Methods and materials for evaluating 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 of related applications This application claims priority to U.S. Provisional Patent Application No. 61 / 578,713, filed on 21 December 2011, and U.S. Provisional Patent Application No. 61 / 654,402, filed on 1 June 2012. The entire contents of these applications are incorporated herein by reference.

[0002] 1. Technical field This document relates to methods and materials for evaluating samples (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 cells (e.g., cancer cells) contain an LOH signature. This document also provides materials and methods for identifying cells (e.g., cancer cells) with homology-directed repair (HDR) deficiencies, as well as materials and methods for identifying cancer patients who are likely to respond to a particular cancer treatment regimen. Throughout this document, unless otherwise specified, HDR deficiency and HRD (homologous repair deficiency) are used synonymously. [Background technology]

[0003] 2. Background information Cancer is a major public health issue, with 562,340 deaths from cancer in the United States alone in 2009. (American Cancer Society, Cancer Facts & Figures 2009 (Non-Patent Literature 1) (available on the American Cancer Society website)). One of the most important challenges in cancer treatment is to identify relevant and clinically useful features of a patient's cancer and then develop a treatment plan best suited to that patient's cancer based on these features. While 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 [Overview of the project]

[0005] overview Overall, one aspect of the present invention is characterized by a method for evaluating LOH in cancer cells or the genomic DNA of said cancer cells. In some embodiments, the method includes, or essentially comprises, (a) detecting LOH regions in at least one pair of human chromosomes from cancer cells (e.g., any pair of human chromosomes other than the human X / Y sex chromosome pair) in cancer cells or 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 that represent the entire genome (e.g., enough chromosomes are analyzed so that the number and size of LOH regions are expected to represent the number and size of LOH regions across the entire genome). In some embodiments, the method further includes the step of determining the total number of LOH regions (indicator 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 megabases (preferably 14 megabases, 15 megabases, 16 megabases, or more megabases, more preferably 15 megabases or more megabases), but shorter than the total length of each chromosome in which an LOH region is located. Or, further, the sum of the lengths of such indicator LOH regions is determined. In some specific embodiments, if the total number of indicator LOH regions or the sum of the lengths of the indicator LOH regions is equal to or greater than a predetermined reference number, the cancer cells or the genomic DNA or a patient having the cancer cells or genomic DNA is identified as having an HDR-deficient LOH signature.

[0006] Alternative methods are also provided for evaluating LOH in cancer cells or their genomic DNA, which include, or essentially consist of, the steps of (a) detecting LOH regions in at least one pair of human chromosomes from cancer cells in cancer cells or genomic DNA obtained from cancer cells, wherein at least one pair of human chromosomes is not a human X / Y sex chromosome pair; and (b) determining the total number and / or total length of LOH regions in at least one pair of human chromosomes, wherein the LOH regions are longer than a first length but shorter than the total length of the chromosome containing the 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 specific embodiments, if the total number or total length is equal to or greater than a predetermined reference number, the cancer cells or the genomic DNA or the patient having the cancer cells or genomic DNA is identified as having an HDR-deficient LOH signature.

[0007] In another aspect, the present invention provides a method for predicting the status of BRCA1 and BRCA2 genes in cancer cells. The method includes, or essentially comprises, the steps of, determining the total number and / or sum of lengths of LOH regions in at least one pair of human chromosomes from cancer cells, wherein the LOH regions are longer than a first length but shorter than the total length of the chromosome containing one LOH region, and at least one 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 a total number or sum of lengths exceeding a reference number with an increased likelihood of deletion in the BRCA1 or BRCA2 gene.

[0008] In another aspect, the present invention provides a method for predicting the HDR status in cancer cells. The method includes, or essentially comprises, the steps of, determining the total number and / or sum of lengths of LOH regions in at least one pair of human chromosomes from the cancer cells, wherein the LOH regions are longer than a first length but shorter than the total length of the chromosome containing one LOH region, and at least one 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 a total number or sum of lengths exceeding a reference number with an increased likelihood of deletion in HDR.

[0009] In another aspect, the present invention provides a method for predicting a cancer patient's response to a cancer treatment regimen comprising a DNA damaging agent, anthracycline, topoisomerase I inhibitor, radiation, and / or PARP inhibitor. The method comprises, or essentially comprises, a step of determining the number and / or total length of LOH regions in at least one pair of human chromosomes from cancer cells derived from a cancer patient, wherein the LOH regions are longer than a first length but shorter than the total length of the chromosome containing one LOH region, and at least one 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 a step of correlating a total number or total length exceeding a reference number with an increased likelihood that the cancer patient may respond to the cancer treatment regimen. In some embodiments, the patient is a previously untreated patient.

[0010] In another aspect, the present invention relates to a method for predicting a cancer patient's response to a treatment regimen. The method includes, or essentially comprises, the steps of, determining the total number and / or sum of lengths of LOH regions in at least one pair of human chromosomes from cancer cells derived from a cancer patient, wherein the LOH regions are longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one 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 a total number or sum of lengths exceeding a reference number with an increased likelihood that the cancer patient will not respond to a treatment regimen comprising paclitaxel or docetaxel.

[0011] In another aspect, the present invention relates to a method for treating cancer. The method includes (a) determining the total number and / or sum of the lengths of LOH regions in at least one pair of human chromosomes from cancer cells derived from a cancer patient or genomic DNA obtained from said cancer cells, wherein the LOH regions are longer than a first length but shorter than the total length of the chromosome containing the LOH regions, and at least one 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 (b) if the total number or sum of the lengths of the LOH regions exceeds a reference number, administering to a cancer patient a cancer treatment regime comprising one or more drugs selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors, or essentially comprising these steps. In some embodiments, the patient is a previously untreated patient.

[0012] In some embodiments of any one or more of the methods described in the foregoing six paragraphs, any one or more of the following can be applied as appropriate. 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 first length may be about 6 megabases, 12 megabases, or about 15 megabases, or more. The reference number may be 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, or 20, or more. At least one pair of human chromosomes may 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 camptothecin, topotecan, or irinotecan, and the PARP inhibitor may be iniparib, olaparib, or velapirib.

[0013] In another aspect, the present invention is characterized by 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 pharmaceutical useful for treating cancer in patients identified as having cancer cells determined to have a total of 5, 8, 9, 10, 12, 15, 17, 20, or more indicator LOH regions. Indicator LOH regions can be determined in at least 2, 5, 10, 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 regions may be approximately 6 megabases, 12 megabases, 15 megabases, or more. The indicator LOH regions may be located on chromosomes other than human chromosome 17. The DNA damaging agent may be a platinum-based chemotherapeutic agent, the anthracycline may be epiruvincin or doxorubicin, the topoisomerase I inhibitor may be campotecin, topotecan, or irinotecan, and the PARP inhibitor may be iniparib, olaparib, or berapirib. In some embodiments, the patient is a treatment-naïve patient.

[0014] In another aspect, the present invention is useful for determining the total number or total length of indicator LOH regions in at least one chromosomal pair from human cancer cells obtained from a cancer patient, and (a) an increased likelihood of a defect in the BRCA1 gene or BRCA2 gene in the cancer cells, (b) an increased likelihood of a defect in HDR in the cancer cells, or (c) an increased likelihood that a cancer patient can respond to a cancer treatment regimen comprising a DNA-damaging agent, anthracycline, topoisomerase I inhibitor, radiation, or PARP inhibitor. It 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. Indicator LOH regions can be determined in at least 2, 5, 10, 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 on 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 comprises, (a) a sample analyzer configured to generate multiple signals for genomic DNA of at least one pair of human chromosomes from cancer cells, and (b) a computer subsystem programmed to calculate the number or total length of indicator LOH regions in at least one pair of human chromosomes based on the multiple signals. The computer subsystem can be programmed to compare the number or total length of indicator LOH regions to a reference number to determine (a) the likelihood of deletion in the BRCA1 and / or BRCA2 genes in the cancer cells, (b) the likelihood of deletion in HDRs in the cancer cells, or (c) the likelihood that the cancer patient may respond to a cancer treatment regimen containing DNA damaging agents, anthracyclines, topoisomerase I inhibitors, radiation, or PARP inhibitors. The system may also include an output module configured to display the likelihood of (a), (b), or (c). The system may also include an output module configured to display recommendations for the use of a cancer treatment regimen. The indicator LOH region can be determined in at least two, five, ten, 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 approximately 6 megabases, 12 megabases, 15 megabases, or more. The indicator LOH region may be located on chromosomes other than human chromosome 17. The DNA damaging agent may be a platinum-based chemotherapeutic agent, the anthracycline may be epiruvincin or doxorubicin, the topoisomerase I inhibitor may be campotecin, topotecan, or irinotecan, and the PARP inhibitor may be iniparib, olaparib, or berapirib.

[0016] In another aspect, the present invention provides a computer program product embedded in a computer-readable medium that 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, wherein 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 shorter than the total length of the chromosome containing the LOH region; and instructions for determining the total number or sum of the lengths of LOH regions in one or more chromosome pairs. The computer program product may include other instructions. Indicator LOH regions can be determined in at least 2 pairs, 5 pairs, 10 pairs, or 21 pairs of human chromosomes. 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, 15 megabases, or more. The indicator LOH region may be located on chromosomes other than human chromosome 17. The DNA damaging agent may be a platinum-based chemotherapeutic agent, the anthracycline may be epiruvincin or doxorubicin, the topoisomerase I inhibitor may be campotecin, topotecan, or irinotecan, and the PARP inhibitor may be iniparib, olaparib, or berapirib.

[0017] In another aspect, the present invention provides a diagnostic kit. The kit comprises, or essentially comprises, at least 500 oligonucleotides that can hybridize to multiple polymorphic regions of human genomic DNA; and a computer program product provided herein. The computer program product may be embedded in a computer-readable medium and, when executed on a computer, 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, wherein the length of the LOH region is about 1.5 megabases or more megabases (or 5 megabases or 10 megabases or more megabases, preferably about 15 megabases or more megabases), but shorter than the total length of the chromosome containing the LOH region; and instructions for determining the total number and / or sum of the lengths of LOH regions in one or more chromosome pairs.

[0018] In another context, this document features a method for evaluating a patient's cancer cells for the presence of an LOH signature. The method includes, or is essentially derived from, (a) detecting the presence of more LOH regions in at least one pair of human chromosomes than the reference number in cancer cells of a cancer patient, wherein the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region, and 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 having cancer cells with an LOH signature.

[0019] In another context, this document features a method for evaluating a patient's cancer cells for the presence of an HDR deficiency. The method includes, or is essentially derived from, (a) detecting the presence of a LOH region in cancer cells from a cancer patient that is longer than a first length but shorter than the total length of the chromosome containing the LOH region, and that 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 having cancer cells with an HDR deficiency.

[0020] In another context, this document features a method for evaluating a patient's cancer cells for the presence of genetic mutations in genes originating from the HDR pathway. The method includes, or is essentially derived from, (a) detecting the presence of LOH regions in cancer cells from a cancer patient that are longer than a first length but shorter than the total length of the chromosome containing the LOH region, and that at least one pair of human chromosomes are not a human X / Y sex chromosome pair, and the first length is approximately 1.5 megabases or more, and (b) identifying the patient as having cancer cells with genetic mutations.

[0021] In another context, this document features a method for determining whether a patient is likely to respond to a cancer treatment regimen, comprising the steps of 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 is essentially comprised of, the steps of (a) detecting the presence of LOH regions in cancer cells of a cancer patient that are longer than a first length but shorter than the total length of the chromosome containing the LOH region, and that at least one pair of human chromosomes are not a human X / Y sex chromosome pair, and that the first length is about 1.5 megabases or more, and (b) identifying the patient as likely to respond to a cancer treatment regimen.

