases, a region of one chromosome in a somatic cell may be deleted. Since only one copy (rather than two copies) of that region is present in the genome of the affected cell, the region that still remains in the other chromosome (the other non-sex chromosome in the male) becomes a LOH region. This LOH region may be of any length (e.g., less than about 1.5 Mb in length to a length equal to the entire length of the chromosome). This type of LOH event reduces the copy number. In other cases, a region of one chromosome (the one non-sex chromosome in the male) in a somatic cell may be replaced with a copy of the region from the other chromosome, thereby eliminating any heterozygosity that may have existed in the replaced region. In such cases, the region that still remains in each chromosome becomes a LOH region, which may also be referred to as a copy neutral LOH region. Copy neutral LOH regions may be of any length (e.g., less than about 1.5 Mb in length to a length equal to the entire length of the chromosome).

[0040] As described herein, the genome of the cell being evaluated is characterized in that it is (a) longer than about 1.5 megabases (e.g., longer than about 2, 2.5, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 75, or 100 megabases (Mb), preferably longer than about 14 or 15 or 16 megabases, more preferably longer than about 15 megabases), and (b) 5 or more (e.g., 6 or more, 7 or more) shorter than the length of an entire chromosome containing the LOH region. A cell sample (e.g., a cancer cell sample) can be identified as having a "positive LOH signature status" (or alternatively referred to as an "HDR-deficient LOH signature") if the cell sample contains 9 or more LOH regions that are (a) longer than about 15 Mb and (b) shorter than the length of the entire chromosome that contains the LOH region. Unless otherwise specified, the term "indicator LOH region" refers to a LOH region in a pair of human chromosomes other than the human X / Y sex chromosome pair, characterized by loss of heterozygosity, and having a length of about 1.5 megabases or more, but less than the length of the entire chromosome containing the LOH region. The length of the entire chromosome containing the LOH region may be determined by examining the length of the shorter chromosome of the corresponding chromosome pair in germline cells or non-tumor somatic cells.In some embodiments, an index LOH region is any LOH region that is about 2, 2.5, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, 30, 35, 40, 45, 50, 75, or 100 megabases (Mb) or more (preferably, greater than about 14 or 15 megabases or more) and less than the length of the entire chromosome containing the LOH region.

[0041] A cell (e.g., a cancer cell) identified as having a positive LOH signature (also referred to herein as an "HDR-deficient LOH signature") can be classified as likely to have HDR deficiency and / or as likely to have a defective state defect in one or more genes of an HDR pathway. For example, a cancer cell identified as having a positive LOH signature status can be classified as likely to have an HDR-deficient status. In some cases, a cancer cell identified as having a positive LOH signature status can be classified as likely to have a defective state in one or more genes of an HDR pathway. As used herein, a defective state of a gene means that the sequence, structure, expression, and / or activity of the gene or its product is defective compared to normal. Examples include, but are not limited to, low or no mRNA expression or protein expression, deleterious mutations, hypermethylation, reduced activity (e.g., enzymatic activity, ability to bind to another biomolecule), and the like. As used herein, a defective state of a pathway (e.g., an HDR pathway) means that at least one gene (e.g., BRCA1) in the pathway is defective. Examples of highly deleterious mutations include frameshift mutations, stop codon mutations, and mutations that lead to changes in RNA splicing. Deficiency of HDR pathway genes can result in defective or reduced activity of homologous recombination repair in cancer cells. Examples of HDR pathway genes include, but are not limited to, the genes listed in Table 1.

[0042] Table 1. Selected HDR pathway genes TIFF2025032160000001.tif234147

[0043] Examples of genetic mutations that may be present in HDR pathway genes include, but are not limited to, the genetic mutations listed in Table 2.

[0044] Table 2. Possible genetic variants in selected genes of the HDR pathway TIFF2025032160000002.tif79128

[0045] In some cases, cell samples (e.g., cancer cell samples) can be identified as having multiple LOH regions (e.g., at least 7, 8, 9, 10 or more LOH regions) covering the entire chromosome.Cells (e.g., cancer cells) identified as having multiple LOH regions covering the entire chromosome can be classified as HDR proficiency, i.e., likely to have intact HDR pathway.For example, cancer cells identified as having multiple LOH regions covering the entire chromosome can be identified as likely to have intact BRCA1 and BRCA2 genes.

[0046] As described herein, identifying LOH loci (as well as the size and number of LOH regions) may include first determining the genotype of the sample at various genomic loci (e.g., SNP loci, single bases in large-scale sequencing), and then determining whether the homozygous loci are due to LOH events. Any suitable technique can be used to determine the genotype at loci of interest in the cell genome. For example, single nucleotide polymorphism (SNP) arrays (e.g., human genome-wide SNP arrays), targeted sequencing of loci of interest (e.g., sequencing of SNP loci and their surrounding sequences), and even non-targeted sequencing (e.g., whole exome, transcriptome, or genome sequencing) can be used to identify loci as homozygous or heterozygous. In some cases, the homozygosity or heterozygosity of loci can be analyzed over the length of a chromosome to determine the length of the homozygous or heterozygous region. For example, SNP array results can be used to evaluate regions of SNP locations spaced apart along a chromosome (e.g., spaced apart from about 25 kb to about 100 kb) to determine the presence of homozygous regions along the chromosome as well as the length of the regions. Results from an SNP array can be used to generate graphs plotting allele dosages along the chromosome. Allele dosage d of SNPi i The two alleles (A i and B. i ) can be calculated from the corrected signal intensity: i =A i / (A i +B i ). An example of such a graph is shown in Figure 1. Numerous variations on nucleic acid arrays useful in the present invention are known in the art. These include the arrays used in the various examples below (e.g., Affymetrix 500K GeneChip arrays in Example 3; Affymetrix OncoScan™ FFPE Express 2.0 Services (formerly MIP CN Services) in Example 4).

[0047] Once the sample is genotyped for multiple loci (e.g., SNPs), common techniques can be used to identify the loci and regions of LOH. One method of determining whether homozygosity is due to LOH events is to compare somatic genotypes with germline. For example, multiple loci (e.g., SNPs) can be genotyped in both germline (e.g., blood) and somatic (e.g., tumor) samples. The genotypes of each sample can be compared (typically by computation) to determine whether the germline genome is heterozygous and the somatic genome is homozygous. Such loci are LOH loci, and regions of such loci are LOH regions.

[0048] Computational methods can also be used to determine whether homozygosity is due to LOH events. Such techniques are particularly useful when germline samples are not available for analysis and comparison. For example, algorithms, such as those described elsewhere, can be used to detect LOH regions using information from SNP arrays (Nannya et al., Cancer Res. (2005) 65:6071-6079(2005)). Typically, these algorithms do not explicitly take into account the contamination of tumor samples with benign tissue. See International Application No. PCT / US2011 / 026098 to Abkevich et al.; Goransson et al., PLoS One(2009)4(6):e6057. This contamination is often severe enough to make it difficult to detect LOH regions. The improved analysis methods of the present invention that identify LOH even in the presence of contamination include the analysis methods incorporated in the following computer software products.

[0049] The following is an example. When the observed ratio of the signals of the two alleles A and B is 2:1, there are two possibilities. The first possibility is that the cancer cells have LOH with allele B deleted in a sample that is 50% contaminated by normal cells. The second possibility is that there is no LOH, but allele A is duplicated in a sample that is not contaminated by normal cells. The algorithm can be implemented as a computer program described herein to reconstruct LOH regions based on genotype (e.g., SNP genotype) data. One feature of this algorithm is that it first reconstructs the allele-specific copy number (ASCN) at each locus (e.g., SNP). The ASCN is the copy number of both the paternal and maternal alleles. The LOH regions are then determined as SNP regions where one of the ASCNs (paternal or maternal) is 0. This algorithm may be based on maximizing a likelihood function and may be conceptually similar to previously described algorithms designed to reconstruct the total copy number (not the ASCN) at each locus (e.g., SNP). See International Application No. PCT / US2011 / 026098 to Abkevich et al. The likelihood function can be maximized over the ASCN of all loci, the level of contamination with benign tissue, the total copy number averaged over the entire genome, and the sample-specific noise level. The input data for the algorithm may include or consist of (1) sample-specific normalized signal intensities of both alleles at each locus, and (2) an assay-specific (specific for different SNP arrays and sequence-based approaches) parameter set defined based on the analysis of a large number of samples with known ASCN profiles.

[0050] In some cases, nucleic acid sequencing methods can be used to identify loci as homozygous or heterozygous. For example, genomic DNA can be extracted and fragmented from cell samples (e.g., cancer cell samples). Genomic nucleic acid can be extracted and fragmented using any suitable method, including but not limited to commercially available kits such as QIAamp™ DNA Mini Kit (Qiagen™), MagNA™ Pure DNA Isolation Kit (Roche Applied Science™), and GenElute™ Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich™). Once extracted and fragmented, targeted or non-targeted sequencing can be performed to determine the genotype of the sample at the locus. For example, whole genome, whole transcriptome, or whole exome sequencing can be performed to determine the genotype at millions or even billions of base pairs (i.e., the base pairs can be the "loci" to be evaluated).

[0051] In some cases, targeted sequencing of known polymorphic loci (e.g., SNPs and surrounding sequences) can be performed as an alternative to microarray analysis. For example, kits designed for this purpose (e.g., Agilent SureSelect™, Illumina TruSeq Capture™, and Nimblegen SeqCap EZ Choice™) can be used to enrich genomic DNA for fragments containing the loci (e.g., SNP positions) to be analyzed. For example, genomic DNA containing the loci to be analyzed can be hybridized to biotinylated capture RNA fragments to form biotinylated RNA / genomic DNA complexes. Alternatively, DNA capture probes can be utilized to form biotinylated DNA / genomic DNA hybrids. Streptavidin-coated magnetic beads and magnetic forces can be used to separate biotinylated RNA / genomic DNA complexes from genomic DNA fragments that are not present in the biotinylated RNA / genomic DNA complex. The resulting biotinylated RNA / genomic DNA complexes can be treated to remove the captured RNA from the magnetic beads, thereby leaving intact genomic DNA fragments containing the loci to be analyzed. These intact genomic DNA fragments containing the loci to be analyzed can be amplified, for example, by PCR.The amplified genomic DNA fragments can be sequenced by high-throughput sequencing technology or next-generation sequencing technology, for example, Illumina HiSeq™, Illumina MiSeq™, Life Technologies SoLID™, or Ion Torrent™, or Roche 454™.

[0052] Similar to the microarray analysis described herein, sequencing results from genomic DNA fragments can be used to identify loci as homozygous or heterozygous. In some cases, the homozygosity or heterozygosity of loci can be analyzed across the length of a chromosome to determine the length of the homozygous or heterozygous region. For example, a region of SNP positions spaced apart along a chromosome (e.g., spaced apart from about 25 kb to about 100 kb) can be assessed by sequencing, and the sequencing results can be used to determine the presence of homozygous regions along the chromosome as well as the length of this LOH region. The resulting sequencing results can be used to generate a graph plotting the allele dosage along the chromosome. The allele dosage of SNPi d i The two alleles (A i and B. i ) can be calculated from the corrected capture probe number: i =A i / (A i +B i ). An example of such a graph is shown in Figure 2. Whether homozygosity is due to an LOH event (as opposed to homozygosity in the germline) can be determined as described herein.

[0053] In some cases, a selection process can be used to select loci (e.g., SNP loci) to be evaluated using assays (e.g., SNP array-based assays and sequencing-based assays) configured to identify loci as homozygous or heterozygous. For example, any human SNP location can be selected for inclusion in an SNP array-based assay or sequencing-based assay configured to identify loci as homozygous or heterozygous in a cellular genome. In some cases, 500,000, 1 million, 1.5 million, 2 million, 2.5 million, or more SNP locations present in the human genome can be evaluated to identify SNPs that (a) are not present on the Y chromosome, (b) are not mitochondrial SNPs, (c) have a minor allele frequency of at least about 5 percent in Caucasians, (d) have a minor allele frequency of at least about 1 percent in three non-Caucasian races (e.g., Chinese, Japanese, and Yoruba), and / or (e) do not significantly deviate from Hardy-Weinberg equilibrium in any of the four races. In some cases, more than 100,000, more than 150,000, or more than 200,000 human SNPs that meet criteria (a)-(e) can be selected. Of the human SNPs that meet criteria (a)-(e), a group of SNPs (e.g., the top 110,000 SNPs) can be selected such that the SNPs have a high allele frequency in Caucasians, cover the human genome with some uniform spacing (e.g., at least one SNP every about 25 kb to about 500 kb), and are not in linkage disequilibrium with another selected SNP in any of the four races. In some cases, about 40,000, 50,000, 60,000, 70,000, 80,000, 90,000, 100,000, 110,000, 120,000, 130,000, or more SNPs can be selected as meeting these criteria and included in an assay configured to identify LOH regions throughout the human genome. For example, about 70,000 to about 90,000 (eg, about 80,000) SNPs can be selected for analysis using an SNP array-based assay method.About 45,000 to about 55,000 (eg, about 54,000) SNPs can be selected for analysis using the sequencing-based assay method.

[0054] As described herein, a cell sample can be evaluated to determine whether the genome of the cells of the sample contains LOH signature, lacks LOH signature, has an increased number of LOH regions covering an entire chromosome, or does not have an increased number of LOH regions covering an entire chromosome. Any suitable type of sample can be evaluated. For example, a sample containing cancer cells can be evaluated to determine whether the genome of the cancer cells contains LOH signature, lacks LOH signature, has an increased number of LOH regions covering an entire chromosome, or does not have an increased number of LOH regions covering an entire chromosome. Examples of samples containing cancer cells that can be evaluated as described herein include, but are not limited to, tumor biopsy samples (e.g., breast tumor biopsy samples), formalin-fixed and paraffin-embedded cancer cell-containing tissue samples, core needle biopsy samples, fine needle aspirates, and samples containing cancer cells sloughed off from tumors (e.g., blood, urine, or other bodily fluids). For formalin-fixed, paraffin-embedded tissue samples, samples can be prepared by DNA extraction using genomic DNA extraction kits optimized for FFPE tissue, including but not limited to those kits described above (e.g., QuickExtract™ FFPE DNA Extraction Kit (Epicentre™) and QIAamp™ DNA FFPE Tissue Kit (Qiagen™)).

[0055] In some cases, a laser dissection can be performed on the tissue sample to minimize the number of non-cancerous cells in the cancer cell sample to be evaluated. In some cases, an antibody-based purification method can be used to enrich for cancer cells and / or to deplete non-cancerous cells. Examples of antibodies that can be used for cancer cell enrichment include, but are not limited to, anti-EpCAM, anti-TROP-2, anti-c-Met, anti-folate binding protein, anti-N-cadherin, anti-CD318, anti-anti-mesenchymal stem cell antigen, anti-Her2, anti-MUCl, anti-EGFR, anti-cytokeratin (e.g., cytokeratin 7, cytokeratin 20, etc.), anti-caveolin-1, anti-PSA, anti-CA125, and anti-surfactant protein antibodies.

[0056] Any type of cancer cell can be evaluated using the methods and materials described herein.For example, breast cancer cell, ovarian cancer cell, liver cancer cell, esophageal cancer cell, lung cancer cell, head and neck cancer cell, prostate cancer cell, colon cancer cell, rectal cancer cell, or colorectal cancer cell, and pancreatic cancer cell can be evaluated to determine whether the genome of cancer cell contains LOH signature, whether it lacks LOH signature, whether the number of LOH regions covering the whole chromosome is increased, or whether the number of LOH regions covering the whole chromosome is not increased.In some embodiments, cancer cell is the primary cancer cell or metastatic cancer cell of ovarian cancer, breast cancer, lung cancer, or esophageal cancer.

[0057] When evaluating the genome of a cancer cell for the presence or absence of LOH signature, one or more pairs of chromosomes (e.g., one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, six, seven, eight, nine, nine, twenty, twenty-one, twenty-two, or twenty-three pairs) can be evaluated. In some cases, the genome of a cancer cell is evaluated for the presence or absence of LOH signature using one or more pairs of chromosomes (e.g., one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, six, seven, eight, nine, nine, twenty, twenty-one, twenty-two, or twenty-three pairs).

[0058] In some cases, it may be beneficial to exclude certain chromosomes from the analysis. For example, in the case of females, the pair to be evaluated may include the X sex chromosome pair. In contrast, in the case of males, any pair of autosomes (i.e., any pair other than the X and Y sex chromosome pair) may be evaluated. As another example, in some cases, the chromosome 17 pair may be excluded from the analysis. In certain cancers, certain chromosomes have been determined to have higher levels of LOH than normal. Therefore, it may be beneficial to exclude such chromosomes from patients with these cancers when analyzing samples as described herein. In some cases, the sample is from a patient with ovarian cancer, and the excluded chromosome is chromosome 17.

[0059] When evaluating the genome of cancer cells for the presence or absence of multiple LOH regions covering the entire chromosome, 10 or more pairs of chromosomes (e.g., 13, 16, 19, or 23 pairs) can be evaluated. For females, the pairs to be evaluated may include the X sex chromosome pair. In contrast, for males, any pair of autosomes (i.e., any pair other than the X and Y sex chromosome pair) can be evaluated. In some cases, the 17th chromosome pair can be excluded from the analysis. In some cases, the sample is from a patient with ovarian cancer, and the excluded chromosome is chromosome 17. In some cases, the genome of cancer cells is evaluated for the presence or absence of multiple LOH regions covering the entire chromosome using 10 or more pairs of chromosomes (e.g., 13, 16, 19, or 23 pairs).

[0060] Thus, a predefined number of chromosomes may be analyzed to determine the total number of index LOH regions, preferably greater than 9 megabases, greater than 10 megabases, greater than 12 megabases, greater than 14 megabases, more preferably greater than 15 megabases in length. Alternatively, or in addition, the sizes of all identified index LOH regions may be summed to obtain the total length of the index LOH regions.

[0061] To classify a positive LOH signature status, the aforementioned reference number for the total number of index LOH regions may be 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, 20 or more, preferably 5, preferably 8, more preferably 9 or 10, most preferably 10. The reference number for the total length (e.g., sum of the lengths) of the index LOH regions may be about 75, 90, 105, 120, 130, 135, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 425, 450, 475, 500 megabases or more, preferably about 75 megabases or more, preferably about 90 or 105 megabases or more, more preferably about 120 or 130 megabases or more, more preferably about 135 megabases or more, and most preferably about 150 megabases or more.

[0062] In some specific embodiments, the total number of LOH regions greater than about 14 or 15 megabases in length is determined and compared to a reference number of about 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, or 20. Alternatively, or in addition, the total length of LOH regions greater than about 14 or 15 megabases in length is determined and compared to a reference number of about 75, 90, 105, 120, 130, 135, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 425, 450, 475, or 500 megabases.

[0063] In some embodiments, the number (or sum of lengths, or test value or score obtained from either) of LOH regions in a patient sample is considered to be "above" the reference if it is at least 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, or 10-fold greater than the reference, and furthermore, in some embodiments, is considered to be "above" the reference if it is at least 1-, 2-, 3-, 4-, 5-, 6-, 7-, 8-, 9-fold, or 10-fold greater than the reference. Conversely, in some embodiments, the number (or sum of lengths, or test value or score obtained from either) of LOH regions in a patient sample is considered to be "not above" the reference if it is less than ½, ⅓, ¼, ⅕, ⅙, ⅙, ⅙, ⅙, or ⅙ of the reference, and furthermore, in some embodiments, it is considered to be "not above" the reference if it is less than 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 standard deviations of the reference.

[0064] In some embodiments, the reference number (or length, value, or score) is obtained from a relevant reference population. Such a reference population may include: (a) patients with the same cancer as the patient being tested, (b) patients with the same cancer subtype, (c) patients with cancer with similar genetic or other clinical or molecular characteristics, (d) patients who have responded to a certain treatment, (e) patients who have not responded to a certain treatment, (f) apparently healthy patients (e.g., patients who do not have cancer or at least patients who do not have the cancer of the patient being tested), etc. The reference number (or length, value, or score) may be (a) representative of the number (or length, value, or score) generally found in the reference population; (b) the average (mean, median, etc.) of the numbers (or lengths, values, or scores) generally found in the reference population or in a particular subpopulation; or (c) representative of the numbers (or lengths, values, or scores) (e.g., average values, e.g., mean or median) found in a tercile, quartile, quintile, etc. of the reference population, as ranked by (i) the respective number (or length, value, or score) or (ii) the clinical feature (e.g., strength of response, prognosis (including time to cancer-specific death), etc.) that they are found to have.

[0065] As described herein, a patient having cancer cells identified as having a positive LOH signature status can be classified as likely to respond to a certain cancer treatment regimen based at least in part on the positive LOH signature status.For example, a patient having cancer cells having a genome containing LOH signature can be classified as likely to respond to a cancer treatment regimen that includes the use of DNA damaging agents, synthetic lethal agents (e.g., PARP inhibitors), radiation, or a combination thereof based at least in part on the positive LOH signature status.Preferably, the patient is a treatment-naive patient.Examples of DNA damaging agents include, but are not limited to, platinum-based chemotherapeutic agents (e.g., cisplatin, carboplatin, oxaliplatin, and picoplatin), anthracyclines (e.g., epirubicin and doxorubicin), topoisomerase I inhibitors (e.g., campothecin, topotecan, and irinotecan), DNA cross-linking agents such as mitomycin C, and triazene compounds (e.g., dacarbazine and temozolomide). The synthetic lethality therapeutic approach typically involves administering an agent that inhibits at least one important component from a biological pathway that is particularly important for the survival of a certain tumor cell.For example, when a tumor cell has a homologous repair pathway deficiency (e.g., a homologous repair pathway deficiency determined according to the present invention), an inhibitor of poly ADP-ribose polymerase (or a platinum drug, a double-strand break repair inhibitor, etc.) may be particularly effective against such tumors.Because two pathways that are important for survival are blocked (one biologically, e.g., by BRCA1 mutation, and the other synthetically, e.g., by administering a pathway drug).The synthetic lethality approach of cancer therapy is described, for example, in O'Brien et al., Converting cancer mutations into therapeutic opportunities, EMBO MOL. MED. (2009)1:297-299.Examples of synthetic lethal agents include, but are not limited to, PARP inhibitors or double-strand break repair inhibitors in homologous repair-deficient tumor cells, PARP inhibitors in PTEN-deficient tumor cells, and methotrexate in MSH2-deficient tumor cells. Examples of PARP inhibitors include, but are not limited to, olaparib, iniparib, and veliparib. Examples of double-strand break repair inhibitors include, but are not limited to, KU55933 (ATM inhibitor) and NU7441 (DNA-PKcs inhibitor). Examples of information that can be used in addition to a positive LOH signature status to form the basis of classification as likely to respond to a particular cancer treatment regimen include, but are not limited to, previous treatment results, germline or somatic DNA mutations, gene or protein expression profiling (e.g., ER / PR / HER2 status, PSA levels), tumor histology (e.g., adenocarcinoma, squamous cell carcinoma, papillary serous carcinoma, mucinous carcinoma, invasive ductal carcinoma, ductal carcinoma in situ (non-invasive), etc.), stage, grade of the tumor or cancer (e.g., well differentiated, moderately differentiated, or poorly differentiated (e.g., Gleason, modified Bloom Richardson), etc.), number of previous courses of treatment, etc.

[0066] Once classified as likely to respond to a certain cancer treatment regimen (e.g., a cancer treatment regimen that includes the use of DNA damaging agents, PARP inhibitors, radiation, or a combination thereof), the cancer patient can be treated with such a cancer treatment regimen. In some embodiments, the patient is a treatment-naive patient. The cancer patient identified as having cancer cells with positive LOH signature status can be treated with any suitable method for treating the cancer in question. For example, a platinum-based chemotherapy drug or a combination of platinum-based chemotherapy drugs can be used to treat cancer as described elsewhere (see, for example, U.S. Patent Nos. 3,892,790, 3,904,663, 7,759,510, 7,759,488, and 7,754,684). In some cases, anthracycline or a combination of anthracyclines can be used to treat cancer as described elsewhere (see, for example, U.S. Patent Nos. 3,590,028, 4,138,480, 4,950,738, 6,087,340, 7,868,040, and 7,485,707). In some cases, topoisomerase I inhibitors or a combination of topoisomerase I inhibitors can be used to treat cancer as described elsewhere (see, for example, U.S. Patent Nos. 5,633,016 and 6,403,563). In some cases, PARP inhibitors or a combination of PARP inhibitors can be used to treat cancer as described elsewhere (see, for example, U.S. Patent Nos. 5,177,075, 7,915,280, and 7,351,701). In some cases, radiation can be used to treat cancer as described elsewhere (see, e.g., U.S. Patent No. 5,295,944). In some cases, cancer can be treated with combinations that include different agents (e.g., combinations that include either platinum-based chemotherapy drugs, anthracyclines, topoisomerase I inhibitors, and / or PARP inhibitors) with or without radiation therapy.In some cases, the combination therapy may include any of the above agents or treatments (e.g., a DNA damaging agent, a PARP inhibitor, radiation, or a combination thereof) along with another agent or treatment, such as a taxane agent (e.g., doxetaxel, paclitaxel, abraxane), a growth factor inhibitor or growth factor receptor inhibitor (e.g., erlotinib, gefitinib, lapatinib, sunitinib, bevacizumab, cetuximab, trastuzumab, panitumumab), and / or antimetabolite (e.g., 5-flourouracil, methotrexate).

[0067] In some cases, the patient identified as having cancer cells with genome lacking LOH signature can be classified as unlikely to respond to a treatment regimen that includes DNA damage agent, PARP inhibitor, radiation, or a combination thereof, based at least in part on negative LOH signature status.Then, such a patient can be classified as likely to respond to a cancer treatment regimen that includes the use of one or more cancer treatment agents that are not related to HDR, such as taxane agents (e.g., doxetaxel, paclitaxel, abraxane), growth factor inhibitors or growth factor receptor inhibitors (e.g., erlotinib, gefitinib, lapatinib, sunitinib, bevacizumab, cetuximab, trastuzumab, panitumumab), and / or antimetabolites (e.g., 5-fluorouracil, methotrexate).In some embodiments, the patient is an untreated patient. Once classified as likely to respond to a particular cancer treatment regimen (e.g., a cancer treatment regimen that includes the use of a cancer treatment agent that is not associated with HDR), the cancer patient can be treated with such a cancer treatment regimen. Any method appropriate for the cancer being treated can be used to treat the cancer patient identified as having cancer cells with a negative LOH signature status. Examples of information that can be used in addition to the negative LOH signature status that is the basis for classification as likely to respond to a particular cancer treatment regimen include, but are not limited to, previous treatment results, germline or somatic DNA mutations, gene or protein expression profiling (e.g., ER / PR / HER2 status, PSA level), tumor histology (e.g., adenocarcinoma, squamous cell carcinoma, papillary serous carcinoma, mucinous carcinoma, invasive ductal carcinoma, ductal carcinoma in situ (non-invasive), etc.), stage, tumor or cancer grade (e.g., well-differentiated, moderately differentiated, or poorly differentiated (e.g., Gleason, modified Bloom Richardson), etc.), number of previous treatment courses, etc.