[0022] In another context, this document features a method for evaluating a patient. The method includes, or is essentially derived from, (a) determining whether a patient has cancer cells having an LOH signature, wherein the presence of LOH regions greater than the reference number 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 being longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one pair of human chromosomes not being a human X / Y sex chromosome pair, and the first length being about 1.5 megabases or more, and (b) diagnosing the patient that he or she has cancer cells having an LOH signature.

[0023] In another context, this document features a method for evaluating a patient. The method includes, or is essentially derived from, (a) determining whether a patient has cancer cells with an HDR deficiency, wherein the presence of LOH regions in the cancer cells of a cancer patient greater than the reference number in at least one pair of human chromosomes indicates that the cancer cells have an HDR deficiency, the LOH region being longer than a first length but shorter than the total length of the chromosome containing the LOH region, at least one pair of human chromosomes not being a human X / Y sex chromosome pair, and the first length being approximately 1.5 megabases or more, and (b) diagnosing the patient that he or she has cancer cells with an HDR deficiency.

[0024] In another context, this document features a method for evaluating a patient. The method includes, or is essentially derived from, (a) determining whether a patient has cancer cells with a gene mutation in a gene derived from the HDR pathway, wherein the presence of a LOH region greater than the reference number in at least one pair of human chromosomes from the cancer patient's cancer cells indicates that the cancer cells have a gene mutation, the LOH region being longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one pair of human chromosomes not being a human X / Y sex chromosome pair, and the first length being approximately 1.5 megabases or more, and (b) diagnosing the patient that he or she has cancer cells with a gene mutation.

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

[0026] In another context, this document features a method for performing a diagnostic analysis of cancer cells in a patient. The method includes, or is essentially derived from, (a) detecting the presence of LOH regions in cancer cells that are greater than a reference number in at least one pair of human chromosomes, wherein the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one pair of human chromosomes are not a human X / Y sex chromosome pair, and the first length is about 1.5 megabases or more, and (b) identifying a patient as having cancer cells with an LOH signature.

[0027] In another context, this document features a method for performing a diagnostic analysis of cancer cells in a patient. The method includes, or is essentially derived from, (a) detecting the presence of LOH regions in at least one pair of human chromosomes greater than the reference number in the cancer cells, wherein the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region, and 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 a patient as having cancer cells with an HDR deficiency.

[0028] In another context, this document features a method for performing a diagnostic analysis of cancer cells in a patient. The method includes, or is essentially derived from, (a) detecting the presence of LOH regions in cancer cells that are greater than the reference number in at least one pair of human chromosomes, wherein the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one pair of human chromosomes is not a human X / Y sex chromosome pair, and the first length is approximately 1.5 megabases or more, and (b) identifying a patient as having cancer cells with genetic mutations in genes derived from the HDR pathway.

[0029] In another context, this document features a method for performing a diagnostic analysis of a patient's cancer cells, comprising the steps of administering radiation or a drug selected from the group consisting of DNA damaging agents, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors, to determine whether the cancer patient is likely to respond to a cancer treatment regimen. The method comprises, or is essentially comprised of, the steps of (a) detecting the presence of LOH regions in cancer cells that are greater than the reference number in at least one pair of human chromosomes, wherein the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one pair of human chromosomes are 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 a cancer treatment regimen.

[0030] In another context, this document features a method for diagnosing a patient having cancer cells with an LOH signature. The method includes, or is essentially derived from, (a) determining that a patient has cancer cells with an LOH signature, wherein the presence of an LOH region greater than the reference number 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 being longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one pair of human chromosomes not being a human X / Y sex chromosome pair, and the first length being about 1.5 megabases or more, and (b) diagnosing the patient having cancer cells with an LOH signature.

[0031] In another context, this document features a method for diagnosing a patient having cancer cells with an HDR deficiency. The method includes, or is essentially derived from, (a) determining that a patient has cancer cells with an HDR deficiency, wherein the presence of LOH regions greater than the 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 deficiency, the LOH region being longer than a first length but shorter than the total length of the chromosome containing the LOH region, and at least one pair of human chromosomes not being a human X / Y sex chromosome pair, and the first length being approximately 1.5 megabases or more; and (b) diagnosing the patient having cancer cells with an HDR deficiency.

[0032] In another context, this document features a method for diagnosing a patient having cancer cells with a gene mutation in a gene derived from the HDR pathway. The method includes, or is essentially derived from, (a) determining that a patient has cancer cells with a gene mutation, wherein the presence of a LOH region greater than the reference number in at least one pair of human chromosomes from the cancer patient's cancer cells indicates that the cancer cells have a gene mutation, the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region, 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 having cancer cells with a gene mutation.

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

[0034] Unless otherwise specified, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art. Similar or equivalent methods and materials may be used in the practice of the present invention, but suitable methods and materials are described below. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference. In case of any conflict, this specification, including definitions, shall prevail. Furthermore, materials, methods, and examples are illustrative and not intended to limit the scope of the invention.

[0035] [Invention 1001] A method for evaluating LOH in cancer cells or the genomic DNA of said cancer cells, comprising the following steps: (a) A step of detecting LOH regions in at least one pair of human chromosomes from cancer cells or genomic DNA obtained from said cancer cells, wherein the at least one pair of human chromosomes is not a human X / Y sex chromosome pair; and (b) A step of determining the total number of LOH regions in at least one pair of human chromosomes, wherein the LOH regions are longer than a first length but shorter than the total length of the chromosome containing the LOH regions, and the first length is about 1.5 megabases or more. [Invention 1002] A method for predicting the status of BRCA1 and BRCA2 genes in cancer cells, including the following steps: A step of determining the total number of LOH regions in at least one pair of human chromosomes from the cancer cells, wherein the LOH region is longer than a first length but shorter than the total length of the 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 A step of correlating the total number, which exceeds the reference number, with an increased likelihood of deletion in the BRCA1 or BRCA2 gene. [Invention 1003] A method for predicting the HDR state in cancer cells, including the following steps: A step of determining the total number of LOH regions in at least one pair of human chromosomes from the cancer cells, wherein the LOH region is longer than a first length but shorter than the total length of the 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 A step of correlating the total number, which exceeds the reference number, with an increased possibility of loss in HDR. [Invention 1004] A method for predicting a cancer patient's response to a cancer treatment regimen comprising a DNA damaging agent, anthracycline, topoisomerase I inhibitor, radiation, and / or PARP inhibitor, comprising the following steps: A step of determining the number of LOH regions in at least one pair of human chromosomes from cancer cells derived from a cancer patient, wherein the LOH region is longer than a first length but shorter than the total length of the 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 A step of correlating a total number exceeding the reference number with an increased likelihood that a cancer patient may respond to the cancer treatment regimen. [Invention 1005] A method for predicting a cancer patient's response to a treatment regimen, including the following steps: A step of determining the total number of LOH regions in at least one pair of human chromosomes from cancer cells derived from a cancer patient, wherein the LOH region is longer than a first length but shorter than the total length of the 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 The step of correlating the total number exceeding the reference number with an increased likelihood that the cancer patient will not respond to a treatment regimen including paclitaxel or docetaxel. [Invention 1006] A method of treating cancer, including the following steps: (a) A step of determining the total number of LOH regions in at least one pair of human chromosomes from cancer cells derived from a cancer patient or genomic DNA obtained from said cancer cells, wherein the LOH region is longer than a first length but shorter than the total length of the 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) If the total number of LOH regions exceeds the reference number, the cancer patient is given 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. [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 pharmaceutical product useful for treating cancer in patients identified as having cancer cells determined to have a total of five or more indicator LOH regions. [Invention 1008] (a) A sample analyzer configured to generate multiple signals for at least one pair of human chromosome genomic DNA from cancer cells, and (b) A computer subsystem programmed to calculate the number of indicator LOH regions in at least one pair of human chromosomes based on the multiple signals. A system for determining the LOH (Long-Onset) status of cancer cells in a cancer patient. [Invention 1009] The computer subsystem compares the number of the index LOH area with the reference number, (a) Possibility of deletion in the BRCA1 gene and / or BRCA2 gene in the cancer cells, (b) the possibility of a defect in HDR in the cancer cells, or (c) The possibility that the cancer patient may respond to a cancer treatment regimen containing DNA damaging agents, anthracyclines, topoisomerase I inhibitors, radiation, or PARP inhibitors. A system according to the present invention 1008, which is programmed to determine the following. [Invention 1010] When running on a computer, A step for detecting the presence or absence of any LOH region along one or more 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 shorter than the length of the entire chromosome containing the LOH region; and The step of determining the total number of LOH regions in one or more chromosome pairs. A computer program product embedded in a computer-readable medium that performs processes including the following. [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 that includes this. [Invention 1012] It is useful for determining the total number of indicator LOH regions in at least one chromosomal pair from human cancer cells obtained from cancer patients, and (a) Increased likelihood of deficiency in the BRCA1 gene or BRCA2 gene in the cancer cells, (b) Increased likelihood of HDR defects in the cancer cells, or (c) Increased likelihood that cancer patients may respond to cancer treatment regimens containing DNA damaging agents, anthracyclines, topoisomerase I inhibitors, radiation, or PARP inhibitors. The use of multiple oligonucleotides that can hybridize to multiple polymorphic regions of human genomic DNA for manufacturing diagnostic kits useful for detecting [the condition]. [Invention 1013] The method according to any one of the present invention 1001 to 1006, wherein the LOH region is determined in at least 2 pairs, 5 pairs, 10 pairs, or 21 pairs of human chromosomes. [Invention 1014] The method according to any one of items 1001 to 1006 of the present invention, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [Invention 1015] The method according to any one of the invention 1001 to 1006, wherein the total number of LOH regions is 9, 15, 20, or more. [Invention 1016] The method according to any one of the present invention 1001 to 1006, wherein the first length is approximately 6 megabases, 12 megabases, 15 megabases, or more. [Invention 1017] The method according to any one of the invention 1001 to 1006, wherein the number of references is 6, 7, 8, 9, 10, 11, 12, or 13 or more. [Invention 1018] Use of the present invention 1007 or 1012, wherein the indicator LOH region is determined in at least two pairs, five pairs, ten pairs, or 21 pairs of human chromosomes. [Invention 1019] Use of the present invention 1007 or 1012, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [Invention 1020] Use of Invention 1007 or 1012, wherein the total number of the aforementioned indicator LOH regions is 9, 15, 20, or more. [Invention 1021] Use of Invention 1007 or 1012, wherein the length of the index LOH region is approximately 6 megabases, 12 megabases, 15 megabases, or more. [Invention 1022] The system of the present invention 1008 or 1009, wherein the indicator LOH region is determined in at least two pairs, five pairs, ten pairs, or 21 pairs of human chromosomes. [Invention 1023] The system of the present invention 1008 or 1009, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [Invention 1024] The system of the present invention 1008 or 1009, wherein the total number of the aforementioned indicator LOH regions is 9, 15, 20, or more. [Invention 1025] The system of the present invention 1008 or 1009, wherein the length of the index LOH region is approximately 6 megabases, 12 megabases, 15 megabases, or more. [Invention 1026] A computer program product of the present invention 1010, wherein the aforementioned indicator LOH region is determined in at least two pairs, five pairs, ten pairs, or 21 pairs of human chromosomes. [Invention 1027] A computer program product of the present invention 1010, wherein the cancer cells are ovarian cancer cells, breast cancer cells, or esophageal cancer cells. [Invention 1028] A computer program product of the present invention 1010, wherein the total number of the aforementioned indicator LOH areas is 9, 15, 20, or more. [Invention 1029] A computer program product of the present invention 1010, wherein the length of the aforementioned index LOH area is approximately 6 megabases, 12 megabases, 15 megabases, or more. [Invention 1030] Any method 1001 to 1006 of the present invention, wherein the at least pair of human chromosomes are not human chromosome 17. [Invention 1031] Use of Invention 1007 or 1012, wherein the aforementioned indicator LOH region is not present on human chromosome 17. [Invention 1032] The system of the present invention 1008 or 1009, wherein the aforementioned indicator LOH region is not present on human chromosome 17. [Invention 1033] A computer program product of the present invention 1010, wherein the aforementioned indicator LOH region is not present on human chromosome 17. [Invention 1034] The method of the present invention 1004 or 1006, wherein the DNA damaging substance is cisplatin, carboplatin, oxalaplatin, or picoplatin; the anthracycline is epiruvinin or doxorubicin; the topoisomerase I inhibitor is campotecin, topotecan, or irinotecan; or the PARP inhibitor is iniparib, olaparib, or berapirib. [Invention 1035] Use of Invention 1012, wherein the DNA damaging agent is a platinum-based chemotherapeutic agent, the anthracycline is epiruvincin or doxorubicin, the topoisomerase I inhibitor is campotecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or berapirib. [Invention 1036] The system of the present invention 1009, wherein the DNA damaging substance is a platinum-based chemotherapeutic agent, the anthracycline is epiruvincin or doxorubicin, the topoisomerase I inhibitor is campotecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or berapirib. [Invention 1037] A computer program product of the present invention 1010, wherein the DNA damaging substance is a platinum-based chemotherapeutic agent, the anthracycline is epiruvincin or doxorubicin, the topoisomerase I inhibitor is campotecin, topotecan, or irinotecan, or the PARP inhibitor is iniparib, olaparib, or berapirib. [Invention 1038] (a) A step of detecting LOH regions in a representative number of human chromosome pairs from cancer cells or genomic DNA obtained from said cancer cells; and (b) A step of determining the number and size of the LOH regions. Methods that include... [Invention 1039] The method of the present invention 1038, wherein the aforementioned representative number of human chromosome pairs represent the entire genome. [Invention 1040] The method of the present invention 1038, further comprising the step of correlating an increase in the number of LOH regions of a particular size with an increase in the likelihood of defects in HDR. [Invention 1041] The method of the present invention 1040, wherein the specific size is longer than approximately 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 shorter than the total length of the chromosome containing the LOH region. [Invention 1042] The method of the present invention 1040 or 1041, wherein six, seven, eight, nine, ten, eleven, twelve, or thirteen or more of the aforementioned LOH regions of a particular size are correlated with an increased likelihood of loss in HDR. [Invention 1043] A method for determining the prognosis in a patient, including the following steps: (a) A step of determining whether a patient has cancer cells having an LOH signature, wherein the presence of an LOH region greater than the reference number in at least one pair of human chromosomes from the cancer patient's cancer cells indicates that the cancer cells have the LOH signature, the LOH region is longer than a first length but shorter than the total length of the 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)(1) Determining, at least in part, on the presence of the LOH signature, that the patient has a relatively good prognosis; or (b)(2) A step of determining that the patient has a relatively poor prognosis, at least in part on the absence of the LOH signature. [Invention 1044] A composition for use in the treatment of a patient for 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, comprising a therapeutic substance selected from the group consisting of DNA damaging substances, anthracyclines, topoisomerase I inhibitors, and PARP inhibitors, wherein the patient has more LOH regions than the reference number in at least one pair of human chromosomes from the patient's cancer cells, the LOH region is longer than a first length but shorter than the total length of the 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. [Invention 1045] The composition of the present invention 1044, wherein the LOH region is determined in at least 2 pairs, 5 pairs, 10 pairs, or 21 pairs of human chromosomes. [Invention 1046] The composition of the present invention 1044, wherein the total number of LOH regions is 9, 15, 20, or more. [Invention 1047] The composition of the present invention 1044, wherein the first length is approximately 6 megabases, 12 megabases, 15 megabases, or more. [Invention 1048] The composition of the present invention 1044, wherein the reference number is 6, 7, 8, 9, 10, 11, 12, or 13, or more. [Invention 1049] A method for treating cancer in a patient, including the following steps: A step of determining the number of LOH regions in at least one pair of human chromosomes from cancer cells of a cancer patient, wherein the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region, indicating that the cancer cell has an LOH signature, and 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; A step of providing test values ​​derived from the number of LOH regions; The steps include comparing the test value with one or more reference values ​​(e.g., mean, median, tercile, quartile, quintile, etc.) derived from the number of LOH regions in the reference population; and A step of administering an anticancer drug to the patient or recommending, prescribing, or initiating a treatment regimen including chemotherapy and / or synthetic lethal agents, based at least in part on a comparing step that reveals that the test value exceeds at least one reference value (for example, at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 times the reference value; or that the test value is at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 standard deviations greater than the reference value); or A step of 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 of determining that the test value does not exceed at least one reference value (for example, less than or equal to 1 / 2, 1 / 3, 1 / 4, 1 / 5, 1 / 6, 1 / 7, 1 / 8, 1 / 9, or 1 / 10 of the reference value; less than or equal to 1 standard deviation, 2 standard deviations, 3 standard deviations, 4 standard deviations, 5 standard deviations, 6 standard deviations, 7 standard deviations, 8 standard deviations, 9 standard deviations, or 10 standard deviations of the reference value). [Invention 1050] The method of the present invention 1049, wherein the LOH region is determined in at least 2 pairs, 5 pairs, 10 pairs, or 21 pairs of human chromosomes. [Invention 1051] The method of the present invention 1049, wherein the total number of LOH regions is 9, 15, 20, or more. [Invention 1052] The method of the present invention 1049, wherein the first length is approximately 6 megabases, 12 megabases, 15 megabases, or more. [Invention 1053] The method of the present invention 1049, wherein the reference number is 6, 7, 8, 9, 10, 11, 12, or 13, or greater than that. [Invention 1054] The method of the present invention 1049, 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. [Invention 1055] The method of the present invention 1049, wherein the DNA damaging substance is cisplatin, carboplatin, oxalaplatin, or picoplatin, the anthracycline is epiruvincin or doxorubicin, the topoisomerase I inhibitor is campotecin, topotecan, or irinotecan, and / or the PARP inhibitor is iniparib, olaparib, or berapirib. Details of one or more embodiments of the present invention are shown in the following description and accompanying drawings. The materials, methods, and examples are illustrative and not intended to limit the invention. Other features, purposes, and advantages of the present invention will be apparent from the description and drawings and the claims. [Brief explanation of the drawing]

[0036] [Figure 1] This graph plots the allele dosages of breast cancer cells derived from breast cancer patients, determined using a SNP array, along chromosome 1. The arrows indicate the transition from the heterozygous region to the LOH region. [Figure 2] This graph, similar to Figure 1, plots the allele levels of breast cancer cells from the same breast cancer patient along chromosome 1, determined using high-throughput sequencing. The arrows indicate the transition from the heterozygous region to the LOH region. [Figure 3] This is an exemplary process flowchart for evaluating the genome of a cell (e.g., a cancer cell) for LOH signatures. [Figure 4] This figure shows an example of a computer device and a mobile computer device that can be used to perform the techniques described herein. [Figure 5]This graph plots the length distribution of LOH regions detected in ovarian cancer cells derived from 62 human patients. The corrected length represents the proportion of chromosome arms covered by the LOH region. [Figure 6] This graph plots the number of LOH regions longer than 15 Mb and shorter than the entire chromosome for a training set of ovarian cancer cell samples containing intact or deficient BRCA1 and BRCA2 genes. The size of the circles is proportional to the number of samples containing such a number of LOH regions. [Figure 7] This graph plots the number of LOH regions longer than 15 Mb and shorter than the entire chromosome for training and validation sets consisting of ovarian cancer cell samples with intact or deficient BRCA1 and BRCA2 genes. The size of the circles is proportional to the number of samples with this number of LOH regions. [Figure 8] This graph plots the number of LOH regions longer than 15 Mb and shorter than the entire chromosome in ovarian cancer cell samples with BRCA somatic mutations, BRCA germline mutations, BRCA1 low expression, or intact BRCA (normal BRCA). The size of the circles is proportional to the number of samples with this number of LOH regions. [Figure 9] This table shows the percentages of BRCA-deficient ovarian cancer samples, HDR-deficient / BRCA-intact ovarian cancer samples, and HDR-intact ovarian cancer samples. [Figure 10] This graph plots the number of LOH regions (longer than 15 Mb and shorter than the entire chromosome) in the displayed cancer cell lines. The size of the circles is proportional to the number of samples that have this number of LOH regions. [Figure 11] This graph plots the number of LOH regions (longer than 15 Mb and shorter than the entire chromosome) in lung cancer samples. [Figure 12]This graph plots the percentage of displayed cancers or cancer cell lines that have HDR deficiency. [Figure 13] The graph includes plots of IC50 values ​​(Log10 (IC50) for camptothecin and the average Log10 (IC50) values ​​for platinum compounds (oxaliplatin, cisplatin, and carboplatin) or anthracyclines (doxorubicin and epirubicin)) upon exposure to 29 breast cancer cell lines with a number of LOH regions longer than 15 Mb and shorter than the entire chromosome, or the IC50 value (Log10 (IC50)) for paclitaxel upon exposure to 27 ovarian cancer cell lines with a number of LOH regions longer than 15 Mb and shorter than the entire chromosome. A dashed line indicates a threshold of 9. [Figure 14] This is a labeled version of the graph from Figure 13, plotting the mean Log10 (IC50) values ​​of platinum compounds (oxaliplatin, cisplatin, and carboplatin) upon exposure to 29 breast cancer cell lines having a number of LOH regions longer than 15 Mb and shorter than the entire chromosome. [Figure 15] This is a flowchart illustrating an exemplary computer computation process for identifying the locus and region of LOH. [Figure 16] This figure shows the proportion of LOH region length and the corrected lengths of these regions for the length of the paired chromosome arms. The maximum corrected value in this figure is equal to 2, which corresponds to the LOH region spanning the entire chromosome. [Figure 17a] The HRD scores in tumor samples are shown. Blue circles: BRCA1-deficient or BRCA2-deficient samples. Red circles: BRCA1 intact samples 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 a corresponding number of LOH regions. HRD scores for the first cohort (46 out of 152 samples were BRCA1-deficient or BRCA2-deficient). [Figure 17b]The HRD scores in tumor samples are shown. Blue circles: BRCA1-deficient or BRCA2-deficient samples. Red circles: BRCA1 intact samples 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 for the second cohort (19 out of 53 samples were BRCA1-deficient or BRCA2-deficient). [Figure 17c] The HRD scores in tumor samples are shown. Blue circles: BRCA1-deficient or BRCA2-deficient samples. Red circles: BRCA1 intact samples 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 for the third cohort (146 out of 435 samples were BRCA1-deficient or BRCA2-deficient). [Figure 17d] HRD scores in tumor samples are shown. Blue circles: BRCA1-deficient or BRCA2-deficient samples. Red circles: BRCA1 intact samples 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 composite data from all three cohorts. Row A: 224 samples with either a BRCA1-deficient gene, a BRCA2-deficient gene, or a RAD51C-deficient gene; B: 84 BRCA1 variants; C: 43 BRCA2 variants; D: 82 samples with BRCA1 low expression or BRCA1 methylation; E: 13 RAD51C methylated 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 with heterozygous BRCA1 or BRCA2 mutations. Purple circles: 2 carriers with homozygous mutations with revertant mutations in either BRCA1 or BRCA2. Blue circles: 7 carriers with homozygous mutations in either BRCA1 or BRCA2, or with methylated BRCA1. The sum of the areas of the green, red, blue, and purple circles is the same. The area of ​​each individual circle is proportional to the number of samples with the corresponding number of LOH regions. [Figure 18b] Kaplan-Meier plots of overall postoperative survival for HRD scores separated by median. These data were prepared using 507 samples from the TCGA dataset, where copy number data and survival information were available. The median overall survival for samples with high HRD scores was 1499 days (95% CI = (1355–1769)) and 1163 days (95% CI = (1081–1354)), respectively. [Figure 19] This figure shows the correlation between LOH scores and HR missings calculated for different LOH region length cutoffs in the first cohort. The corresponding log10 (p-value) is on the y-axis. The relationship between LOH region size cutoffs and the significance of the correlation between LOH scores and HR missings were examined. This figure shows that the LOH length cutoff can easily be between 11 and 21 Mb. In some preferred embodiments, a 15 Mb cutoff, which lies approximately in the middle of this interval, can be used because the 15 Mb cutoff has been found to be highly sensitive to statistical noise present in the data. [Figure 20]This shows a comparison of LOH scores in three groups of BRCA1-deficient and BRCA2-deficient samples, using composite 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] This shows a comparison of LOH scores for BRCA1-deficient, BRCA2-deficient, and RAD51C-deficient samples. Blue circles correspond to BRCA1-deficient samples, red circles to BRCA2-deficient samples, and green circles to RAD51C-deficient samples. The sum of the areas of the red, blue, and green circles is the same. The area of ​​each individual circle is proportional to the number of samples with the corresponding number of LOH regions. [Modes for carrying out the invention]