[0068] Upon treatment for a certain period of time (e.g., 1 month to 6 months), the patient can be evaluated to determine whether the treatment regimen is effective. If a beneficial effect is detected, the patient can continue the same or similar cancer treatment regimen. If minimal or no beneficial effect is detected, adjustments can be made to the cancer treatment regimen. For example, the dose, frequency of administration, or duration of treatment can be increased. In some cases, additional anticancer agents can be added to the treatment regimen, and certain anticancer agents can be replaced with one or more different anticancer agents. As appropriate, the treated patient can continue to be monitored, and changes can be made to the cancer treatment regimen as appropriate.

[0069] In addition to predicting likely therapeutic response or selecting desired therapeutic regimen, LOH signature can be used to determine patient prognosis.As shown in Example 3 below (particularly in FIG. 18b), patients with tumors that have LOH signatures show significantly longer survival than patients with tumors that do not have such LOH signatures.Thus, in one aspect, this document features a method for determining patient prognosis, based at least in part on detecting the presence or absence of LOH signatures in samples from patients. The method includes, or consists essentially of, (a) determining whether the patient contains cancer cells having an LOH signature as described herein (e.g., the presence of more than a reference number of LOH regions in at least one pair of human chromosomes from the cancer cells of the cancer patient indicates that the cancer cells have an LOH signature, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, wherein the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more), and (b)(1) determining that the patient has a relatively good prognosis based at least in part on the presence of the LOH signature, or (b)(2) determining that the patient has a relatively poor prognosis based at least in part on the absence of the LOH signature. A prognosis may include the patient's chance of survival (e.g., progression-free survival, overall survival), with a relatively good prognosis including an increased chance of survival when compared to some reference population (e.g., the average patient with this patient's cancer type / subtype, the average patient without the LOH signature, etc.). Conversely, a relatively poor prognosis in terms of survival includes a decreased chance of survival when compared to some reference population (e.g., the average patient with this patient's cancer type / subtype, the average patient with the LOH signature, etc.).

[0070] As described herein, this document provides a method for evaluating a patient for cells (e.g., cancer cells) that have genomes that contain LOH signatures. In some embodiments, the patient is an untreated patient. For example, one or more clinicians or medical practitioners can determine whether the patient contains cancer cells that have genomes that contain LOH signatures. In some cases, one or more clinicians or medical practitioners can obtain a cancer cell sample from the patient, and evaluate the genomes of the cancer cells of the cancer cell sample to determine whether LOH signatures are present as described herein, thereby determining whether the patient contains cancer cells that have genomes that contain LOH signatures.

[0071] In some cases, one or more clinicians or medical practitioners can obtain cancer cell samples from patients and provide the samples to a testing institution capable of evaluating the genome of cancer cells of the cancer cell samples to indicate the presence or absence of LOH signature as described herein. In some embodiments, the patient is an untreated patient. In such a case, one or more clinicians or medical practitioners can determine whether the patient contains cancer cells with genomes containing LOH signatures by directly or indirectly receiving information about the presence or absence of LOH signatures from a testing institution. For example, after the testing institution evaluates the genome of cancer cells for the presence or absence of LOH signatures as described herein, the testing institution can provide or allow the clinician or medical practitioner to access a written, electronic, or verbal report or medical record that indicates the presence or absence of LOH signatures of a particular patient being evaluated. With such written, electronic, or verbal report or medical record, one or more clinicians or medical practitioners can determine whether the particular patient being evaluated contains cancer cells with genomes containing LOH signatures.

[0072] When a clinician or a medical practitioner or a group of clinicians or medical practitioners determine that a certain patient being evaluated contains cancer cells with genomes that contain LOH signature, the clinician or a medical practitioner (or group) can classify the patient as having cancer cells with genomes that contain the presence of LOH signature.In some embodiments, the patient is an untreated patient.In some cases, the clinician or a medical practitioner or a group of clinicians or medical practitioners can diagnose the patient that is determined to have cancer cells with genomes that contain LOH signature as having cancer cells that are likely to be HDR defective.This diagnosis can be based simply on determining that a certain patient being evaluated contains cancer cells with genomes that contain LOH signature, or can be based at least in part on determining that a certain patient being evaluated contains cancer cells with genomes that contain LOH signature. For example, a patient determined to have cancer cells with a genome that includes the presence of an LOH signature can be diagnosed as likely HDR deficient based on a combination of a positive LOH signature status and a deletion status in one or more tumor suppressor genes (e.g., BRCA1 / 2, RAD51C), a family history of cancer, or the presence of behavioral risk factors (e.g., smoking).

[0073] In some cases, a clinician or medical practitioner or a group of clinicians or medical practitioners can diagnose a patient who is determined to have a cancer cell with genome comprising the presence of LOH signature as having cancer cells that are likely to contain one or more genetic mutations of HDR pathway genes.In some embodiments, the patient is an untreated patient.Such a diagnosis can be based simply on determining that a certain patient being evaluated has a cancer cell with genome comprising LOH signature, or can be based at least in part on determining that a certain patient being evaluated has a cancer cell with genome comprising LOH signature.For example, a patient who is determined to have a cancer cell with genome comprising LOH signature can be diagnosed as having cancer cells that are likely to contain one or more genetic mutations of HDR pathway genes based on a combination of positive LOH positive status and family history of cancer, or the presence of behavioral risk factors (e.g., smoking).

[0074] In some cases, a clinician or medical practitioner or a group of clinicians or medical practitioners can diagnose a patient who is determined to have cancer cells with genomes that include the presence of LOH signature as having cancer cells that are likely to respond to a certain cancer treatment regimen. In some embodiments, the patient is an untreated patient. Such a diagnosis can be based simply on determining that a certain patient being evaluated has cancer cells with genomes that include LOH signatures, or can be based at least in part on determining that a certain patient being evaluated has cancer cells with genomes that include LOH signatures. For example, a patient who is determined to have cancer cells with genomes that include the presence of LOH signatures can be diagnosed as having high likelihood of responding to a certain cancer treatment regimen based on a combination of positive LOH signature status and the deletion status of one or more tumor suppressor genes (e.g., BRCA1 / 2, RAD51), family history of cancer, or the presence of behavioral risk factors (e.g., smoking). As described herein, the patient who is determined to have cancer cells with genome comprising the presence of LOH signature can be diagnosed as likely to respond to cancer treatment regimen, including the use of platinum-based chemotherapy drugs, such as cisplatin, carboplatin, oxaliplatin or picoplatin, anthracycline, such as epirubicin or doxorubicin, topoisomerase I inhibitor, such as campothecin, topotecan or irinotecan, PARP inhibitor, radiation, combination thereof, or any of the above in combination with another anticancer drug.In some embodiments, the patient is an untreated patient.

[0075] When a clinician or a medical practitioner or a group of clinicians or medical practitioners determine that a certain patient being evaluated contains cancer cells with genomes that lack LOH signature, the clinician or a medical practitioner (or a group) can classify the patient as having cancer cells with genomes that include the absence of LOH signature.In some embodiments, the patient is an untreated patient.In some cases, the clinician or a medical practitioner or a group of clinicians or medical practitioners can diagnose a patient that is determined to have cancer cells that contain genomes that lack the presence of LOH signature as having cancer cells that are likely to have functional HDR.In some cases, the clinician or a medical practitioner or a group of clinicians or medical practitioners can diagnose a patient that is determined to have cancer cells that contain genomes that lack the presence of LOH signature as having cancer cells that are unlikely to contain genetic mutations in one or more HDR pathway genes. In some cases, a clinician or medical practitioner or a group of clinicians or medical practitioners can diagnose a patient who is determined to have cancer cells that contain genomes that lack the presence of LOH signatures or contain multiple LOH regions that cover entire chromosomes as having cancer cells that are less likely to respond to platinum-based chemotherapy drugs, such as cisplatin, carboplatin, oxalaplatin, or picoplatin, anthracyclines, such as epirubincin or doxorubicin, topoisomerase I inhibitors, such as campothecin, topotecan, or irinotecan, PARP inhibitors, or radiation, and / or are more likely to respond to cancer treatment regimens that include the use of cancer treatments that are not related to HDR, such as one or more taxanes, growth factor inhibitors or growth factor receptor inhibitors, antimetabolites, etc. In some embodiments, the patient is an untreated patient.

[0076] As described herein, this document also provides a method for carrying out diagnostic analysis of a nucleic acid sample (e.g., genomic nucleic acid sample or amplified genomic nucleic acid sample) of a cancer patient to determine whether the cancer cells in the patient have a genome that contains LOH signature and / or multiple LOH regions that cover the entire chromosome.In some embodiments, the patient is an untreated patient.For example, one or more laboratory technicians or laboratory personnel can detect the presence or absence of LOH signature in the genome of the patient's cancer cells, or the presence or absence of multiple LOH regions that cover the entire chromosome in the genome of the patient's cancer cells. In some cases, one or more laboratory technicians or laboratory workers (a) receive a cancer cell sample obtained from a patient, receive a genomic nucleic acid sample obtained from a cancer cell obtained from a patient, or receive an enriched and / or amplified genomic nucleic acid sample obtained from a cancer cell obtained from a patient, and (b) use the received material to perform an analysis (e.g., an SNP array-based assay or a sequencing-based assay) to detect the presence or absence of an LOH signature or the presence or absence of a number of LOH regions covering an entire chromosome in the genome of the cancer cell of the patient by detecting the presence or absence of an LOH signature or the presence or absence of a number of LOH regions covering an entire chromosome as described herein.In some cases, one or more laboratory technicians or laboratory workers can receive the sample to be analyzed (e.g., a cancer cell sample obtained from a patient, a genomic nucleic acid sample obtained from a cancer cell obtained from a patient, or an enriched and / or amplified genomic nucleic acid sample obtained from a cancer cell obtained from a patient) directly or indirectly from a clinician or medical worker.In some embodiments, the patient is an untreated patient.

[0077] When a laboratory technician or testing worker or group of laboratory technicians or testing workers detects the presence of an LOH signature as described herein, the laboratory technician or testing worker (or group) can identify a patient having cancer cells detected as having an LOH signature as having cancer cells having a positive LOH signature status. For example, one or more laboratory technicians or testing workers can identify a patient having cancer cells detected as having an LOH signature as having cancer cells having a positive LOH signature status by associating the positive LOH signature status or the results (or summary results) of a diagnostic analysis performed with the corresponding patient's name, medical record, symbolic / numeric identifier, or a combination thereof. In some cases, a laboratory technician or a group of laboratory technicians or laboratory technicians can identify a patient having cancer cells detected as having LOH signature as having cancer cells that are potentially HDR-deficient by associating positive LOH signature status, potential deficiency in HDR status, or the result (or summary of results) of diagnostic analysis performed with the corresponding patient's name, medical record, symbol / number identifier, or a combination thereof. Such identification can be based solely on detecting the presence of LOH signature, or can be based at least in part on detecting the presence of LOH signature. For example, a laboratory technician or a group of laboratory technicians can identify a patient having cancer cells detected as having LOH signature as having cancer cells that are potentially HDR-deficient based on a combination of positive LOH signature status and the results of other genetic and biochemical tests performed in a testing facility. In some embodiments, the patient is an untreated patient.

[0078] In some cases, a laboratory technician or a laboratory worker or a group of laboratory technicians or laboratory workers can identify a patient having cancer cells detected as having LOH signature as having cancer cells potentially containing one or more HDR pathway gene mutations by associating positive LOH signature status, the potential presence of one or more HDR pathway gene mutations, or the results (or summary results) of diagnostic analysis performed with the corresponding patient's name, medical record, symbol / number identifier, or a combination thereof. Such identification can be based solely on detecting the presence of LOH signature, or can be based at least in part on detecting the presence of LOH signature. For example, a laboratory technician or a laboratory worker can identify a patient having cancer cells detected as having LOH signature as having cancer cells potentially containing one or more HDR pathway gene mutations based on a combination of positive LOH signature status and the results of other genetic and biochemical tests performed in a testing facility. In some embodiments, the patient is an untreated patient.

[0079] In some cases, a laboratory technician or a group of laboratory technicians or laboratory technicians can identify a patient having cancer cells detected with LOH signature as having cancer cells that are likely to respond to a certain cancer treatment regimen by associating the positive LOH signature status, the potentially missing HDR status, the potential presence of one or more HDR pathway gene missing status, or the result (or summary of results) of the diagnostic analysis performed with the corresponding patient's name, medical record, symbol / number identifier, or a combination thereof. Such identification can be based solely on detecting the presence of LOH signature, or can be based at least in part on detecting the presence of LOH signature. For example, a laboratory technician or a group of laboratory technicians can identify a patient having cancer cells detected with LOH signature as having cancer cells that are likely to respond to a certain cancer treatment regimen based on a combination of the positive LOH signature status and the results of other genetic and biochemical tests performed in the testing facility. In some embodiments, the patient is an untreated patient.

[0080] When a laboratory technician or testing worker or group of laboratory technicians or testing workers detects the absence of an LOH signature, the laboratory technician or testing worker (or group) can identify patients having cancer cells detected as lacking an LOH signature as having cancer cells having a negative LOH signature status. For example, one or more laboratory technicians or testing workers can identify patients having cancer cells detected as lacking an LOH signature as having cancer cells having a negative LOH signature status by associating the negative LOH signature status or the results (or summary results) of the diagnostic analysis performed with the corresponding patient's name, medical record, symbolic / numeric identifier, or a combination thereof. In some cases, a laboratory technician or testing personnel or group of laboratory technicians or testing personnel can identify patients having cancer cells detected as lacking an LOH signature as having cancer cells with potentially intact HDR by correlating the negative LOH signature status, potentially intact HDR status, or the results (or summary results) of the diagnostic analysis performed with the corresponding patient's name, medical record, symbolic / numeric identifier, or a combination thereof. In some embodiments, the patient is an untreated patient.

[0081] In some cases, a laboratory technician or testing practitioner, or group of laboratory technicians or testing practitioners, can identify patients having cancer cells detected as lacking an LOH signature as having cancer cells with potentially intact HDR pathway genes by correlating the negative LOH signature status, the potential absence of genetic mutations in HDR pathway genes, or the results (or summary results) of diagnostic analyses performed with the corresponding patient name, medical record, symbolic / numeric identifier, or a combination thereof.