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

[0038] Generally, by comparing sequences at the same locus on each chromosome (each autosome in males), it is possible to determine whether a particular locus is homozygous or heterozygous within the cell genome. Typically, since an individual receives one copy from their biological father and one copy from their biological mother, 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 can become homozygous as a result of inheriting identical copies from both 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 there is only one copy (instead of two) of that region in the genome of the affected cell, the region that remains on the other chromosome (the other non-sex chromosome in males) becomes the LOH region. This LOH region can be of any length (for example, from less than approximately 1.5 Mb to the length of the chromosome). This type of LOH event reduces the copy number. In other cases, a region of one chromosome in a somatic cell (one non-sex chromosome in males) may be replaced by a copy of the same region from the other chromosome, thereby eliminating any heterozygosity that may have existed in the replaced region. In such cases, the region that remains on each chromosome becomes the LOH region, sometimes called a copy-neutral LOH region. A copy-neutral LOH region can be of any length (for example, from less than approximately 1.5 Mb to the length of the chromosome).

[0040] As described herein, the genome of the cell being evaluated 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, 15, or 16 megabases, more preferably longer than about 15 megabases), and (b) shorter than the total length of the chromosome containing the LOH region, with five or more (e.g., six or more, seven or more) A cell sample (e.g., a cancer cell sample) can be identified as having a “positive LOH signature state” (or, instead, referred to as an “HDR-deficient LOH signature”) if it contains 8 or more, 9 or more, 10 or more, 11 or more, 12 or more, 13 or more, 14 or more, 15 or more, 16 or more, 17 or more, 18 or more, 19 or more, or 20 or more LOH regions. In some cases, a cancer cell sample can be identified as having a positive LOH signature state if the genome of the cells being evaluated contains 9 or more LOH regions that are (a) longer than approximately 15 Mb and (b) shorter than the total length of the chromosome containing the LOH region. Unless otherwise specified, the term “indicator LOH region” refers to an LOH region located 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 approximately 1.5 megabases or more, but shorter 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, the indicator LOH region is any LOH region that is approximately 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 larger (preferably longer than approximately 14 or 15 megabases) and shorter than the total length of the chromosome containing the LOH region.

[0041] Cells identified as having a positive LOH signature (also referred to herein as an "HDR-deficient LOH signature") (e.g., cancer cells) can be classified as likely to have HDR deficiency and / or deficiency in one or more genes of the HDR pathway. For example, cancer cells identified as having a positive LOH signature can be classified as likely to have HDR deficiency. In some cases, cancer cells identified as having a positive LOH signature can be classified as likely to have deficiency in one or more genes of the HDR pathway. As used herein, gene deficiency means that the sequence, structure, expression, and / or activity of a gene or its product is deficient compared to normal. Examples include, but are not limited to, low or absent mRNA or protein expression, harmful mutations, hypermethylation, and reduced activity (e.g., enzyme activity, ability to bind to another biomolecule). As used herein, pathway deficiency (e.g., HDR pathway) means that at least one gene in that pathway (e.g., BRCA1) is deficient. Examples of highly detrimental mutations include frameshift mutations, stop codon mutations, and mutations that lead to alterations in RNA splicing. Deficiencies in HDR pathway genes can result in deficiencies or reduced activity of homologous recombination repair in cancer cells. Examples of HDR pathway genes include, but are not limited to, those listed in Table 1.

[0042] (Table 1) Selected HDR pathway genes TIFF0007844597000001.tif234147

[0043] Examples of gene mutations that may exist within HDR pathway genes include, but are not limited to, those listed in Table 2.

[0044] (Table 2) Possible gene mutations in selected genes of the HDR pathway TIFF0007844597000002.tif79128

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

[0046] As described herein, identifying LOH loci (and the size and number of LOH regions) may first involve determining the genotype of a sample at various genomic loci (e.g., SNP loci, individual bases in large-scale sequencing), and then determining whether homozygous loci are due to LOH events. Genotype at loci of interest within the cellular genome can be determined using any suitable technique. 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 surrounding sequences), and even untargeted sequencing (e.g., whole exome, transcriptome, or genome sequencing) can be used to identify whether a locus is homozygous or heterozygous. In some cases, the homozygosity or heterozygosity of a locus can be analyzed over the length of a chromosome to determine the length of the homozygous or heterozygous region. For example, using SNP array results, it is possible to evaluate regions consisting of SNP locations spaced apart along the chromosome (e.g., spaced approximately 25kb to 100kb apart), determining not only the presence of homozygous regions along the chromosome but also their length. Using the results from the SNP array, a graph plotting allele amounts along the chromosome can be created. (Example: Allele amount d of SNPi) i It consists of two alleles (A i and B i It can be calculated from the corrected signal intensity of ):d i =A i / ( A i +B i An example of such a graph is shown in Figure 1. The great many variations on nucleic acid arrays useful in the present invention are known in the art. These include the arrays used in the following various examples (e.g., the Affymetrix 500K GeneChip array in Example 3; the Affymetrix OncoScan® FFPE Express 2.0 Services (formerly MIP CN Services) in Example 4).

[0047] Once the genotype of a sample is determined for multiple loci (e.g., SNPs), LOH loci and regions can be identified using common techniques. One method for determining whether homozygosity is due to an LOH event is to compare the somatic cell genotype with that of the germline. For example, the genotypes of multiple loci (e.g., SNPs) can be determined in both germline (e.g., blood) and somatic cell (e.g., tumor) samples. By comparing the genotypes of each sample (typically by computer calculation), it can be determined whether the germline cell genome is heterozygous and the somatic cell genome is homozygous. Such loci are LOH loci, and the regions of such loci are LOH regions.

[0048] Computer methods can also be used to determine whether homozygosity is due to a LOH event. Such techniques are particularly useful when germline samples are not available for analysis and comparison. For example, algorithms, e.g., 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 clearly account for contamination of tumor samples by 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 detection of LOH regions difficult. Improved analytical methods according to the present invention that identify LOH even in the presence of contamination include analytical methods incorporated into the computer software products described below.

[0049] The following is an example. When the observed ratio of signals from two alleles A and B is 2:1, there are two possibilities. The first possibility is that cancer cells have LOH with a deletion of allele B in a sample contaminated 50% with normal cells. The second possibility is that in a sample not contaminated with normal cells, there is no LOH, but allele A is duplicated. The algorithm can be executed as a computer program described herein to reconstruct the LOH region 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 region is then determined as an SNP region where one of the ASCNs (paternal or maternal) is 0. This algorithm may be based on maximizing the likelihood function and may be 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 number PCT / US2011 / 026098 to Abkevich et al. The likelihood function can be maximized over ASCN for all loci, the level of contamination by benign tissue, the total copy number averaged across the entire genome, and the sample-specific noise level. The input data for the algorithm may include, or consist of, (1) sample-specific standardized signal intensities for both alleles at each locus and (2) assay-specific (different SNP array and sequence-based approach specific) parameter sets defined based on analysis of a large number of samples with known ASCN profiles.

[0050] In some cases, nucleic acid sequencing can be used to identify whether a locus is homozygous or heterozygous. For example, genomic DNA can be extracted and fragmented from a cell sample (e.g., a cancer cell sample). Genomic nucleic acids can be extracted and fragmented using any suitable method, including but not limited to commercially available kits such as the QIAamp® DNA Mini Kit (Qiagen®), the MagNA® Pure DNA Isolation Kit (Roche Applied Science®), and the GenElute® Mammalian Genomic DNA Miniprep Kit (Sigma-Aldrich®). Once extracted and fragmented, targeted or untargeted 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 genotypes at millions, or even billions, of base pairs (i.e., the base pairs may also be the “locus” 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, using kits designed for this purpose (e.g., Agilent SureSelect®, Illumina TruSeq Capture®, and Nimblegen SeqCap EZ Choice®), genomic DNA can be enriched with fragments containing the locus to be analyzed (e.g., SNP locations). For example, genomic DNA containing the locus to be analyzed can be hybridized with a biotinylated capture RNA fragment to form a biotinylated RNA / genomic DNA complex. Alternatively, a DNA capture probe may be used to form a biotinylated RNA / genomic DNA hybrid. The biotinylated RNA / genomic DNA complex can be separated from genomic DNA fragments not present in the complex using streptavidin-coated magnetic beads and magnetic force. The resulting biotinylated RNA / genomic DNA complex can be processed to remove the captured RNA from the magnetic beads, thereby leaving intact genomic DNA fragments containing the locus to be analyzed. These intact genomic DNA fragments containing the locus to be analyzed can be amplified, for example, using PCR. The amplified genomic DNA fragments can then be sequenced using high-throughput sequencing or next-generation sequencing technologies, such as Illumina HiSeq®, Illumina MiSeq®, Life Technologies SoLID®, Ion Torrent®, or Roche 454®.

[0052] Similar to the microarray analysis described herein, loci can be identified as being homozygous or heterozygous using sequencing results from genomic DNA fragments. 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, a region consisting of SNP positions spaced along a chromosome (e.g., spaced at intervals of about 25 kb to about 100 kb) can be evaluated by sequencing, and using the sequencing results, not only can the presence of homozygous regions along the chromosome be determined, but also the length of this LOH region can be determined. Using the obtained sequencing results, a graph plotting the allele amounts along the chromosome can be created. The allele amount d i of SNPi can be calculated from the corrected number of capture probes for two alleles (A i and B i ): d i = A i / (A i + B i ). An example of such a graph is shown in FIG. 2. Whether homozygosity is due to an LOH event (in contrast 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 assay methods (e.g., SNP array-based assays and sequencing-based assays) configured to identify whether a locus is 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 whether a locus is homozygous or heterozygous within the cellular genome. In some cases, 500,000, 1,000,000, 1,500,000, 2,000,000, 2,500,000 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 approximately 5 percent in Caucasians, (d) have a minor allele frequency of at least approximately 1 percent in three non-Caucasian races (e.g., Chinese, Japanese, and Yoruba), and / or (e) do not significantly deviate from the 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) to (e) can be selected. From the human SNPs that meet criteria (a) to (e), a group of SNPs (e.g., the top 110,000 SNPs) can be selected such that the SNPs have high allele frequencies in Caucasians, cover the human genome at somewhat uniform intervals (e.g., at least one SNP every 25kb to 500kb), and are not in linkage disequilibrium with other selected SNPs in any of the four races. In some cases, approximately 40,000, 50,000, 60,000, 70,000, 80,000, 90,000, 100,000, 110,000, 120,000, 130,000, or more SNPs that meet these criteria can be selected and included in assays configured to identify LOH regions across the entire human genome. For example, approximately 70,000 to 90,000 (e.g., approximately 80,000) SNPs can be selected for analysis using an SNP array-based assay method.For analysis using sequencing-based assays, approximately 45,000 to 55,000 SNPs (for example, approximately 54,000) can be selected.

[0054] As described herein, cell samples can be evaluated to determine whether the genome of the cells in the sample contains or lacks an LOH signature, whether the number of LOH regions covering the entire chromosome is increased, or whether the number of LOH regions covering the entire chromosome is not increased. 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 or lacks an LOH signature, whether the number of LOH regions covering the entire chromosome is increased, or whether the number of LOH regions covering the entire chromosome is not increased. Examples of cancer cell-containing samples that can be evaluated as described herein include, but are not limited to, tumor biopsy samples (e.g., mammary gland tumor biopsy samples), formalin-fixed and paraffin-embedded cancer cell-containing tissue samples, core needle biopsy materials, aspiration fluid, and samples containing cancer cells detached from a tumor (e.g., blood, urine, or other body fluids). In the case of formalin-fixed and paraffin-embedded tissue samples, the samples can be prepared by DNA extraction using genomic DNA extraction kits optimized for FFPE tissue, including, but not limited to, the aforementioned kits (e.g., QuickExtract® FFPE DNA Extraction Kit (Epicentre®) and QIAamp® DNA FFPE Tissue Kit (Qiagen®)).

[0055] In some cases, laser dissection can be performed on tissue samples to minimize the number of non-cancerous cells in the cancer cell sample to be evaluated. In some cases, antibody-based purification methods can be used to enrich cancer cells and / or 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-antimesencymal 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 cells, ovarian cancer cells, liver cancer cells, esophageal cancer cells, lung cancer cells, head and neck cancer cells, prostate cancer cells, colon cancer cells, rectal cancer cells, or colorectal cancer cells, and pancreatic cancer cells can be evaluated to determine whether the genome of the cancer cells contains or lacks an LOH signature, whether the number of LOH regions covering the entire chromosome is increased, or whether the number of LOH regions covering the entire chromosome is not increased. In some embodiments, the cancer cells are primary or metastatic cancer cells of ovarian cancer, breast cancer, lung cancer, or esophageal cancer.

[0057] When evaluating the genome of cancer cells for the presence or absence of a LOH signature, one or more pairs of chromosomes (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, or 23 pairs) can be evaluated. In some cases, the genome of cancer cells is evaluated for the presence or absence of a LOH signature using one or more pairs of chromosomes (e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, or 23 pairs).

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

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

[0060] Therefore, a predetermined number of chromosomes may be analyzed to determine the total number of indicator LOH regions, preferably those with a length of more than 9 megabases, more than 10 megabases, more than 12 megabases, more than 14 megabases, and more preferably more than 15 megabases. Alternatively, or further, the total length of all indicator LOH regions may be obtained by summing the sizes of all identified indicator LOH regions.

[0061] To classify positive LOH signature states, the aforementioned reference numbers 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, and most preferably 10. The reference number of the total length (e.g., sum of lengths) of the index LOH region may be about 75, 90, 105, 120, 130, 135, 150, 175, 200, 225, 250, 275, 300, 325, 350, 375, 400, 425, 450, 475, or 500 megabases, 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 longer than approximately 14 or 15 megabases is determined and compared to a reference number of approximately 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, or 20. Alternatively, or further, the total length of LOH regions longer than approximately 14 or 15 megabases is determined and compared to a reference number of approximately 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 of LOH regions in a patient sample (or the total length, or any test value or score obtained from either) is considered to "exceed" the reference if it is at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 times the reference, and in some embodiments, it is considered to "exceed" the reference if it is at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 standard deviations greater than the reference. Conversely, in some embodiments, the number of LOH regions in a patient sample (or the sum of their lengths, or any test value or score obtained from them) is considered "not exceeding" the reference if it is less than or equal to 1 / 2, 1 / 3, 1 / 4, 1 / 5, 1 / 6, 1 / 7, 1 / 8, 1 / 9, or 1 / 10 of the reference, and furthermore, in some embodiments, it is considered "not exceeding" the reference if it is less than or equal to 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 derived 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 having similar genetic or other clinical or molecular characteristics, (d) patients who responded to a particular treatment, (e) patients who did not respond to a particular treatment, and (f) clearly healthy patients (e.g., patients without cancer or at least patients without cancer as the patient being tested). A reference number (or length, value, or score) may (a) be representative of a number (or length, value, or score) generally observed in the reference population; (b) be the average (mean, median, etc.) of numbers (or lengths, values, or scores) generally observed in the reference population or in a particular subpopulation; or (c) be representative of a number (or length, value, or score) (e.g., mean, median, etc.) found in the tercile, quartile, quintile, etc., of the reference population, ranked by (i) each number (or length, value, or score) or (ii) the clinical features found to be present (e.g., response strength, prognosis (including time to cancer-specific death), etc.).

[0065] As described herein, patients with cancer cells identified as having a positive LOH signature can be classified, at least partially, as likely to respond to a particular cancer treatment regimen based on the positive LOH signature. For example, patients with cancer cells having a genome containing the LOH signature can be classified, at least partially, as likely to respond to a cancer treatment regimen including the use of DNA damaging agents, synthetic lethal agents (e.g., PARP inhibitors), radiation, or a combination thereof. Preferably, the patient is untreated. 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., campotecan, topotecan, and irinotecan), DNA crosslinkers (e.g., mitomycin C), and triazene compounds (e.g., dacarbazine and temozolomide). A synthetic lethal therapeutic approach typically involves administering an agent that inhibits at least one key component of a biological pathway particularly important for the survival of certain tumor cells. For example, when tumor cells have homologous repair pathway deficiencies (e.g., homologous repair pathway deficiencies as defined in this invention), poly-ADP-ribose polymerase inhibitors (or platinum drugs, double-strand break repair inhibitors, etc.) may be particularly effective against such tumors because two pathways important for survival are blocked (one biologically, e.g., by BRCA1 mutations, and the other synthetically, e.g., by administration of pathway drugs). A synthetic lethal approach to 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 the positive LOH signature status, which forms the basis for classifying a patient as likely to respond to a particular cancer treatment regimen, include, but are not limited to, previous treatment outcomes, 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, serous papillary carcinoma, mucinous carcinoma, invasive ductal carcinoma, non-invasive ductal carcinoma in situ, etc.), disease stage, tumor or cancer malignancy grade (e.g., well-differentiated, moderately differentiated, or poorly differentiated (e.g., Gleason, modified Bloom Richardson), etc.), and the number of previous treatment courses.

[0066] Cancer patients can be treated with a cancer treatment regimen if they are classified as likely to respond to a particular cancer treatment regimen (e.g., a regimen involving the use of DNA damaging agents, PARP inhibitors, radiation, or a combination thereof). In some embodiments, the patient is a previously untreated patient. Cancer patients identified as having cancer cells with a positive LOH signature can be treated with any appropriate method for treating the cancer in question. For example, cancer can be treated with platinum-based chemotherapy agents or combinations of platinum-based chemotherapy agents as described elsewhere (see, for example, U.S. Patents 3,892,790, 3,904,663, 7,759,510, 7,759,488, and 7,754,684). In some cases, cancer can be treated with anthracyclines or combinations of anthracyclines as described elsewhere (see, for example, U.S. Patents No. 3,590,028, No. 4,138,480, No. 4,950,738, No. 6,087,340, No. 7,868,040, and No. 7,485,707). In some cases, cancer can be treated with topoisomerase I inhibitors or combinations of topoisomerase I inhibitors as described elsewhere (see, for example, U.S. Patents No. 5,633,016 and No. 6,403,563). In some cases, cancer can be treated with PARP inhibitors or combinations of PARP inhibitors as described elsewhere (see, for example, U.S. Patents No. 5,177,075, No. 7,915,280, and No. 7,351,701). In some cases, radiation can be used to treat cancer elsewhere as described (see, for example, U.S. Patent No. 5,295,944). In some cases, cancer can be treated with or without radiation therapy using combinations of different agents (for example, combinations of platinum-based chemotherapeutic agents, anthracyclines, topoisomerase I inhibitors, and / or PARP inhibitors).In some cases, combination therapy may include one of the aforementioned activators or therapies (e.g., DNA damaging agents, PARP inhibitors, radiation, or a combination thereof) together with another activator or therapy, such as taxanes (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-flourouracil, methotrexate).

[0067] In some cases, patients identified as having cancer cells with a genome lacking a LOH signature can be classified as less likely to respond to treatment regimens including DNA damaging agents, PARP inhibitors, radiation, or a combination thereof, at least partially based on their negative LOH signature state. Such patients can then be classified as more likely to respond to cancer treatment regimens including the use of one or more non-HDR-related cancer treatment agents, e.g., taxanes (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-flourouracil, methotrexate). In some embodiments, the patients are untreated. When a cancer patient is classified as likely to respond to a particular cancer treatment regimen (e.g., a regimen that includes the use of cancer treatment substances not associated with HDR), that cancer patient can be treated with such a regimen. Cancer patients identified as having cancer cells with a negative LOH signature can be treated using any method appropriate to the cancer being treated. Examples of information that can be used in addition to the negative LOH signature that forms the basis for classification as likely to respond to a particular cancer treatment regimen include, but are not limited to, previous treatment outcomes, 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, serous papillary carcinoma, mucinous carcinoma, invasive ductal carcinoma, ductal carcinoma in situ (non-invasive), etc.), stage, malignancy of the tumor or cancer (e.g., well-differentiated, moderately differentiated, or poorly differentiated (e.g., Gleason, modified Bloom Richardson), etc.), and the number of previous treatment courses.

[0068] After treating a patient for a specific period (e.g., 1 to 6 months), it is possible to evaluate the patient and determine whether the treatment regimen is effective. If a beneficial effect is detected, the patient can continue with the same or a similar cancer treatment regimen. If a minimal beneficial effect is detected, or if 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 may be increased. In some cases, additional anticancer drugs may be added to the treatment regimen, and a particular anticancer drug may be replaced with one or more different anticancer drugs. Patients being treated can be monitored as appropriate, and changes to the cancer treatment regimen can be made as needed.