[0082] In some cases, a laboratory technician or a group of laboratory technicians or technicians may identify patients having cancer cells detected as lacking an LOH signature as being indicative of a disease that is not ... The patient may be identified as having cancer cells that are less likely to respond to certain cancer treatment regimens (e.g., cancer treatment regimens that include the use of cancer treatments that are not associated with HDR), such as platin, oxalaplatin, or picoplatin, anthracyclines such as epirubicin or doxorubicin, topoisomerase I inhibitors such as campothecin, topotecan, or irinotecan, PARP inhibitors such as iniparib, olaparib, or velapirib, or radiation). In some embodiments, the patient is a treatment-naive patient.

[0083] When a technologist or a group of technologists ...

[0084] The results of any analysis according to the present invention are often communicated to a physician, genetic counselor, and / or patient (or other parties such as researchers) in a communicable form that can be communicated or communicated to any of the aforementioned parties. Such forms may vary and may be tangible or intangible. The results may be embodied in a description, diagram, photograph, chart, image, or any other visual form. For example, graphs or diagrams showing genotype or LOH (or HRD status) information may be used in describing the results. The descriptions and visual forms may be recorded on a tangible medium, such as paper, a computer readable medium, such as a floppy disk, compact disk, flash memory, etc., or on an intangible medium, such as an electronic medium in the form of an e-mail or a website on the Internet or an intranet. Additionally, the results may be recorded in audio form and communicated via any suitable medium, such as analog or digital cable lines, fiber optic cables, etc., telephone, facsimile, wireless mobile phone, Internet telephone, etc.

[0085] Thus, the information and data on the test results can be generated anywhere in the world and transmitted to different locations. For example, when the assay method is performed outside the United States, the information and data on the test results can be generated, cast in a transmittable form as described above, and then imported into the United States. Thus, the present invention also encompasses a method for generating a transmittable form of information on the LOH signature for at least one patient sample. The method includes the steps of (1) determining an LOH signature according to the method of the present invention; and (2) embodying the result of the determining step in a transmittable form. The transmittable form is the result of such a method.

[0086] Some embodiments of the invention described herein involve correlating an LOH signature according to the invention (e.g., a total number of LOH regions in at least one pair of human chromosomes from said cancer cell, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, wherein the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and wherein the first length is about 1.5 megabases or more) with a certain clinical feature (e.g., an increased likelihood of defects in the BRCA1 or BRCA2 genes; an increased likelihood of HDR defects; an increased likelihood of response to a therapeutic regimen including a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, and / or a PARP inhibitor, etc.) if the number is greater than some reference (or, optionally, correlating with another feature if the number is less than some reference). Throughout this document, wherever such an embodiment is described, another embodiment of the invention may involve one or both of the following steps in addition to or instead of the correlating step: (a) concluding that the patient has the clinical trait based at least in part on the presence or absence of the LOH signature; or (b) communicating that the patient has the clinical trait based at least in part on the presence or absence of the LOH signature.

[0087] By way of example, one embodiment described in this document is a method for predicting the response of a cancer patient to a cancer treatment regimen that includes, but is not limited to, a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, and / or a PARP inhibitor. The method includes the steps of: (1) determining the number of LOH regions in at least one pair of human chromosomes from a cancer cell derived from the cancer patient, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (2) correlating the total number that is greater than a reference number with an increased likelihood that the cancer patient may respond to the cancer treatment regimen. According to the above paragraph, this description of this embodiment is understood to include the description of two related embodiments, namely, a method of predicting a cancer patient's response to a cancer treatment regimen comprising a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, and / or a PARP inhibitor, comprising the steps of: (1) determining, in a cancer cell derived from the cancer patient, a number of LOH regions in at least a pair of human chromosomes from the cancer cell of the cancer patient, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (2)(a) concluding that the cancer patient is likely to respond to the cancer treatment regimen based at least in part on the total number exceeding a reference number, or (2)(b) conveying that the cancer patient is likely to respond to the cancer treatment regimen based at least in part on the total number exceeding a reference number.

[0088] In each embodiment described herein involving correlating the output of a particular assay or analysis (e.g., total number of LOH regions greater than a reference number, etc.) with some probability (e.g., increased, no increase, decreased, etc.) of some clinical feature (e.g., response to a particular treatment, cancer-specific mortality, etc.), or additionally or alternatively concluding or communicating such clinical feature based at least in part on the output of such particular assay or analysis, such correlating, concluding or communicating may include assigning a risk or likelihood of the clinical feature occurring based at least in part on the output of the particular assay or analysis. In some embodiments, such risk is a percentage probability of an event or outcome occurring. In some embodiments, patients are assigned to a risk group (e.g., low risk, medium risk, high risk, etc.). In some embodiments, "low risk" is any percentage probability less than 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, or 50%. In some embodiments, "medium risk" is any percent probability greater than 5%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, or 50% and less than 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, or 75%. In some embodiments, "high risk" is any percent probability greater than 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, or 99%.

[0089] As used herein, "communicating" a particular piece of information means making such information known to another person or transferring such information to something (e.g., a computer). In some methods of the present invention, a patient's prognosis or likelihood of response to a particular treatment is communicated. In some embodiments, information used to arrive at such a prognosis or response prediction (e.g., LOH signature according to the present invention, etc.) is communicated. This communication may be audible (e.g., verbal), visual (e.g., written), electronic (e.g., data transferred from one computer system to another computer system), etc. In some embodiments, communicating the cancer classification (e.g., prognosis, likelihood of response, appropriate treatment, etc.) includes creating a report that communicates the cancer classification. In some embodiments, the report is a paper report, an audible report, or an electronic record. In some embodiments, the report is displayed and / or stored on a computing device (e.g., a handheld device, a desktop computer, a smart device, a website, etc.). In some embodiments, the cancer classification is communicated to a physician (e.g., a report that communicates the classification is provided to the physician). In some embodiments, the cancer classification is communicated to a patient (e.g., a report that communicates the classification is provided to the patient). Cancer classification may also be conveyed by transferring information (e.g., data) embodying the classification to a server computer and making such information accessible to an intermediate or end user (e.g., by viewing the information as displayed from the server, by downloading the information in the form of one or more files transferred from the server to the intermediate or end user's device, etc.).

[0090] Whenever an embodiment of the invention involves concluding some fact (e.g., a patient's prognosis or the patient's likelihood of responding to a particular treatment regimen), this may in some embodiments include a computer program that concludes such a fact, typically after running an algorithm that applies information about LOH regions in accordance with the invention.

[0091] In each of the embodiments described herein involving a number of LOH regions (e.g., LOH indicator regions) or the total length of such LOH regions, the invention encompasses related embodiments involving test values ​​or scores (e.g., HRD scores, LOH scores, etc.) derived from, incorporating, and / or reflecting, at least to some extent, such numbers or lengths. In other words, in various methods, systems, etc. of the invention, the number or length of bare LOH regions need not be used, but rather a test value or score derived from such numbers or lengths may be used.For example, one embodiment of the present invention provides a method of treating cancer in a patient, comprising the steps of: (1) determining in a sample from the patient a number of LOH regions in at least a pair of human chromosomes from a cancer cell of the cancer patient, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions indicative of the cancer cell having an LOH signature, wherein the at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; (2) providing a test value derived from the number of LOH regions; (3) comparing the test value to one or more reference values ​​(e.g., mean, median, tertile, quartile, quintile, etc.) derived from the number of LOH regions in a reference population; and (4)(a) determining whether the test value exceeds at least one of the reference values ​​(e.g., at least 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, or 10-fold the reference value; or (4)(b) administering an anti-cancer drug to the patient or recommending or prescribing or initiating a treatment regimen that includes chemotherapy and / or synthetic lethal agents based, at least in part, on the comparing step that reveals that the test value is not above at least one of the reference values ​​(e.g., not more than ½, ⅓, ¼, ⅕, ⅙ ... The invention encompasses corresponding embodiments in which, mutatis mutandis, the test value or score is used to determine a patient's prognosis, the likelihood that a patient will respond to a particular therapeutic regimen, the likelihood that a patient or patient sample has a BRCA1 deficiency, a BRCA2 deficiency, a RAD51C deficiency, or an HDR deficiency, etc.

[0092] FIG. 15 illustrates an exemplary process by which a computational system (or a computer program (e.g., software) including computer executable instructions) can identify LOH loci or regions from genotype data as described herein. When the observed ratio of signals of the two alleles A and B is 2:1, there are two possibilities. The first possibility is that the cancer cells have LOH with allele B deleted in a sample that is 50% contaminated by normal cells. The second possibility is that there is no LOH, but allele A is duplicated in a sample that is not contaminated by normal cells. The process starts at box 1500, where the following data is collected by the computational system: (1) sample-specific normalized signal intensities of both alleles at each locus, and (2) an assay-specific (specific to different SNP arrays and sequence-based approaches) parameter set defined based on the analysis of a large number of samples with known ASCN profiles. As described herein, loci along a chromosome can be assessed for homozygosity or heterozygosity using any suitable assay, such as an SNP array-based assay or a sequencing-based assay. In some cases, a system including a signal detector and a computer can be used to collect data (e.g., fluorescent signals or sequencing results) regarding homozygosity or heterozygosity of multiple loci (e.g., sample-specific normalized signal intensities of both alleles at each locus). In box 1510, an allele-specific copy number (ASCN) is reconstructed at each locus (e.g., each SNP). The ASCN is the copy number of both the paternal and maternal alleles. In box 1530, a likelihood function is used to determine whether a homozygous locus or region of a homozygous locus is due to LOH. This may be conceptually similar to previously described algorithms designed to reconstruct the total copy number (not the ASCN) at each locus (e.g., SNP). See International Application No. PCT / US2011 / 026098 to Abkevich et al.The likelihood function can be maximized over the ASCN of all loci, the level of contamination by benign tissue, the total copy number averaged over the whole genome, and the sample-specific noise level. At box 1540, the LOH region is determined as the SNP region where one of the ASCNs (paternal or maternal) is 0. In some embodiments, the computer process further comprises the step of querying whether the patient is untreated or the step of determining whether the patient is untreated.

[0093] FIG. 3 shows an exemplary process by which a computing system can determine the presence or absence of an LOH signature. The process starts at box 300. In box 300, data on homozygosity or heterozygosity of a plurality of loci along a chromosome is collected by a computing system. As described herein, any suitable assay method, such as an SNP array-based assay method or a sequencing-based assay method, can be used to evaluate the homozygosity or heterozygosity of the loci along a chromosome. In some cases, a system comprising a signal detector and a computer can be used to collect data (e.g., fluorescent signals or sequencing results) on the homozygosity or heterozygosity of the plurality of loci. In box 310, the data on homozygosity or heterozygosity of the plurality of loci and the position or spatial relationship of each locus are evaluated by the computing system to determine the length of any LOH regions present along the chromosome. In box 320, the data regarding the number of detected LOH regions and the length of each detected LOH region is evaluated by the computing system to determine the number of LOH regions that (a) are greater than or equal to a pre-set number of Mb (e.g., 15 Mb) and (b) have a length less than the full length of the chromosome containing the LOH region. Alternatively, the computing system can determine the full length of the LOH or determine the sum of the LOH lengths as described above. In box 330, the computing system formats an output indicating the presence or absence of the LOH signature. Once formatted, the computing system can present the output to a user (e.g., a laboratory technician, clinician, or medical professional). As described herein, the presence or absence of the LOH signature can be used to indicate the likely HDR status of the patient, to indicate the likely presence or absence of genetic mutations in HDR pathway genes, and / or to indicate a possible cancer treatment regimen.

[0094] 4 is a diagram of an example of a computing device 1400 and a mobile computing device 1450 that may be used with the techniques described herein. The computing device 1400 is intended to be various types of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The computing device 1450 is intended to be various types of mobile devices, such as personal digital assistants, mobile phones, smartphones, and other similar computing devices. The components shown here, their connections and relationships, and their functions are intended to be merely exemplary and are not intended to limit the practice of the invention described and / or claimed in this document.

[0095] Computing device 1400 includes processor 1402, memory 1404, storage device 1406, high-speed interface 1408 connecting memory 1404 and high-speed expansion port 1410, and low-speed interface 1415 connecting low-speed bus 1414 and storage device 1406. Each of components 1402, 1404, 1406, 1408, 1410, and 1415 may be interconnected using various buses and mounted on a common motherboard or in other manners as appropriate. Processor 1402 can process instructions for execution within computing device 1400, including instructions stored in memory 1404 or storage device 1406, to display external input / output devices, such as graphical information for a GUI on display 1416 in communication with high-speed interface 1408. In other implementations, multiple processors and / or multiple buses may be used, along with multiple memories and types of memories, as appropriate. Also, multiple computing devices 1400 may be connected, each providing a portion of the required operations (eg, as a server bank, a collection of blade servers, or a multi-processor system).

[0096] The memory 1404 stores information within the computing device 1400. In one implementation, the memory 1404 is a volatile memory unit. In another implementation, the memory 1404 is a non-volatile memory unit. The memory 1404 may also be another type of computer readable medium, such as a magnetic disk or an optical disk.

[0097] The storage device 1406 can provide mass storage for the computing device 1400. In one implementation, the storage device 1406 can be or comprise a number of different devices, including computer-readable media, such as a floppy disk drive, a hard disk drive, an optical disk drive, or a tape drive, a flash memory or other similar solid-state memory device, or a device in a storage area network or other configuration. The computer program product can be tangibly embodied in an information carrier. The computer program product can also include instructions that, when executed, perform one or more methods, such as the methods described herein. The information carrier is a computer-readable or machine-readable medium, such as the memory 1404, the storage device 1406, a memory installed in the processor 1402, or a propagating signal.