[0069] In addition to predicting a likely therapeutic response or selecting a desirable treatment regimen, LOH signatures can be used to determine a patient's prognosis. As shown in Example 3 below (particularly Figure 18b), patients with tumors exhibiting an LOH signature have a significantly longer survival time than patients with tumors that do not exhibit such an LOH signature. Therefore, in one aspect, this document features a method for determining a patient's prognosis based at least in part on detecting the presence or absence of an LOH signature in a sample from the patient. The method comprises (a) determining whether a patient has cancer cells having an LOH signature as described herein (for example, the presence of more LOH regions than the reference number 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 total length of the chromosome containing the LOH region, 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) determining that the patient has a relatively good prognosis, at least partially based on the presence of an LOH signature, or (b) determining that the patient has a relatively poor prognosis, at least partially based on the absence of an LOH signature. Prognosis may include the likelihood of patient survival (e.g., progression-free survival, overall survival). A relatively good prognosis includes an increased likelihood of survival compared to a subset of reference populations (e.g., the average patient with this patient's cancer type / subtype, the average patient without a LOH signature, etc.). Conversely, a relatively poor prognosis in terms of survival includes a decreased likelihood of survival compared to a subset of reference populations (e.g., the average patient with this patient's cancer type / subtype, the average patient with a LOH signature, etc.).

[0070] As described herein, this document provides a method for evaluating a patient for cells (e.g., cancer cells) having a genome containing the LOH signature. In some embodiments, the patient is a previously untreated patient. For example, one or more clinicians or healthcare professionals can determine whether a patient has cancer cells with a genome containing the LOH signature. In some cases, one or more clinicians or healthcare professionals can determine whether a patient has cancer cells with a genome containing the LOH signature by obtaining a cancer cell sample from the patient and evaluating the genome of the cancer cells in the cancer cell sample to determine the presence or absence of the LOH signature as described herein.

[0071] In some cases, one or more clinicians or healthcare professionals may obtain cancer cell samples from a patient and provide the samples to a testing laboratory capable of evaluating the genome of the cancer cells in the cancer cell samples to indicate the presence or absence of the LOH signature as described herein. In some embodiments, the patient is a previously untreated patient. In such cases, one or more clinicians or healthcare professionals may determine whether the patient contains cancer cells with a genome containing the LOH signature by receiving information about the presence or absence of the LOH signature directly or indirectly from the testing laboratory. For example, after the testing laboratory has evaluated the genome of cancer cells for the presence or absence of the LOH signature as described herein, it may provide or make available to the clinician or healthcare professional a written, electronic, or oral report or medical record indicating the presence or absence of the LOH signature in a particular patient being evaluated. With such written, electronic, or oral reports or medical records, one or more clinicians or healthcare professionals may determine whether the particular patient being evaluated contains cancer cells with a genome containing the LOH signature.

[0072] If a clinician or healthcare professional, or a group of clinicians or healthcare professionals, determines that a particular patient being evaluated has cancer cells with a genome containing the LOH signature, the clinician or healthcare professional (or group) can classify that patient as having cancer cells with a genome containing the LOH signature. In some embodiments, the patient is a previously untreated patient. In some cases, a clinician or healthcare professional, or a group of clinicians or healthcare professionals, can diagnose a patient determined to have cancer cells with a genome containing the LOH signature as having cancer cells that are likely to be HDR-deficient. Such a diagnosis may be based solely on the determination that a particular patient being evaluated has cancer cells with a genome containing the LOH signature, or at least in part on the determination that a particular patient being evaluated has cancer cells with a genome containing the LOH signature. For example, a patient determined to have cancer cells with a genome containing a LOH signature may be diagnosed as likely to have HDR deficiency based on a combination of positive LOH signature status and deficiency 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 healthcare professional, or a group of clinicians or healthcare professionals, may diagnose a patient who has been determined to have cancer cells with a genome containing the presence of a LOH signature as having cancer cells that are likely to contain mutations in one or more HDR pathway genes. In some embodiments, the patient is a previously untreated patient. Such a diagnosis may be based solely on determining that a particular patient being evaluated has cancer cells with a genome containing the LOH signature, or at least in part on determining that a particular patient being evaluated has cancer cells with a genome containing the LOH signature. For example, a patient who has been determined to have cancer cells with a genome containing the presence of a LOH signature may be diagnosed as having cancer cells that are likely to contain mutations in one or more HDR pathway genes based on a combination of positive LOH-positive status and a family history of cancer, or the presence of behavioral risk factors (e.g., smoking).

[0074] In some cases, a clinician or healthcare professional, or a group of clinicians or healthcare professionals, may diagnose a patient who has been determined to have cancer cells with a genome containing the LOH signature as having cancer cells that are likely to respond to a particular cancer treatment regimen. In some embodiments, the patient is untreated. Such a diagnosis may be based solely on determining that a particular patient being evaluated has cancer cells with a genome containing the LOH signature, or at least in part on determining that a particular patient being evaluated has cancer cells with a genome containing the LOH signature. For example, a patient who has been determined to have cancer cells with a genome containing the LOH signature may be diagnosed as likely to respond to a particular cancer treatment regimen based on a combination of positive LOH signature status and deficiency status of one or more tumor suppressor genes (e.g., BRCA1 / 2, RAD51), a family history of cancer, or the presence of behavioral risk factors (e.g., smoking). As described herein, a patient determined to have cancer cells with a genome containing the presence of a LOH signature may be diagnosed as likely to respond to a cancer treatment regimen including the use of platinum-based chemotherapeutic agents, e.g., cisplatin, carboplatin, oxaliplatin, or picoplatin; anthracyclines, e.g., epirubicin or doxorubicin; topoisomerase I inhibitors, e.g., campotecan, topotecan, or irinotecan; PARP inhibitors; radiation; a combination thereof; or a combination of any of the aforementioned with another anticancer agent. In some embodiments, the patient is a previously untreated patient.

[0075] If a clinician or healthcare professional, or a group of clinicians or healthcare professionals, determines that a particular patient being evaluated has cancer cells with a genome lacking the LOH signature, the clinician or healthcare professional (or group) can classify that patient as having cancer cells with a genome that includes the absence of the LOH signature. In some embodiments, the patient is treatment-naïve. In some cases, a clinician or healthcare professional, or a group of clinicians or healthcare professionals, can diagnose a patient determined to have cancer cells with a genome lacking the LOH signature as having cancer cells that are likely to have functional HDR. In some cases, a clinician or healthcare professional, or a group of clinicians or healthcare professionals, can diagnose a patient determined to have cancer cells with a genome lacking the 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 healthcare professional, or a group of clinicians or healthcare professionals, may diagnose a patient who has cancer cells containing genomes lacking the presence of a LOH signature, or who has numerous LOH regions covering entire chromosomes, as having cancer cells less likely to respond to platinum-based chemotherapy agents, e.g., cisplatin, carboplatin, oxalaplatin, or picoplatin; anthracyclines, e.g., epiruvincin or doxorubicin; topoisomerase I inhibitors, e.g., campotecan, topotecan, or irinotecan; PARP inhibitors; or radiation, and / or a cancer treatment regime that includes the use of cancer treatment agents not related to HDR, e.g., one or more taxanes, growth factor inhibitors or growth factor receptor inhibitors, antimetabolites, etc. In some embodiments, the patient is a previously untreated patient.

[0076] As described herein, this document also provides a method for performing diagnostic analysis of nucleic acid samples from cancer patients (e.g., genomic nucleic acid samples or amplified genomic nucleic acid samples) to determine whether cancer cells in the patient have a genome containing a LOH signature and / or numerous LOH regions covering entire chromosomes. In some embodiments, the patient is a previously untreated patient. For example, one or more laboratory technicians or laboratory personnel can detect the presence or absence of a LOH signature in the genome of the patient's cancer cells, or the presence or absence of numerous LOH regions covering entire chromosomes in the genome of the patient's cancer cells. In some cases, one or more laboratory technicians or laboratory personnel can (a) receive a cancer cell sample obtained from a patient, or a genomic nucleic acid sample obtained from cancer cells obtained from a patient, or an enriched and / or amplified genomic nucleic acid sample obtained from cancer cells obtained from a patient, and (b) perform analysis (e.g., SNP array-based assay or sequencing-based assay) using the received material to detect the presence or absence of an LOH signature or a large number of LOH regions covering an entire chromosome, as described herein. In some cases, one or more laboratory technicians or laboratory personnel can receive the sample to be analyzed (e.g., a cancer cell sample obtained from a patient, a genomic nucleic acid sample obtained from cancer cells obtained from a patient, or an enriched and / or amplified genomic nucleic acid sample obtained from cancer cells obtained from a patient) directly or indirectly from a clinician or healthcare professional. In some embodiments, the patient is a previously untreated patient.

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

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

[0079] In some cases, a laboratory technician or laboratory worker, or a group of laboratory workers, can identify a patient with cancer cells detected to have the LOH signature as having cancer cells that are likely to respond to a particular cancer treatment regimen by associating the presence of a positive LOH signature, a potentially deficient HDR status, a potential deficiency in one or more HDR pathway genes, or the results (or a summary of the results) of a diagnostic analysis performed with the corresponding patient's name, medical record, symbol / numeric identifier, or a combination thereof. Such identification may be based solely on the detection of the presence of the LOH signature, or at least partially on the detection of the presence of the LOH signature. For example, a laboratory technician or laboratory worker may identify a patient with cancer cells detected to have the LOH signature as having cancer cells that are likely to respond to a particular cancer treatment regimen based on a combination of a positive LOH signature status and the results of other genetic and biochemical tests performed at the laboratory. In some embodiments, the patient is a previously untreated patient.

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

[0081] In some cases, a laboratory technician or laboratory worker, or a group of laboratory technicians or laboratory workers, may identify a patient with cancer cells detected as lacking a LOH signature as having cancer cells with potentially intact HDR pathway genes by associating the negative LOH signature status, the potential absence of genetic mutations in HDR pathway genes, or the results (or summaries of results) of a diagnostic analysis performed with the corresponding patient's name, medical record, symbol / numeric identifier, or a combination thereof.

[0082] In some cases, a laboratory technician or laboratory worker, or a group of laboratory technicians or laboratory workers, may associate a patient with cancer cells detected as lacking a LOH signature with the corresponding patient's name, medical record, symbol / numeric identifier, or a combination thereof, by associating the negative LOH signature status, potentially intact HDR status, potential absence of genetic mutations in HDR pathway genes, or the results (or summary of results) of the diagnostic analysis performed with the patient's name, medical record, symbol / numeric identifier, or a combination thereof. It is possible to identify cancer cells that are less likely to respond to platin, oxalaplatin, or picoplatin, anthracyclines (e.g., epiruvincin or doxorubicin), topoisomerase I inhibitors (e.g., campotecan, topotecan, or irinotecan), PARP inhibitors (e.g., iniparib, olaparib, or berapirib, or radiation), and / or are more likely to respond to a particular cancer treatment regimen (e.g., a cancer treatment regimen that includes the use of cancer treatment substances not associated with HDR). In some embodiments, the patient is a previously untreated patient.

[0083] When a laboratory technician or laboratory worker, or a group of laboratory technicians or laboratory workers, detects the presence of numerous LOH regions covering an entire chromosome, the laboratory technician or laboratory worker (or group) can identify a patient with cancer cells that have been detected to have numerous LOH regions covering an entire chromosome as likely to have cancer cells with intact BRCA1, BRCA2, and / or RAD51C status or intact HDR pathways. For example, one or more laboratory technicians or laboratory workers can identify a patient with cancer cells that have been detected to have numerous LOH regions covering an entire chromosome as likely to have cancer cells with intact BRCA1 and BRCA2 status by associating the presence of numerous LOH regions covering an entire chromosome or the results (or summary of results) of the diagnostic analysis performed with the corresponding patient's name, medical record, symbol / numeric identifier, or a combination thereof. In some embodiments, the patient is a previously untreated patient.