[0098] The high-speed controller 1408 manages bandwidth-intensive operations for the computing device 1400, while the low-speed controller 1415 manages less bandwidth-intensive operations. This allocation of functions is merely exemplary. In one implementation, the high-speed controller 1408 is coupled to the memory 1404, the display 1416 (e.g., via a graphics processor or accelerator), and a high-speed expansion port 1410, which may accept various expansion cards (not shown). In this implementation, the low-speed controller 1415 is coupled to the storage device 1406 and the low-speed expansion port 1414. The low-speed expansion port may comprise various communication ports (e.g., USB, Bluetooth, Ethernet, or wireless Ethernet) and may be coupled, for example, via a network adapter, to one or more input / output devices, such as a keyboard, a pointing device, a scanner, an optical reader, a fluorescent signal detector, or a networking device, such as a switch or a router.

[0099] As shown in the figure, the computing device 1400 can be implemented in many different ways. For example, the computing device 1400 can be implemented as a standard server 1420, or multiple times in a collection of such servers. The computing device 1400 can be implemented as part of a rack server system 1424. Additionally, the computing device 1400 can be implemented in a personal computer, such as a laptop computer 1422. Or, components from the computing device 1400 can be combined with other components, such as device 1450, in a mobile device (not shown). Each such device may comprise one or more of the computing devices 1400, 1450, or the entire system may consist of multiple computing devices 1400, 1450 in communication with each other.

[0100] Computing device 1450 includes a processor 1452, memory 1464, input / output devices such as a display 1454, a communication interface 1466, and a transceiver 1468, among other components (e.g., scanner, optical reader, fluorescent signal detector). Device 1450 may also be provided with a storage device such as a microdrive or other device to provide additional storage. Components 1450, 1452, 1464, 1454, 1466, and 1468 are each interconnected using various buses, and some of the components may be mounted on a common motherboard or in other manners as appropriate.

[0101] The processor 1452 can execute instructions within the computing device 1450, including instructions stored in the memory 1464. The processor may be implemented as a chipset of chips with separate and multiple analog and digital processors. The processor may provide, for example, control of a user interface, applications run by the device 1450, and wireless communication by the device 1450, for example, to coordinate other components of the device 1450.

[0102] The processor 1452 can communicate with a user via a control interface 1458 and a display interface 1456 coupled to a display 1454. The display 1454 can be, for example, a TFTLCD (thin film transistor liquid crystal display) or an OLED (organic light emitting diode) display or other suitable display technology. The display interface 1456 can include suitable circuitry for driving the display 1454 to present graphical and other information to a user. The control interface 1458 can receive commands from a user and translate the commands for transmission to the processor 1452. Additionally, an external interface 1462 can communicate with the processor 1452 to enable near area communication between the device 1450 and other devices. The external interface 1462 can, for example, provide for wired communication in some implementations and wireless communication in other implementations, and multiple interfaces can be used.

[0103] The memory 1464 stores information within the computing device 1450. The memory 1464 may be implemented as one or more of a computer readable medium, a volatile memory unit, or a non-volatile memory unit. An expansion memory 1474 may be provided and connected to the device 1450 via an expansion interface 1472. The expansion interface 1472 may include, for example, a SIMM (single in-line memory module) card interface. Such expansion memory 1474 may provide extra storage space to the device 1450 and may store applications or other information for the device 1450. For example, the expansion memory 1474 may include instructions that perform or supplement the processes described herein and may also include secure information. Thus, for example, the expansion memory 1474 may be provided as a security module for the device 1450 and may be programmed with instructions that enable secure use of the device 1450. Additionally, secure applications may be provided via the SIMM card along with additional information, such as placing identifying information on the SIMM card so that it cannot be hacked.

[0104] The memory may include, for example, flash memory and / or NVRAM memory, as discussed below. In one implementation, the computer program product is tangibly embodied in an information carrier. The computer program product includes instructions that, when executed, perform one or more methods, such as the methods described herein. The information carrier is a computer-readable or machine-readable medium, such as memory 1464, expansion memory 1474, memory on-board the processor 1452, or a propagated signal that may be received, such as a propagated signal that may be received by the transceiver 1468 or the external interface 1462.

[0105] The device 1450 may communicate wirelessly via a communication interface 1466. The communication interface 1466 may include digital signal processing circuitry, if necessary. The communication interface 1466 may provide communications under various modes or protocols, such as GSM voice calls, SMS, EMS, or MMS messaging, CDMA, TDMA, PDC, WCDMA, CDMA2000, or GPRS, among others. Such communications may occur, for example, via a radio frequency transceiver 1468. Additionally, short-range communications may occur, for example, using Bluetooth, WiFi, or other such transceivers (not shown). Additionally, a GPS (Global Positioning System) receiver module 1470 may provide further navigation-related and location-related wireless data to the device 1450, which may be used by applications operating on the device 1450, as appropriate.

[0106] Device 1450 may also communicate audibly using audio codec 1460. Audio codec 1460 may receive voice information from a user and convert it into usable digital information. Additionally, audio codec 1460 may generate audible sounds to a user, for example, at a handset of device 1450, for example, via a speaker. Such sounds may include sounds from voice telephone calls, may include recorded sounds (e.g., voice mail, music files, etc.), and may also include sounds created by applications running on device 1450.

[0107] As shown in the figure, computing device 1450 may be implemented in a number of different ways. For example, computing device 1450 may be implemented as a mobile phone 1480. Computing device 1450 may also be implemented as part of a smartphone 1482, personal digital assistant, or other similar mobile device.

[0108] Various implementations of the systems and techniques described herein can be realized in digital electronic circuitry, integrated circuits, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include execution in one or more computer programs executable and / or interpretable on a programmable system including at least one programmable processor, which may be special purpose or general purpose, coupled to receive data and instructions from a storage system, at least one input device, and at least one output device, and to transmit data and instructions to the storage system, at least one input device, and at least one output device.

[0109] These computer programs (also known as programs, software, software applications, or code) contain machine instructions for a programmable processor and may be executed in high-level procedural and / or object-oriented programming languages ​​and / or assembly / machine languages. As used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal (e.g., magnetic disks, optical disks, memory, and programmable logic devices (PLDs)). The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0110] The systems and techniques described herein for providing user interaction can be executed on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user, as well as a keyboard and pointing device (e.g., a mouse or trackball) to allow the user to provide input to the computer. Other types of devices can also be used to provide user interaction. For example, feedback provided to the user can be any type of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback). Input from the user can be received in any form, including acoustic input, speech input, or tactile input.

[0111] The systems and techniques described herein can be implemented in a computing system that includes a back-end component (e.g., as a data server), or includes a middleware component (e.g., an application server), or includes a front-end component (e.g., a client computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communications network). Examples of communications networks include a local area network ("LAN"), a wide area network ("WAN"), and the Internet.

[0112] A computing system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0113] In some cases, the computing system provided herein may be configured to include one or more sample analyzers. The sample analyzer may be configured to generate a plurality of signals for genomic DNA of at least one pair of human chromosomes from cancer cells. For example, the sample analyzer may generate a signal that can be interpreted in a manner that identifies homozygosity or heterozygosity of loci along the chromosome. In some cases, the sample analyzer may be configured to perform one or more steps of a SNP array-based assay or a sequencing-based assay, and may be configured to generate and / or capture signals from such assays. In some cases, the computing system provided herein may be configured to include a computing device. In such cases, the computing device may be configured to receive signals from the sample analyzer. The computing device may include computer executable instructions or a computer program (e.g., software) that includes computer executable instructions for performing one or more of the methods or steps described herein. In some cases, such computer executable instructions can instruct the computing device to analyze signals from a sample analyzer, another computing device, a SNP array-based assay, or a sequencing-based assay.Such signals can be analyzed to determine genotype, homozygosity at a particular locus, regions of homozygosity, number of LOH regions, determine the size of the LOH regions, determine the number of LOH regions of a particular size or size range, determine whether a sample is LOH signature positive, determine the number of indicator LOH regions in at least a pair of human chromosomes, determine the likelihood of defects in the BRCA1 and / or BRCA2 genes, determine the likelihood of defects in HDR, determine the likelihood that a cancer patient will respond to a particular cancer treatment regimen (e.g., a regimen including a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, a PARP inhibitor, or a combination thereof), or a combination of these items.

[0114] In some cases, the computational system provided herein may comprise computer executable instructions or computer programs (e.g., software) for formatting an output that indicates the number of LOH regions, the size of LOH regions, the number of LOH regions with a certain size or size range, whether a sample is LOH signature positive, the number of indicator LOH regions in at least a pair of human chromosomes, the likelihood of defects in BRCA1 and / or BRCA2 genes, the likelihood of defects in HDR, the likelihood that a cancer patient will respond to a certain cancer treatment regimen (e.g., a regimen that includes DNA damaging agents, anthracyclines, topoisomerase I inhibitors, radiation, PARP inhibitors, or a combination thereof), or a combination of these items.In some cases, the computational system provided herein may comprise computer executable instructions or computer programs (e.g., software) for determining the cancer treatment regimen that is desirable for a certain patient based at least in part on the presence or absence of LOH signature or the number of indicator LOH regions.

[0115] In some cases, the computational system provided herein may include a pre-processing device configured to process a sample (e.g., cancer cells) so that it can perform an SNP array-based assay or a sequencing-based assay. Examples of pre-processing devices include, but are not limited to, devices configured to enrich a cell population for cancer cells as opposed to non-cancerous cells, devices configured to lyse cells and / or extract genomic nucleic acid, and devices configured to enrich a sample for specific genomic DNA fragments.

[0116] This document also provides a kit for evaluating a sample (e.g., cancer cells) as described herein. For example, this document provides a kit for evaluating cancer cells for the presence of LOH signatures or for determining the number of indicator LOH regions in at least a pair of human chromosomes. The kit provided herein may include either SNP probes (e.g., an array of SNP probes for carrying out the SNP array-based assay method described herein) or primers (e.g., primers designed to sequence SNP regions via sequencing-based assay method) in combination with a computer program product that includes computer executable instructions for carrying out one or more of the methods or steps described herein (e.g., computer executable instructions for determining the number of LOH regions with a particular size or size range). In some cases, the kit provided herein may include at least 500, 1000, 10,000, 25,000, or 50,000 SNP probes that can hybridize to polymorphic regions of human genomic DNA. In some cases, the kits provided herein may include at least 500, 1000, 10,000, 25,000, or 50,000 primers capable of sequencing polymorphic regions of human genomic DNA. In some cases, the kits provided herein may include one or more other components for performing SNP array-based or sequencing-based assays. Examples of such other components include, but are not limited to, buffers, sequencing nucleotides, enzymes (e.g., polymerases), and the like. This document also provides for the use of any suitable number of materials provided herein in the manufacture of a kit for performing one or more of the methods or steps described herein. For example, this document provides for the use of a collection of SNP probes (e.g., a collection of 10,000 to 100,000 SNP probes) and a computer program product provided herein in the manufacture of a kit for evaluating cancer cells for the presence of a LOH signature.As another example, this document provides the use of a collection of primers (e.g., a collection of 10,000-100,000 primers for sequencing SNP regions) and the computer program product provided herein in the manufacture of a kit for evaluating cancer cells for the presence of an LOH signature.

[0117] The invention is further described in the following examples, which do not limit the scope of the invention described in the claims. EXAMPLES

[0118] Example 1 - Evaluation of LOH regions and HDR Two sets of tumors from patients with advanced ovarian cancer were used. The first set of 94 tumors (training set) was used to derive candidate signatures, and the second set of 40 tumors (validation set) was used to validate the signatures. The entire coding region of the BRCA1 and BRCA2 genes was sequenced to detect germline and somatic mutations. BRCA1 and BRCA2 mRNA expression levels were measured and Affymetrix SNP microarrays were performed.

[0119] A computer program was used to reconstruct the LOH signature state based on the allele intensities obtained from the microarray data. An algorithm was developed and implemented as a computer program to reconstruct LOH regions based on genotype (e.g., SNP genotype) data.

[0120] One feature of this algorithm was to first reconstruct the allele-specific copy number (ASCN) at each locus (e.g., SNP). The ASCN is the copy number of both the paternal and maternal alleles. LOH regions were then determined as SNP regions where one of the ASCNs (paternal or maternal) was 0. This algorithm was based on maximizing a likelihood function and was conceptually similar to previously described algorithms designed to reconstruct the total copy number (rather than the ASCN) at each locus (e.g., SNP). See International Application No. PCT / US2011 / 026098 to Abkevich et al. The likelihood function was maximized over the ASCN of all loci, the level of contamination with benign tissue, the total copy number averaged over the entire genome, and the sample-specific noise level. Input data for the algorithm included (1) sample-specific normalized signal intensities of both alleles at each locus and (2) an assay-specific (specific for different SNP arrays and sequence-based approaches) parameter set defined based on the analysis of a large number of samples with known ASCN profiles.

[0121] Tumors were defined as HDR-deficient for the purposes of this analysis if they had one or more deleterious mutations in the BRCA1 and / or BRCA2 genes or if they had low BRCA1 mRNA expression. The remaining tumors were defined as likely to be HDR-nondeficient for the purposes of this analysis.

[0122] The distribution of the length of LOH regions was examined (Figure 5). Three classifications of LOH regions were used: (1) LOH affecting the entire chromosome; (2) large LOH regions (longer than about 15 Mb), typically affecting part of a chromosome arm or the entire chromosome arm; and (3) multiple short LOH regions (shorter than about 15 Mb). Next, using only the training set, the number of LOH regions in one of these three classifications was evaluated for possible correlation with HDR loss. It was found that (1) the number of short LOH regions did not significantly correlate with HDR loss (p>0.05); (2) LOH covering the entire chromosome was weakly correlated with HDR loss (p=0.0011); and (3) the number of large LOH regions was significantly correlated with HDR loss (p=1.9e-8). More specifically, it was found that all HDR-deficient tumors had a large number of large LOH regions (e.g., 9 or more), whereas the majority of tumors likely to be HDR-nondeficient had a small number of large LOH regions (Figures 6-8). Tumors likely to be HDR-nondeficient were in fact HDR-deficient due to other genetic alterations except for BRCA1 and BRCA2 mutations and low mRNA expression. In addition to the number of large LOH regions, the total length of these regions was also significantly correlated with HDR deficiency.

[0123] Using the validation set, these results were confirmed: (1) the number of short LOH regions was not significantly correlated with HDR loss (p>0.05); (2) LOH covering the entire chromosome was weakly correlated with HDR loss (p=0.05); and (3) the number of large LOH regions was significantly correlated with HDR loss (p=3.9e-6).

[0124] The 134 tumors were divided into three groups from the combined training and validation datasets: (1) BRCA-deficient if they had one or more deleterious mutations in the BRCA1 and / or BRCA2 genes or low BRCA1 mRNA expression; (2) HDR-deficient / BRCA-intact if they had 9 or more large LOH regions (longer than 15 Mb but shorter than the length of an entire chromosome); (3) HDR-intact if they had less than 9 large LOH regions (longer than 15 Mb but shorter than the length of an entire chromosome). The results of this analysis are shown in Figure 9. Figure 9 shows that BRCA-deficient as well as HDR-deficient not due to BRCA-deficient are frequent among ovarian tumors.

[0125] Figure 10 shows the distribution of large LOH regions (longer than 15 Mb but shorter than the length of the entire chromosome) for different types of cancer cell lines. The size of the circle is proportional to the number of samples with such a number of large LOH regions. The frequency of HDR deficiency (cell lines with at least 9 of such large LOH regions) is highest among breast and esophageal cancer cell lines. No HDR deficiency was observed among colon cancer cell lines. Validating previous findings in ovarian tumors, all BRCA-deficient cell lines were also found to be HDR deficient.

[0126] Figure 11 shows the distribution of large LOH regions (longer than 15 Mb but shorter than the length of an entire chromosome) in a publicly available lung tumor dataset (GSE19399 from Gene Expression Omnibus). A very high frequency of HDR loss (defined as having at least 9 large LOH regions) was observed among lung tumors (39%).

[0127] Figure 12 summarizes the results of the analysis of various tumors and cell lines. For several tumors and cell lines, the frequency of HDR loss, defined as the percentage of samples with at least nine large LOH regions (longer than 15 Mb but shorter than the length of the entire chromosome), is shown. This frequency was as high as 50% in ovarian tumors and was not observed at all in brain and colon cell lines. Thus, HDR loss appears to play an important role in the majority of cancers.

[0128] Example 2 - Chemical Toxicity Reactions In preparation for chemical toxicity reaction experiments, all cell lines were incubated at 75 cm 2 Culture in tissue culture flasks (VWR International, Inc. Cat. No. 353136) and recommended growth medium at 37°C + 5% CO 2 Prior to each experiment, each cell line was trypsinized (Invitrogen Corporation Catalog No. 25200-056), counted, and plated at 2500 cells or 5000 cells in 100 μL of medium per well in columns 2-12 of a 96-well clear-bottom polystyrene microplate (Perkin Elmer Catalog No. 600518) in Advanced RPMI 1640 (Invitrogen Corporation Catalog No. 12633-020), 3% FBS, 1% penicillin / streptomycin (Invitrogen Corporation Catalog No. 15140-122). Column 1 contained 100 μL of medium only per well. The plate with the cells was then incubated at 37°C + 5% CO 2 The mixture was incubated overnight at 4°C.

[0129] Two different final drug concentration working stocks were prepared. Advanced RPMI 1640 was used as the diluent for the highest concentration when 100% DMSO was required for drug dissolution. The lower concentration used Advanced RPMI 1640 + a pre-determined amount of DMSO equal to the total DMSO in the high concentration working stock. The lowest concentration used up to 60% DMSO. This was done to keep the DMSO concentration per well equal and prevent non-specific cell death as a result of DMSO. The lower of the two drug concentrations was placed in rows A-D, column 12 of a 96-well thin-walled PCR cycle plate (Robbins Scientific catalogue no. 1055-00-0), whereas the higher concentration was placed in rows E-H, column 12 of the same plate. Serial dilutions of 1:2 or 1:3 were performed in descending order in columns 12-3. Columns 1 and 2 were used for no cells / no drug and no drug controls. This allowed quadruplet data points for each drug concentration. Once the dilutions were complete, 5 μL was transferred from the dilution plate to the corresponding wells of the seeded cell plate. The drug-containing plates were then incubated at 37° C. + 5% CO 2 The mixture was incubated at RT for 3 or 6 days.

[0130] After the 3-day or 6-day dosing regimen, the ATPlite assay (Perkin Elmer Cat. No. 6016941) was performed on each well of each plate following the ATPLite assay protocol. Luminescence was then read on the FUSION instrument and saved as a .CSV file. For each cell line and drug combination, the four replicates of the no drug control were averaged and divided by 100 to create a "normalization factor" used to calculate normalized percent survival. The normalized percent survival of the no drug control was 100%. For each drug concentration, the four replicates of the cell + drug wells were averaged and divided by the normalization factor. The percent survival for each drug concentration, starting with a concentration equal to 0, was used to calculate the IC using proprietary software. 50was calculated.

[0131] Figure 13 shows the response of breast and ovarian cancer cell lines to chemotherapy. The y-axis shows the Log of various chemotherapy drugs (camptothecin, and the mean of the results for platinum compounds (oxaliplatin, cisplatin, and carboplatin) or anthracyclines (doxorubicin and epirubicin)) upon exposure to 29 breast cancer cell lines. 10 (I C 50 ) values, as well as the Log 10 (I C 50 On the x-axis, the number of large LOH regions longer than 15 Mb and shorter than the entire chromosome for these cell lines is displayed. The threshold of 9 is indicated by the dashed line.

[0132] Figure 14 is a version of the graph from Figure 13 showing the specificity and sensitivity among responders and non-responders to treatment with platinum compounds (oxaliplatin, cisplatin, and carboplatin) upon exposure to 29 breast cancer cell lines. The dashed line indicates the threshold of 9 for the number of large LOH regions longer than 15 Mb and shorter than the entire chromosome. The solid line separates the cell lines into responders and non-responders.

[0133] Example 3 - Further validation of the HR deficiency assay material and method Ovarian tumor samples Three independent human ovarian cancer cohorts were used: 1: 152 unselected ovarian cancer samples; 2: 53 high-grade serous ovarian tumors; 3: Publicly available data from 435 serous ovarian cancer samples with complete information available were downloaded from The Cancer Genome Atlas (TCGA) Network website on October 31, 2011. All cohorts were obtained under protocols approved by the Institutional Review Board (IRB). Patient and tumor characteristics are shown in Table 2. A variable number of samples was used in the described assay (Table 3).

[0134] Table 2 Patient and cancer characteristics TIFF2025032160000003.tif176156TIFF2025032160000004.tif175156

[0135] Table 3. Number of samples used in each assay TIFF2025032160000005.tif126161

[0136] cell line Sixty-seven cancer cell lines were analyzed (29 ovarian, 34 breast, 3 colon, and 1 pancreatic cancer cell line). Three breast cancer cell lines were obtained from DSMZ (Braunschweig, Germany). Colon, pancreatic, and the remaining breast cancer cell lines were obtained from ATCC (Manassas, VA). Cancer cell lines were plated in T75 flasks and cultured at approximately 5 × 10 cells / ml at 37 °C in RPMI + 10% FBS + 1% penicillin / streptomycin medium. 6 Cells were grown to cell density. Exceptions were cell lines requiring non-standard media, L-glutamine, or insulin. Cells grown in suspension were centrifuged in 1.5 mL centrifuge tubes for 5 min at 1700 rpm, and the supernatant was discarded. Cells grown in monolayer were aspirated to remove medium, washed with PBS, and trypsin solution was added. Cells were detached, collected in medium, transferred to 1.5 mL microcentrifuge tubes, and centrifuged at 1700 rpm for 5 min. The supernatant was discarded. Isolated cells were resuspended in 200 μL PBS.

[0137] Extraction of genomic DNA and total RNA from frozen tumors and cell lines 10 μm frozen sections were cut and microdissected. Tissues were homogenized (TissueRuptor, Qiagen) after addition of QIAzol lysis reagent, followed by RNA isolation using the Qiagen miRNAeasy Mini Kit according to the manufacturer's protocol. DNA was isolated using the QIAamp DNA Mini Kit (Qiagen) following overnight lysis incubation at 56°C and RNaseA treatment according to the manufacturer's protocol.

[0138] BRCA1 and BRCA2 sequencing BRCA1 and BRCA2 were sequenced as described in Hennessy at al., 2010. Identified mutations were included in the analysis only if they were classified as deleterious or suspected to be deleterious based on previously described criteria ( Beaudet and Tsui, 1993 ).

[0139] Promoter methylation qPCR assay Methylation levels were quantified using the Methyl-Profiler DNA Methylation PCR Array System (SABiosciences) according to the manufacturer's recommended protocol. DNA methylation-sensitive and -dependent restriction enzymes were used to selectively digest unmethylated or methylated genomic DNA, respectively. Post-digestion DNA was quantified by real-time PCR using primers flanking the regions of interest. Relative concentrations of differentially methylated DNA were determined by comparing the amount of each digest to that of a mock digest.

[0140] BRCA1 promoter methylation sequencing assay 50-300ng of DNA was incubated at 60°C for approximately 5 hours in the presence of bisulfite under acidic conditions and briefly ramped to 95°C. After incubation, the reaction was bound to a spin column, washed under basic conditions to remove bisulfite, and the converted DNA was then eluted in 15μL. The lowercase region of the primer is specific to the genomic region being amplified. The uppercase region of the primer corresponds to the 454 Titanium chemistry tail and the 4bp barcode (the last 4 bases before the region-specific bases). By combining forward and reverse primers in multiple combinations, up to 100 samples can be multiplexed in a single sequencing reaction.

[0141] BRCA1 and cell cycle progression signature expression assays RNA was treated with Amplification Grade Deoxyribonuclease I (Sigma-Aldrich Inc.) with an extended incubation time of 30 min according to the manufacturer's protocol. Reverse transcription was performed using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions.

[0142] Replicate preamplifications were performed independently in 5 μL reaction volumes using the Taqman PreAmp Master Mix Kit (Applied Biosystems) protocol. For the cell cycle gene assays, preamplification repeats were performed for 8 and 18 cycles, respectively. Only the BRCA1 assay had three preamplification repeats for 18 cycles. Postamplification products were diluted 1:5 in low EDTA Tris-EDTA (TE). Quantitative polymerase chain reaction (qPCR) was then performed and evaluated on Gene Expression M48 Dynamic Arrays (Fluidigm, South San Francisco, CA) according to the manufacturer's protocol. Comparative cycle threshold (C T Relative gene expression was calculated using the ) method. C T s was centered by the average of the genes in the replicate that were common to all replicates. The resulting values ​​were first multiplied by the average C T s, then the normalized C for each assay for all samples from the first cohort T Normalized by the average of s, ΔΔC T The CCP scores and relative BRCA1 expression were calculated using the ΔΔC T It was calculated as the average of the negative values ​​of s.

[0143] Identifying samples with lost BRCA1 expression: Samples with anti-correlated CCP and BRCA1 expression were defined as BRCA1-deficient. A threshold for identifying patients with abnormal BRCA1 expression was defined using robust linear regression in a large set of ovarian cancer samples (n=300). BRCA1 expression was regressed against CCP score using iteratively weighted least squares (IWLS). Points outside the lower end of the 99% prediction interval were considered abnormal. This method is described in further detail in International Application No. PCT / US2011 / 054369 to Timms et al.

[0144] Affymetrix 500K GeneChip Array Affymetrix GeneChip Mapping NspI or StyI Assay Kit was used to generate biotinylated DNA for Affymetrix Mapping 500K NspI or StyI microarray hybridization (each assay was prepared separately). Genomic DNA (250 ng) was digested with NspI or StyI restriction enzyme, and adapters were added to the restriction fragment ends using T4 DNA ligase. The adapter-modified samples were PCR amplified using Clontech Titanium Taq. This generated amplification products with an average size of 200-1,100 bp. The amplification products were purified using a Clontech DNA Amplification and Purification Kit. 90 μg of purified DNA was fragmented using Affymetrix Fragmentation Reagent. The fragmented samples were biotinylated using GeneChip DNA Labeling Reagent. Biotin-labeled DNA was hybridized on NspI or StyI Affymetrix microarrays for 16-18 hours at 49°C in an Affymetrix rotation oven. After hybridization, probe array washing and staining procedures were performed on automated Affymetrix Fluidics Stations according to the manufacturer's manual. Microarrays were scanned and raw data were collected by an Affymetrix GeneChip Scanner 3000.

[0145] CN and LOH analysis of SNP microarray data This algorithm is designed to determine the most likely allele-specific copy number at each SNP position. The corresponding probability explicitly takes into account contamination of cancer DNA samples with non-cancer stromal cell DNA. A similar algorithm for CN analysis is detailed in International Application No. PCT / US2011 / 026098 (Publication No. WO / 2011 / 106541) to Abkevich et al. The algorithm used in this document was implemented in two versions, one for the analysis of internally generated Affymetrix 500K GeneChip array data, and the other for the analysis of GenomeWideSNP6 Affymetrix array data downloaded from the TCGA website (http: / / tcga-data.nci.nih.gov / tcga / dataAccessMatrix.htm?diseaseType=OV). In addition to the SNP probes, the latter array contains a large number of probes for non-polymorphic positions throughout the human genome. These probes are useful for CN analysis, but not directly useful for LOH analysis.

[0146] statistical analysis P values ​​in this publication were calculated using the Kolmogorov-Smirnov test unless otherwise specified.