[0084] The results of any analysis according to the present invention are often communicated to a physician, genetic counselor, and / or patient (or other party such as a researcher) in a communicable form that can be conveyed or transmitted to any of the aforementioned parties. Such form may vary and may or may not be tangible. The results can be embodied in descriptions, diagrams, photographs, charts, images, or any other visual form. For example, when explaining the results, graphs or diagrams showing genotype or LOH (or HRD status) information may be used. Descriptions and visual forms can be recorded in tangible media, such as paper; computer-readable media, such as floppy disks, compact disks, flash memory, etc.; or in non-tangible media, such as electronic media in the form of email or on a website on the internet or intranet. Furthermore, the results can also be recorded in audio form and transmitted via any suitable medium, such as analog or digital cable lines, fiber optic cables, telephones, facsimile machines, wireless mobile phones, internet phones, etc.

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

[0086] Some aspects of the present invention described herein involve correlating a number of LOH signatures according to the present invention (e.g., the total number of LOH regions in at least one pair of human chromosomes from the cancer cells, where each LOH region is longer than a first length but shorter than the total length of the 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) with a particular clinical feature (e.g., an increased likelihood of deletion in the BRCA1 or BRCA2 gene; an increased likelihood of HDR deletion; an increased likelihood of response to a treatment regimen including DNA damaging agents, anthracyclines, topoisomerase I inhibitors, radiation, and / or PARP inhibitors, etc.) if the number is greater than any reference (or optionally, correlating a number with another feature if the number is smaller than any reference). Throughout this document, whenever such embodiments are described, other embodiments of the present invention may include, in addition to or instead of the correlating step, one or both of the following steps: (a) concluding that a patient has a clinical feature, at least in part on the presence or absence of a LOH signature; or (b) informing a patient that they have a clinical feature, at least in part on the presence or absence of a LOH signature.

[0087] As an example, one aspect described herein is a method for predicting a cancer patient's response to a cancer treatment regimen comprising a DNA damaging agent, anthracycline, topoisomerase I inhibitor, radiation, and / or PARP inhibitor. The method includes (1) determining the number of LOH regions in at least one pair of human chromosomes from cancer cells derived from the cancer patient, wherein the LOH region is longer than a first length but shorter than the total length of the 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 (2) correlating the total number exceeding a reference number with an increased likelihood that the cancer patient may respond to the cancer treatment regimen. According to the preceding paragraph, this description of this aspect is understood to include a description of two related aspects, namely 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: (1) determining the number of LOH regions in at least one pair of human chromosomes from cancer cells derived from the cancer patient, wherein the LOH region is longer than a first length but shorter than the total length of the 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 (2) (a) concluding that the cancer patient is likely to respond to the cancer treatment regimen, at least partially based on a total number exceeding a reference number, or (2) (b) conveying that the cancer patient is likely to respond to the cancer treatment regimen, at least partially based on a total number exceeding a reference number.

[0088] In each embodiment described in this document, such correlation, conclusion, or communication may involve correlating the output of a particular assay or analysis (e.g., the total number of LOH regions greater than a reference number) 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 further, or instead, concluding or communicating such clinical features based at least partially on the output of such particular assay or analysis, and such correlation, conclusion, or communication may include assigning a risk or probability of the clinical feature occurring based at least partially on the output of the particular assay or analysis. In some embodiments, such risk is the percent probability of the event or outcome occurring. In some embodiments, patients are assigned to risk groups (e.g., low risk, medium risk, high risk, etc.). In some embodiments, “low risk” is any percent 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” certain information means making such information known to another person or transferring such information to an object (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 prediction of prognosis or response (e.g., LOH signature according to the present invention) is communicated. This communication may be auditory (e.g., verbal), visual (e.g., written), electronic (e.g., data transferred from one computer system to another), etc. In some embodiments, communicating a 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 auditory report, or an electronic record. In some embodiments, the report is displayed and / or stored on a computer 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 communicating the classification is provided to the physician). In some embodiments, the cancer classification is communicated to a patient (e.g., a report communicating the classification is provided to the patient). Cancer classification can also be communicated by transferring information that embodies the classification (e.g., data) to a server computer and making such information accessible to an intermediate user or end user (e.g., by viewing the information when it is displayed on the server, or by downloading the information in the form of one or more files transferred from the server to the intermediate user or end user's device).

[0090] Wherever an aspect of the present invention involves concluding any facts (e.g., patient prognosis or the likelihood of a patient responding to a particular treatment regimen), this may, in some aspects, include a computer program that concludes such facts after performing an algorithm that typically applies information about the LOH region in accordance with the present invention.

[0091] In each aspect described herein involving a number of LOH regions (e.g., LOH index regions) or the entire length of such LOH regions, the present invention encompasses related aspects involving test values ​​or scores (e.g., HRD scores, LOH scores, etc.) derived from, incorporating, and / or reflecting, such numbers or lengths at least to some extent. In other words, in the various methods, systems, etc. of the present invention, it is not necessary to use the bare number or length of LOH regions, but rather test values ​​or scores derived from such numbers or lengths may be used.For example, one aspect of the present invention is a method for treating cancer in a patient, comprising the steps of: (1) determining the number of LOH regions in at least one pair of human chromosomes from cancer cells of a cancer patient in a sample derived from the patient, wherein the LOH region is longer than a first length but shorter than the total length of the chromosome containing the LOH region indicating that the cancer cell has an LOH signature, and 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; (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, quartile, quartile, quintile, etc.) derived from the number of LOH regions in a reference population; and (4) (a) the test value is greater than at least one of the reference values ​​(e.g., at least 2, 3, 4, 5, 6, 7, 8, 9, or 10 times the reference value; The method provides a step of administering an anticancer drug to the patient or recommending, prescribing, or initiating a treatment regimen that includes chemotherapy and / or synthetic lethal agents, based at least in part on the comparison step of determining that the test value is at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 standard deviations greater than the reference value; or (4)(b) recommending, prescribing, or initiating a treatment regimen that does not include chemotherapy and / or synthetic lethal agents, based at least in part on the comparison step of determining that the test value is at least 1 / 2, 1 / 3, 1 / 4, 1 / 5, 1 / 6, 1 / 7, 1 / 8, 1 / 9, or 1 / 10 of the reference value; and at least 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 standard deviations greater than the reference value. The present invention includes corresponding embodiments, with modifications as necessary, that use test values ​​or scores to determine the patient's prognosis, the patient's likelihood of responding to a particular treatment regimen, the likelihood that the patient or patient sample has BRCA1 deficiency, BRCA2 deficiency, RAD51C deficiency, or HDR deficiency, etc.

[0092] Figure 15 shows an exemplary process by which a computer computing system (or a computer program containing computer executable instructions (e.g., software)) can identify a LOH locus or region from genotype data as described herein. When the observed ratio of signals of two alleles A and B is 2:1, there are two possibilities. The first possibility is that cancer cells have LOH with a deletion of allele B in a sample contaminated 50% with normal cells. The second possibility is that there is no LOH, but allele A is duplicated in a sample not contaminated with normal cells. The process starts in box 1500. In box 1500, the following data are collected by the computer computing system: (1) sample-specific normalized signal intensities of both alleles at each locus and (2) assay-specific (different SNP array and sequence-based approach specific) parameter sets defined based on analysis of a large number of samples with known ASCN profiles. As described herein, loci can be evaluated for homozygosity or heterozygosity along chromosomes using any suitable assay method, such as SNP array-based assays or sequencing-based assays. In some cases, a system including a signal detector and a computer can be used to collect data on the homozygosity or heterozygosity of multiple loci (e.g., fluorescence signals or sequencing results) (e.g., sample-specific normalized signal intensities of both alleles at each locus). In box 1510, allele-specific copy numbers (ASCNs) are reconstructed at each locus (e.g., each SNP). ASCNs are the copy numbers of both the paternal and maternal alleles. In box 1530, a likelihood function is used to determine whether a locus or region of a homozygosity locus is due to LOH. This may be conceptually analogous to previously described algorithms designed to reconstruct the total copy number (rather than ASCNs) at each locus (e.g., SNP). See International Patent 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 across the entire genome, and the sample-specific noise level. In box 1540, the LOH region is determined as a SNP region where one of the ASCNs (paternal or maternal) is 0. In some embodiments, the computer process further includes the step of querying whether the patient is untreated or determining whether the patient is untreated.

[0093] Figure 3 shows an exemplary process in which a computer system can determine the presence or absence of a LOH signature. The process begins in box 300, where the computer system collects data on the homozygosity or heterozygosity of multiple loci along the chromosome. As described herein, any suitable assay method, such as an SNP array-based assay or a sequencing-based assay, can be used to evaluate the homozygosity or heterozygosity of loci along the chromosome. In some cases, a system comprising a signal detector and a computer can be used to collect data on the homozygosity or heterozygosity of multiple loci (e.g., fluorescence signals or sequencing results). In box 310, the computer system evaluates the data on the homozygosity or heterozygosity of multiple loci, as well as the position or spatial relationship of each locus, to determine the length of any LOH region present along the chromosome. In box 320, the computer system evaluates data regarding the number of detected LOH regions and the length of each detected LOH region to determine the number of LOH regions that (a) are longer than or equal to a preset number of Mb (e.g., 15 Mb) and (b) are shorter than the total length of the chromosome containing the LOH region. Alternatively, the computer system can determine the total length of the LOH as described above or calculate the sum of the lengths of the LOH regions. In box 330, the computer system formats an output indicating the presence or absence of LOH signatures. Once formatted, the computer system can present the output to a user (e.g., a laboratory technician, clinician, or healthcare professional). As described herein, the presence or absence of LOH signatures can be used to indicate a patient's probable HDR status, the likelihood of genetic mutations in HDR pathway genes, and / or potential cancer treatment regimens.

[0094] Figure 4 shows examples of computer devices 1400 and mobile computer devices 1450 that may be used in conjunction with the techniques described herein. Computer device 1400 is intended to be various types of digital computers, e.g., laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Computer device 1450 is intended to be various types of mobile devices, e.g., personal digital assistants, mobile phones, smartphones, and other similar computer devices. The components shown herein, their connections and relationships, and their functions are intended to be illustrative only and are not intended to limit the execution of the invention as described and / or claimed herein.

[0095] The computer calculator 1400 comprises a processor 1402, memory 1404, storage device 1406, a high-speed interface 1408 connected to memory 1404 and a high-speed expansion port 1410, and a low-speed bus 1414 and a low-speed interface 1415 connected to storage device 1406. Components 1402, 1404, 1406, 1408, 1410, and 1415 are interconnected using various buses and may be mounted on a common motherboard or, as appropriate, in other ways. The processor 1402 can process instructions to be executed within the computer calculator 1400, including instructions stored in memory 1404 or storage device 1406 for displaying graphical information for a GUI on an external input / output device, such as a display 1416 connected to the high-speed interface 1408. In other executions, multiple processors and / or multiple buses may be used, as appropriate, with multiple memories and memory types. Furthermore, multiple computer devices 1400 may be connected, each providing some of the necessary operations (for example, as a server bank, a collection of blade servers, or a multiprocessor system).

[0096] Memory 1404 stores information within the computer computing device 1400. In one execution, memory 1404 is a volatile memory unit. In another execution, memory 1404 is a non-volatile memory unit. 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 large-capacity storage to the computer calculator 1400. In one execution, the storage device 1406 may be a variety of devices, including computer-readable media such as floppy disk drives, hard disk drives, optical disk drives, or tape drives, flash memory or other similar solid-state memory devices, or devices in a storage network or other configuration. Computer program products can be incorporated into an information carrier so that they can be tactile. Computer program products may also include instructions that, when executed, perform one or more actions, for example, the actions described herein. The information carrier is computer-readable or machine-readable media such as memory 1404, storage device 1406, memory mounted on processor 1402, or propagating signals.