[0147] result HR-deficient tumors Tumor samples were considered HR deficient if they had germline or somatic mutations in BRCA1 or BRCA2 or methylation or low mRNA expression of BRCA1. Thirty-one of 152 samples from the first cohort were carriers of BRCA1 and / or BRCA2 mutations, along with 14 of 53 samples from the second cohort and 83 of 435 samples from the third cohort (two of which were excluded from further analysis, see below). The mutations are summarized in Table 4.

[0148] Table 4. BRCA1, BRCA2, and RAD51C abnormalities detected in the study cohort TIFF2025032160000006.tif53158 1 Two of these mutations were excluded from the analysis because one copy of BRCA2 remained intact.

[0149] Methylation was measured in the promoter CpG islands of both BRCA1 and BRCA2. Methylation was observed in multiple samples for BRCA1 but not for BRCA2. Eleven of 126 samples from the first cohort, three of 34 from the second cohort, and 64 of 435 from the third cohort were defined as HR deficient with high levels of BRCA1 promoter methylation. No deleterious BRCA1 or BRCA2 mutations were observed in any of these samples, except one from the third cohort.

[0150] Low mRNA expression of BRCA1 or BRCA2 may also lead to HR deficiency and may be the result of mechanisms other than promoter methylation. BRCA1 and BRCA2 expression levels were measured in 137 samples from the first cohort and 53 samples from the second cohort. BRCA1 expression was abnormally low in 20 samples. Only five samples with abnormally low BRCA1 expression were flagged as HR deficient by BRCA1 promoter methylation. No abnormally low expression was observed for BRCA2.

[0151] One intact copy of BRCA1 or BRCA2 is required for function. For all BRCA1-deficient samples, the BRCA1 gene is contained within a region of LOH. Furthermore, for all but two BRCA2-deficient samples, the BRCA2 gene is observed within a region of LOH. These two BRCA2-deficient samples were not considered HR-deficient in our analysis.

[0152] Distribution of LOH region length The initial hypothesis was that regions with LOH of different lengths may appear in cancer genomes through different pathways, and thus the relationship between LOH and HR defects may depend on the length of the LOH region. The distribution of the length of the LOH regions corrected for the length of the chromosome arm on which these LOH regions are observed is shown in Figure 16. Chromosomes 13, 14, 15, and 22 were excluded because no SNPs are available on the short arms of these chromosomes. Three peculiar features were observed in this distribution. First, there are many short LOH regions. Second, there is a long, flat LOH region tail that is up to the length of one chromosome arm. There are very few LOH regions that are longer than one chromosome arm but shorter than the entire chromosome. Finally, there is a high peak that corresponds to LOH across the entire chromosome. The observed distribution is quite different from the similar distribution obtained for CN variation (Beroukhim et al. 2010). This suggests that CN variation and LOH regions may arise via different mechanisms.

[0153] Correlation between samples with HR deficiency and samples with LOH The first cohort of samples was used as the "discovery" cohort. Since LOH was observed across chromosome 17 in almost all samples, LOH regions on this chromosome were excluded from the analysis, presumably because genes important for ovarian cancer progression are thought to reside on this chromosome. We investigated the correlation between HR defects and the number of short LOH regions (<15 Mb), the number of long LOH regions (>15 Mb but shorter than the entire chromosome), and the number of LOH regions covering the entire chromosome. A variety of different LOH region length cutoffs can be used. The impact of this cutoff on the detection of HR defects was investigated in Figure 19 and its accompanying discussion, and 15 Mb was found to be generally preferred. There was no significant correlation between the number of short LOH regions and HR defects. The number of LOH regions covering the entire chromosome was significantly higher in tumors with intact BRCA1 or BRCA2 (p=4×10 -5The number of long LOH regions (referred to as the "HRD score" following this Example 3 and throughout this document) was significantly higher in tumors with deleted BRCA1 or deleted BRCA2 (p=9×10 -11 ) (Figure 17a).

[0154] The second and third cohorts were used to validate the results obtained for the first cohort. The correlation between HR loss and the number of LOH regions covering the entire chromosome was not verified in the second cohort, probably due to the small number of samples, but was significantly greater among tumors with intact BRCA1 and BRCA2 in the third cohort (p=3×10 -11 A strong and significant correlation was observed between HRD score and HR defect for both cohorts (p = 2 × 10 -7 and p = 9 × 10 -30 ), whereas HRD scores were significantly lower among ovarian tumors with intact BRCA1 and BRCA2 (FIGS. 17b and 17c).

[0155] Alterations in RAD51C and other HR pathway genes Available data suggest that BRCA1 and BRCA2 are the main genes responsible for HR defects in ovarian cancer. However, many other genes may also be important, for example, RAD51C (Meindl et al., 2010) and RAD51D (Loveday et al., 2011) have recently been implicated as predisposing genes for ovarian cancer. In the first cohort, the promoter CpG islands of eight additional genes involved in the HR pathway were targeted for methylation measurement (Table 5). Only RAD51C had high levels of promoter methylation (3 out of 89 samples). In the third cohort, 11 out of 435 samples had RAD51C promoter methylation. All RAD51C methylation-positive samples from both cohorts were homozygous at the RAD51C locus due to LOH. To test whether HRD scores were higher in samples with RAD51C promoter methylation, these samples from both cohorts were compared with BRCA intact samples lacking RAD51C methylation. Consistent with our observations for the BRCA1 and BRCA2 genes, HRD scores were significantly higher among samples with RAD51C methylation (p=0.0003).

[0156] Table 5. Promoter methylation assay used (SABiosciences) TIFF2025032160000007.tif94158

[0157] In a third cohort, deleterious mutations and methylation of HR pathway genes have been reported (TCGA, 2011). Mutations were investigated and the analysis was limited to aberrations with a high probability of being deleterious (e.g., nonsense and frameshift mutations). This resulted in a total of eight deleterious mutations in six genes (ATM, ATR, FANCA, FANCD2, FANCM, and PALB2). Five additional samples had methylation of HR pathway genes. Loss of the second allele was only detected in one of 13 samples (FANCM nonsense mutation). Since inactivation of both alleles is required to eliminate tumor suppressor function, most of these 13 samples are expected to have intact HR. Not surprisingly, the majority of these samples did not have high HRD scores.

[0158] Analyzing Complex Data The correlation between HRD scores and HR deficiency (defined as BRCA1 deficiency, BRCA2 deficiency, or RAD51C deficiency) in all three cohorts is shown in Figure 17d. A highly significant association was observed (p=2×10 -54 ).

[0159] An important question is whether the distribution of HRD scores is the same for HR deficiencies due to different genomic loci. To answer this, the distribution of HRD scores for BRCA1-deficient, BRCA2-deficient, and RAD51C-deficient tumors was analyzed separately (Figure 21). A significant difference was observed (p=7×10 -5 ), the mean HRD score for BRCA1-deficient samples (16.1; SD = 4.3) was higher than for BRCA2-deficient samples (13.0; SD = 3.9). The differences in HRD scores between either BRCA1 or BRCA2 and RAD51C (14.5; SD = 5.1) were not significant.

[0160] Normal tissue was available from some samples from the first two cohorts and all samples from the third cohort. This was used to determine whether the BRCA1 and BRCA2 mutations were germline or somatic. There is no significant difference between somatic and germline in the distribution of HRD scores for either BRCA1 or BRCA2 defects (Figure 20).

[0161] HRD scores in BRCA1- and BRCA2-deficient cell lines Unselected breast (n=34) and ovarian (n=29) cell lines were obtained from multiple sources. Three additional colon and one pancreatic cell line obtained from NCI60 with published BRCA1 and BRCA2 status were analyzed. Of these 67 cell lines, seven had homozygous deleterious mutations or BRCA1 promoter methylation, two had homozygous mutations with apparent functional reversion, and six had heterozygous mutations. Figure 18a shows the distribution of HRD scores for these three mutant groups as well as for wild-type samples. The distribution of HRD scores is not significantly different between wild-type ovarian tumors and wild-type cancer cell lines. Among cancer cell lines with heterozygous mutations, the distribution of HRD scores is similar to wild-type cancer cell lines, possibly because cells become HR-deficient only when both copies of BRCA1 or BRCA2 become nonfunctional. For cancer cell lines with functional loss of both copies of either BRCA1 or BRCA2, we observed high HRD scores similar to those observed for ovarian tumors with defective BRCA1, BRCA2, or RAD51C genes. Cancer cell lines with BRCA1 and BRCA2 mutation reversions also had high HRD scores, supporting the initial hypothesis that HR defects result in irreversible LOH changes. The difference between the distribution of HRD scores in wild-type or heterozygous mutant cell lines and those in cell lines with homozygous mutations (with or without reversions) or BRCA1 promoter methylation was highly significant (p=10 -5Importantly, after removing ovarian cancer cell lines from the dataset, there is a significant correlation between HRD score and BRCA1 and BRCA2 deficiency (p=0.01), suggesting that the relationship between HRD score and HR deficiency is not limited to ovarian cancer.

[0162] Correlation of HR deficiency with overall survival (OS) and progression-free survival (PFS) For the third cohort, where patients with a high HRD score had improved survival, PFS (p=0.03) and OS (p=6×10 -5 ) was observed (Figure 18b). P values ​​were calculated using the Cox model. This result is consistent with and expands on previously reported data showing that germline mutations in BRCA1 and BRCA2 are associated with improved outcomes in ovarian cancer (Rubin et al., 1996, Boyd et al., 2000; Cass et al., 2003; Tan et al., 2008, Hennessy et al., 2010).

[0163] Consideration The HRD score was validated in two independent ovarian cancer datasets and reflected mutations causing HR deficiency also in breast and pancreatic cell lines.

[0164] Table 6. Mean HRD scores for BRCA1 and BRCA2-deficient and BRCA1 and BRCA2-intact tumors and corresponding p-values TIFF2025032160000008.tif127159

[0165] The size of the intermediate class of LOH, longer than 15 Mb but shorter than the whole chromosome, correlated strongly and positively with defective HR genes, suggesting that most, but not all, of this type of LOH class exist because they incorporate double-stranded DNA breaks as part of their development and require repair by HR. In contrast, the frequency of LOH at the whole chromosome level was significantly lower in HR-deficient tumors. One possible explanation is that LOH at the whole chromosome level arises through another competing mechanism that does not involve double-stranded DNA breaks.

[0166] In addition to BRCA1 and BRCA2 aberrations, RAD51C promoter methylation was observed in ovarian tumors. High HRD scores were significantly associated with RAD51C deficiency in two data sets. In three data sets, only one additional HR-deficient tumor was identified, in which a FANCM nonsense mutation caused loss of the second allele by LOH. The HRD scores associated with FANCM mutations (8) are within the normal distribution for samples with high HRD scores.

[0167] Among tumors with apparently intact BRCA1, BRCA2, and RAD51C, a significant proportion of samples have high HRD scores. One possible explanation is that many of these samples have a significant proportion of abnormalities in other HR pathway genes. Another explanation is that contamination of tumors with normal tissue complicates the detection of abnormalities. The data suggest that the HRD score is less sensitive to contamination than other assays, and that undetected abnormalities may account for a significant proportion of high HRD score samples (see Supplementary Results).

[0168] Published studies have demonstrated that secondary reversions that restore BRCA2 function can occur in BRCA2-mutated cell lines after exposure to platinum agents (Sakai et al., 2009; Sakai et al., 2008; Edwards et al., 2008). Norquist et al., (2011) observed that approximately 28% of recurrent tumors harbor secondary mutations that restore BRCA function. Reversions were found primarily in individuals with prior exposure to platinum agents and were predictive of platinum resistance. The HRD score results from cumulative aberrations that occur in the tumor genome. DNA-based HR deficiency markers are functionally linked to HR deficiency and therefore likely to be strongly associated with HR deficiency. As a result, the HRD score is a highly robust measure of HR deficiency. However, its persistence means that the score is likely to be less sensitive to reversions. Post-treatment samples were not available from the tumors used in this study. However, data from cell lines are consistent with this hypothesis. Failure to detect reversions results in a false positive. This is likely to affect very few tumors in the neoadjuvant or adjuvant setting (Norquist et al., 2011) and is less important than false negatives that erroneously identify individuals as likely non-responders.

[0169] A high HRD score strongly correlates with HR deficiency. This score can be used to identify patients likely to respond to DNA damaging agents and PARP inhibitors (among other agents). Such tests have clear clinical utility in breast and ovarian cancer and can be used to extend the use of PARPi and platinum salts to other cancers where HR deficiency is less well characterized.

[0170] Example 4 - Further validation of the HR deficiency assay material and method The patient cohort analyzed in this example included 56 breast cancer patients. All of these breast cancer patients had BRCA mutation positive or triple negative breast cancer (most were TNBC). Stages I-III were included (most were II or III). Patients received 6 cycles of neoadjuvant gemcitabine + iniparib + carboplatin. Response was measured as a relatively low residual cancer burden after treatment.

[0171] Fifty-six fresh-frozen breast tumors were analyzed. The median contamination level was 60%. Nine samples were at least 90% contaminated. Eleven of these tumors were carriers of deleterious BRCA1 mutations and three were carriers of deleterious BRCA2 mutations. All of these tumors had LOH in the deleted gene. One of the carriers of a deleterious BRCA1 mutation also had a deleterious BRCA2 mutation. However, this sample did not have LOH in the BRCA2 gene.

[0172] Thirty samples were obtained from patients who responded to treatment (residual cancer burden 0 or 1). Thirteen of these were BRCA1 / 2 defective. Twenty-six samples were obtained from non-responders (residual cancer burden 2 or 3). One of these was BRCA1 defective. Genotyping analysis was performed by Affymetrix using Affymetrix MIP arrays (described in U.S. Pat. No. 6,858,412; U.S. Patent Application Publication No. US20060234264; Hardenbol et al., Nature Biotechnology (2003) 21: 673-678; Wang et al., BMC Med Genomics (2009) 2:8; each of which is incorporated herein by reference in its entirety). HRD scores were calculated as described above.

[0173] result The mean HRD score for responders was 16.5. The mean HRD scores for BRCA1 / 2 intact and BRCA1 / 2 defective responders were the same. The mean HRD score for non-responders was 11.4. The mean HRD score for BRCA1 / 2 intact non-responders was 11.6, and the mean HRD score for BRCA1 defective non-responders was 8. By Mann-Whitney U test, the p-value for the relationship between response to treatment and HRD score was 0.004. When the BRCA1 / 2 defective samples were excluded, the relationship between response to treatment and HRD score remained significant (p-value=0.02).

[0174] The difference in HRD scores was not significant between samples with residual cancer burden 0 and 1. Similarly, the difference in HRD scores was not significant between samples with residual cancer burden 2 and 3. The correlation between response to treatment and clinical parameters (stage, grade) was not significant.

[0175] Other Aspects Although the present invention has been described in conjunction with the detailed description of the invention, it should be understood that the foregoing description is for illustrative purposes only and is not intended to limit the scope of the invention, the scope of which is defined by the appended claims. Other aspects, advantages, and modifications are within the scope of the appended claims.

Claims

1. A method for assessing LOH in a cancer cell or genomic DNA of said cancer cell, comprising the steps of: (a) detecting in a cancer cell or in genomic DNA obtained from the cancer cell, a region of LOH in at least a pair of human chromosomes from the cancer cell, wherein the at least a pair of human chromosomes is not a human X / Y sex chromosome pair; and (b) determining the total number of LOH regions in the at least one pair of human chromosomes, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, and wherein the first length is about 1.5 megabases or more.

2. A method for predicting the status of BRCA1 and BRCA2 genes in cancer cells, comprising the steps of: determining, in said cancer cell, a total number of LOH regions in at least a pair of human chromosomes from said cancer cell, wherein said LOH regions are longer than a first length but shorter than the length of the entire chromosome containing said LOH regions, said at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and said first length is about 1.5 megabases or more; and Correlating said total number above a reference number with an increased likelihood of a defect in the BRCA1 or BRCA2 gene.

3. A method for predicting HDR status in cancer cells, comprising the steps of: determining, in said cancer cell, a total number of LOH regions in at least a pair of human chromosomes from said cancer cell, wherein said LOH regions are longer than a first length but shorter than the length of the entire chromosome containing said LOH regions, said at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and said first length is about 1.5 megabases or more; and Correlating said total number above a reference number with an increased likelihood of a defect in the HDR.

4. 1. A method for predicting a cancer patient's response to a cancer treatment regimen comprising a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, and / or a PARP inhibitor, comprising the steps of: determining, in cancer cells derived from the cancer patient, the number of LOH regions in at least one pair of human chromosomes from the cancer cells of the cancer patient, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and Correlating a total number above the reference number with an increased likelihood that the cancer patient may respond to the cancer treatment regimen.

5. 1. A method for predicting a cancer patient's response to a treatment regimen, comprising the steps of: determining, in cancer cells derived from the cancer patient, a total number of LOH regions in at least one pair of human chromosomes from the cancer cells of the cancer patient, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and Correlating said total number above a reference number with an increased likelihood that said cancer patient will not respond to a treatment regimen including paclitaxel or docetaxel.

6. 1. A method of treating cancer, comprising the steps of: (a) determining in a cancer cell derived from a cancer patient or in genomic DNA obtained from the cancer cell a total number of LOH regions in at least a pair of human chromosomes from the cancer cell, wherein the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more; and (b) administering to the cancer patient a cancer treatment regimen comprising one or more drugs selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors if the total number of LOH regions exceeds the reference number.

7. The use of one or more drugs selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors to manufacture a medicament useful for treating cancer in a patient identified as having cancer cells determined to have a total of five or more index LOH regions.

8. (a) a sample analyzer configured to generate a plurality of signals for genomic DNA of at least one pair of human chromosomes from a cancer cell; and (b) a computer subsystem programmed to calculate the number of indicator LOH regions in the at least one pair of human chromosomes based on the plurality of signals. A system for determining the LOH status of cancer cells of a cancer patient, comprising:

9. the computer subsystem compares the number of index LOH regions to a reference number; (a) a possible defect in the BRCA1 and / or BRCA2 genes in the cancer cells; (b) a possible defect in HDR in the cancer cell; or (c) the likelihood that the cancer patient will respond to a cancer treatment regimen that includes a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, or a PARP inhibitor. The system of claim 8, programmed to determine:

10. When running on a computer, detecting the presence or absence of any LOH regions along one or more of the human chromosomes other than the human X and Y sex chromosomes, wherein the length of the LOH region is about 1.5 megabases or more, but less than the length of the entire chromosome containing the LOH region; and determining the total number of said LOH regions in one or more chromosome pairs A computer program product embodied in a computer-readable medium, the computer program product performing steps including:

11. at least 500 oligonucleotides capable of hybridizing to multiple polymorphic regions of human genomic DNA; and A computer program product according to claim 10. A diagnostic kit comprising:

12. It is useful for determining the total number of index LOH regions in at least one chromosome pair from a human cancer cell obtained from a cancer patient, and (a) an increased likelihood of a defect in the BRCA1 or BRCA2 gene in the cancer cells; (b) an increased likelihood of a defect in HDR in the cancer cell; or (c) increasing the likelihood that the cancer patient will respond to a cancer treatment regimen that includes a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, radiation, or a PARP inhibitor. The use of a plurality of oligonucleotides capable of hybridizing to a plurality of polymorphic regions of human genomic DNA to prepare a diagnostic kit which is useful for detecting

13. The method of any one of claims 1 to 6, wherein the LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes.

14. The method of any one of claims 1 to 6, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells.

15. The method of any one of claims 1 to 6, wherein the total number of LOH regions is 9, 15, 20 or more.

16. 7. The method of any one of claims 1 to 6, wherein the first length is about 6 megabases, 12 megabases, or 15 megabases, or more.

17. 7. The method of any one of claims 1 to 6, wherein the reference number is 6, 7, 8, 9, 10, 11, 12, or 13 or more.

18. The use according to claim 7 or 12, wherein the indicator LOH regions are determined in at least 2, 5, 10 or 21 pairs of human chromosomes.

19. 13. The use according to claim 7 or 12, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells.

20. The use according to claim 7 or 12, wherein the total number of the index LOH regions is 9, 15, 20 or more.

21. The use according to claim 7 or 12, wherein the length of the indicator LOH region is about 6 megabases, 12 megabases, or 15 megabases or more.

22. 10. The system of claim 8 or 9, wherein the indicator LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes.

23. The system of claim 8 or 9, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells.

24. The system of claim 8 or 9, wherein the total number of index LOH regions is 9, 15, 20, or more.

25. The system of claim 8 or 9, wherein the length of the index LOH region is about 6 megabases, 12 megabases, or 15 megabases, or more.

26. 11. The computer program product of claim 10, wherein the indicator LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes.

27. 11. The computer program product of claim 10, wherein the cancer cell is an ovarian cancer cell, a breast cancer cell, or an esophageal cancer cell.

28. 11. The computer program product of claim 10, wherein the total number of index LOH regions is 9, 15, 20, or more.

29. 11. The computer program product of claim 10, wherein the index LOH region is about 6 megabases, 12 megabases, or 15 megabases in length, or more.

30. The method of any one of claims 1 to 6, wherein the at least one pair of human chromosomes is not human chromosome 17.

31. The use according to claim 7 or 12, wherein the indicator LOH region is not present in human chromosome 17.

32. The system according to claim 8 or 9, wherein the indicator LOH region is not present in human chromosome 17.

33. The computer program product of claim 10, wherein the indicator LOH region is not present in human chromosome 17.

34. 7. The method of claim 4 or 6, wherein the DNA damaging agent is cisplatin, carboplatin, oxalaplatin, or picoplatin; the anthracycline is epirubicin or doxorubicin; the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan; or the PARP inhibitor is iniparib, olaparib, or velapirib.

35. 13. The use of claim 12, wherein the DNA damaging agent is a platinum-based chemotherapy drug, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or velapirib.

36. The system of claim 9, wherein the DNA damaging agent is a platinum-based chemotherapy drug, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or velapirib.

37. 11. The computer program product of claim 10, wherein the DNA damaging agent is a platinum-based chemotherapy drug, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or velapirib.

38. (a) detecting, in cancer cells or genomic DNA obtained from the cancer cells, LOH regions in a representative number of human chromosome pairs from the cancer cells; and (b) determining the number and size of the LOH regions. A method comprising:

39. 39. The method of claim 38, wherein said representative number of human chromosome pairs is representative of the entire genome.

40. 39. The method of claim 38, further comprising correlating an increase in the number of LOH regions of a particular size with an increased likelihood of deletion in the HDR.

41. 41. The method of claim 40, wherein the specified size is greater than about 1.5 megabases, 2 megabases, 2.5 megabases, 3 megabases, 4 megabases, 5 megabases, 6 megabases, 7 megabases, 8 megabases, 9 megabases, 10 megabases, 11 megabases, 12 megabases, 13 megabases, 14 megabases, 15 megabases, 16 megabases, 17 megabases, 18 megabases, 19 megabases, 20 megabases, 25 megabases, 30 megabases, 35 megabases, 40 megabases, 45 megabases, 50 megabases, 75 megabases, or 100 megabases, and is less than the length of an entire chromosome containing the LOH region.

42. 42. The method of claim 40 or 41, wherein 6, 7, 8, 9, 10, 11, 12, or 13 or more LOH regions of the particular size are correlated with an increased likelihood of deletion in HDR.

43. A method for determining prognosis in a patient, comprising the steps of: (a) determining whether the patient contains cancer cells having an LOH signature, wherein the presence of more LOH regions than a reference number in at least one pair of human chromosomes from a cancer cell of the cancer patient indicates that the cancer cell has the LOH signature, wherein the LOH region is longer than a first length but shorter than the length of the entire chromosome containing the LOH region, wherein the at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and wherein the first length is about 1.5 megabases or more; and (b)(1) determining that the patient has a relatively good prognosis based, at least in part, on the presence of the LOH signature; or (b)(2) determining that the patient has a relatively poor prognosis based, at least in part, on the absence of the LOH signature.

44. 1. A composition comprising a therapeutic agent selected from the group consisting of a DNA damaging agent, an anthracycline, a topoisomerase I inhibitor, and a PARP inhibitor for use in treating a disease, a cancer selected from the group consisting of breast cancer, ovarian cancer, liver cancer, esophageal cancer, lung cancer, head and neck cancer, prostate cancer, colon cancer, rectal cancer, colorectal cancer, and pancreatic cancer, in a patient, wherein the patient has more LOH regions than a reference number in at least a pair of human chromosomes from the patient's cancer cells, the LOH regions are longer than a first length but shorter than the length of the entire chromosome containing the LOH regions, the at least a pair of human chromosomes are not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more.

45. The composition of claim 44, wherein the LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes.

46. The composition of claim 44, wherein the total number of LOH regions is 9, 15, 20, or more.

47. 45. The composition of claim 44, wherein the first length is about 6 megabases, 12 megabases, or 15 megabases, or more.

48. 45. The composition of claim 44, wherein the reference number is 6, 7, 8, 9, 10, 11, 12, or 13 or more.

49. 1. A method of treating cancer in a patient, comprising the steps of: determining in a sample derived from said patient a number of LOH regions in at least one pair of human chromosomes from a cancer cell of said cancer patient, said LOH regions being longer than a first length but shorter than the length of the entire chromosome containing said LOH regions, indicating said cancer cell has an LOH signature, said at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and said first length is about 1.5 megabases or more; providing a test value derived from the number of LOH regions; Comparing the test value to one or more reference values ​​(e.g., mean, median, tercile, quartile, quintile, etc.) derived from the number of LOH regions in a reference population; and administering an anti-cancer drug to the patient or recommending or prescribing or initiating a treatment regimen including chemotherapy and / or synthetic lethal agents based at least in part on the comparing step revealing that the test value exceeds at least one of the reference values ​​(e.g., at least 2-fold, 3-fold, 4-fold, 5-fold, 6-fold, 7-fold, 8-fold, 9-fold, or 10-fold the reference value; at least 1-, 2-, 3-, 4-, 5-, 6-, 7-, 8-, 9-fold, or 10-fold greater than the reference value); or recommending, prescribing or initiating a treatment regimen that does not include chemotherapy and / or synthetic lethal agents based at least in part on the comparing step revealing that the test value does not exceed at least one of the reference values ​​(e.g., is less than ½, ⅓, ¼, ⅕, ⅙, ⅙, ⅙, or ⅙ of the reference value; is less than 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 standard deviations of the reference value).

50. 50. The method of claim 49, wherein the LOH regions are determined in at least 2, 5, 10, or 21 pairs of human chromosomes.

51. 50. The method of claim 49, wherein the total number of LOH regions is 9, 15, 20, or more.

52. 50. The method of claim 49, wherein the first length is about 6 megabases, 12 megabases, or 15 megabases, or more.

53. 50. The method of claim 49, wherein the reference number is 6, 7, 8, 9, 10, 11, 12, or 13 or more.

54. 50. The method of claim 49, wherein the chemotherapy is selected from the group consisting of DNA damaging agents, anthracyclines, and topoisomerase I inhibitors, and / or the synthetic lethal agent is a PARP inhibitor.

55. 50. The method of claim 49, wherein the DNA damaging agent is cisplatin, carboplatin, oxalaplatin, or picoplatin, the anthracycline is epirubicin or doxorubicin, the topoisomerase I inhibitor is campothecin, topotecan, or irinotecan, and / or the PARP inhibitor is iniparib, olaparib, or velapirib.

Images