[0098] The high-speed controller 1408 manages bandwidth-intensive operations for the computer calculator 1400, while the low-speed controller 1415 manages less bandwidth-intensive operations. This assignment of functions is illustrative only. In one execution, the high-speed controller 1408 is connected to memory 1404, a 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 execution, the low-speed controller 1415 is connected to storage device 1406 and the low-speed expansion port 1414. The low-speed expansion port may have various communication ports (e.g., USB, Bluetooth, Ethernet, or wireless Ethernet) and may be connected, for example, via a network adapter to one or more input / output devices, e.g., a keyboard, pointing device, scanner, optical reader, fluorescence signal detector, or networking device, e.g., a switch or router.

[0099] As shown in the drawings, the computer calculator 1400 can be run in a number of different forms. For example, the computer calculator 1400 may run as a standard server 1420, and may run multiple times in a collection of such servers. The computer calculator 1400 may run as part of a rack server system 1424. Furthermore, the computer calculator 1400 may run in a personal computer, such as a laptop computer 1422. Alternatively, components from the computer calculator 1400 may be combined with other components in a mobile device (not shown), such as a device 1450. Each such device may comprise one or more computer calculators 1400, 1450, and the entire system may consist of multiple computer calculators 1400, 1450 communicating with each other.

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

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

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

[0103] Memory 1464 stores information within the computer computing device 1450. Memory 1464 can function as one or more computer-readable media, volatile memory units, or non-volatile memory units. An extended memory 1474 may be installed via an extended interface 1472 and connected to the device 1450. The extended interface 1472 may include, for example, a SIMM (Single In-Line Memory Module) card interface. Such an extended 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 extended memory 1474 may include instructions to perform or supplement the processes described herein and may also include secure information. Thus, for example, the extended 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. Furthermore, a secure application may be provided via the SIMM card along with additional information, such as storing identification information on the SIMM card to prevent hacking.

[0104] The memory may include, for example, flash memory and / or NVRAM memory, as discussed below. In one execution, the computer program product is incorporated into an information carrier so that it can be tangibly accessed. The computer program product includes instructions that, when executed, perform one or more actions, for example, the actions described herein. The information carrier is a computer-readable or machine-readable medium, for example, memory 1464, extended memory 1474, memory mounted on processor 1452, or a propagating signal that can be received, for example, a propagating signal that can be received by transceiver 1468 or external interface 1462.

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

[0106] The device 1450 may also communicate using an audio codec 1460. The audio codec 1460 can receive voice information from the user and convert it into usable digital information. Furthermore, the audio codec 1460 can generate audible sound for the user, for example, through a speaker, or on the handset of the device 1450. Such sound may include sounds from voice telephone calls, recorded audio (e.g., voicemails, music files, etc.), or audio created by applications running on the device 1450.

[0107] As shown in the drawings, the computer calculator 1450 can be implemented in a number of different forms. For example, the computer calculator 1450 may be implemented as a mobile phone 1480. The computer calculator 1450 may also be implemented as part of a smartphone 1482, a personal digital assistant, or other similar mobile device.

[0108] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuits, integrated circuits, specially designed ASICs (application-specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations may also include implementations in one or more computer programs executable and / or interpretable on a programmable system, which includes at least one programmable processor, which may be for special or general use, connected 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 can be executed in high-level procedural programming languages ​​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, device, and / or apparatus (e.g., magnetic disks, optical disks, memory, and programmable logic devices (PLDs)) that contains machine-readable medium that receives machine instructions as machine-readable signals. 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 performed on a computer having a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) and a keyboard and pointing device (e.g., a mouse or trackball) that allows the user to input into the computer. User interaction can also be provided using other types of devices. For example, the feedback provided to the user may 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, voice input, or tactile input.

[0111] The systems and techniques described herein can be executed in a computer computing system comprising a backend component (e.g., as a data server), a middleware component (e.g., an application server), or a frontend component (e.g., a client computer having a graphical user interface or web browser for a user to interact with the execution of the systems and techniques described herein), or any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by digital data communication (e.g., a communication network) in any form or medium. Examples of communication networks include local area networks ("LANs"), wide area networks ("WANs"), and the Internet.

[0112] A computer computing system may include clients and servers. Clients and servers are generally geographically separated and typically interact via a communication network. The client-server relationship arises from computer programs running on each computer that have a client-server relationship with one another.

[0113] In some cases, the computer computing systems provided herein may be configured to include one or more sample analyzers. The sample analyzers may be configured to generate multiple signals for genomic DNA of at least one pair of human chromosomes from cancer cells. For example, the sample analyzer may generate signals that can be interpreted in a way that identifies homozygosity or heterozygosity of loci along the chromosomes. In some cases, the sample analyzer may be configured to perform one or more steps of an 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 computer computing systems provided herein may be configured to include a computer computing device. In such cases, the computer computing device may be configured to receive signals from the sample analyzer. The computer computing device may include computer executable instructions or computer programs (e.g., software) containing computer executable instructions for performing one or more of the methods or steps described herein. In some cases, such computer executable instructions may instruct the computer computing device to analyze signals from the sample analyzer, another computer computing device, an SNP array-based assay, or a sequencing-based assay.By analyzing these signals, it is possible to determine the genotype, homozygosity at a particular locus, homozygous regions, the number of LOH regions, the size of LOH regions, the number of LOH regions of a specific size or size range, whether a sample is LOH signature positive, the number of indicator LOH regions in at least one pair of human chromosomes, the possibility of deletion in the BRCA1 and / or BRCA2 genes, the possibility of deletion in HDRs, the likelihood of a cancer patient responding to a particular cancer treatment regimen (e.g., a regimen including DNA damaging agents, anthracyclines, topoisomerase I inhibitors, radiation, PARP inhibitors, or a combination thereof), or a combination of these items.

[0114] In some cases, the computer computing system provided herein may include computer executable instructions or computer programs (e.g., software) that include computer executable instructions for formatting an output showing the number of LOH regions, the size of LOH regions, the number of LOH regions having a particular size or size range, whether a sample is LOH signature positive, the number of indicator LOH regions in at least one pair of human chromosomes, the possibility of deletion in the BRCA1 gene and / or BRCA2 gene, the possibility of deletion in HDR, the possibility that a cancer patient will respond to a particular cancer treatment regimen (e.g., a regimen including DNA damaging agents, anthracyclines, topoisomerase I inhibitors, radiation, PARP inhibitors, or a combination thereof), or combinations thereof. In some cases, the computer computing system provided herein may include computer executable instructions or computer programs (e.g., software) that include computer executable instructions for determining a cancer treatment regimen desirable for a particular patient, at least in part based on the presence or absence of an LOH signature or the number of indicator LOH regions.

[0115] In some cases, the computer systems provided herein may include preprocessing devices configured to process a sample (e.g., cancer cells) so that an SNP array-based assay or a sequencing-based assay can be performed. Examples of preprocessing devices include, but are not limited to, devices configured to enrich a cell population with respect to cancer cells as opposed to non-cancerous cells, devices configured to lyse cells and / or extract genomic nucleic acids, and devices configured to enrich a sample with respect to a specific genomic DNA fragment.

[0116] This document also provides kits for evaluating samples (e.g., cancer cells) as described herein. For example, this document provides kits for evaluating cancer cells for the presence of LOH signatures, or for determining the number of indicator LOH regions in at least one pair of human chromosomes. The kits provided herein may include either SNP probes (e.g., arrays of SNP probes for performing the SNP array-based assay methods described herein) or primers (e.g., primers designed to sequence SNP regions via a sequencing-based assay method) in combination with a computer program product that includes computer-executable instructions for performing one or more of the methods or steps described herein (e.g., computer-executable instructions for determining the number of LOH regions having a particular size or size range). In some cases, the kits provided herein may include at least 500, 1,000, 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, 1,000, 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 assays or sequencing-based assays. Examples of such other components include, but are not limited to, buffers, sequencing nucleotides, and enzymes (e.g., polymerases). This document also provides the use of any appropriate number of materials provided herein in the preparation of kits for performing one or more of the methods or steps described herein. For example, this document provides the use of SNP probe collections (e.g., collections of 10,000 to 100,000 SNP probes) and computer program products provided herein in the preparation of a kit for evaluating cancer cells for the presence of a LOH signature.As another example, this document provides the use of primer collections (e.g., collections of 10,000 to 100,000 primers for sequencing SNP regions) and computer program products provided herein in the manufacture of a kit for evaluating cancer cells for the presence of LOH signatures.

[0117] The present invention is further illustrated in the following embodiments. The following embodiments do not limit the scope of the present invention as defined in the claims. [Examples]

[0118] Example 1 - Evaluation of LOH area and HDR Two sets of tumors from patients with advanced ovarian cancer were used. Candidate signatures were derived using the first set (training set), consisting of 94 tumors, and the signatures were validated using the second set (validation set), consisting of 40 tumors. All coding regions of the BRCA1 and BRCA2 genes were sequenced to detect germline and somatic mutations. mRNA expression levels of BRCA1 and BRCA2 were measured and analyzed using Affymetrix SNP microarrays.

[0119] Using a computer program, the LOH signature state was reconstructed based on allele intensities obtained from microarray data. An algorithm was developed to reconstruct the LOH region based on genotype (e.g., SNP genotype) data, and this was executed as a computer program.

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

[0121] For the purposes of this analysis, a tumor was defined as HDR-deficient if it had one or more harmful mutations in the BRCA1 and / or BRCA2 genes, or if it had low BRCA1 mRNA expression. The remaining tumors were defined as likely to be HDR-non-deficient for the purposes of this analysis.

[0122] The distribution of LOH region lengths was examined (Figure 5). Three classifications of LOH regions were used: (1) LOH affecting the entire chromosome; (2) large LOH regions (longer than approximately 15 Mb) typically affecting part or the entire chromosome arm; and (3) multiple short LOH regions (shorter than approximately 15 Mb). Next, using only the training set, the number of LOH regions in each of these three classifications was evaluated for its potential correlation with HDR deficiency. It was found that (1) the number of short LOH regions did not significantly correlate with HDR deficiency (p>0.05); (2) LOH covering the entire chromosome was weakly correlated with HDR deficiency (p=0.0011); and (3) the number of large LOH regions was significantly correlated with HDR deficiency (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), while the majority of tumors likely to be HDR-non-deficient had a small number of large LOH regions (Figures 6-8). Tumors that were likely to be HDR-free were almost certainly HDR-deficient due to other genetic changes besides BRCA1 and BRCA2 mutations and low mRNA expression. In addition to the number of large LOH regions, the full length of these regions also significantly correlated with HDR deficiency.

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

[0124] 134 tumors were divided into three groups based on a combination of training and validation datasets: (1) BRCA deficiency if there was one or more harmful mutations in the BRCA1 and / or BRCA2 genes or low BRCA1 mRNA expression; (2) HDR deficiency / BRCA intact if there were nine or more large LOH regions (longer than 15 Mb but shorter than the total chromosome length); and (3) HDR intact if there were fewer than nine large LOH regions (longer than 15 Mb but shorter than the total chromosome length). The results of this analysis are shown in Figure 9. Figure 9 shows that BRCA deficiency and HDR deficiency not related to BRCA deficiency are frequent in ovarian tumors.

[0125] Figure 10 shows the distribution of large LOH regions (longer than 15 Mb but shorter than the total length of the chromosome) in different types of cancer cell lines. The size of the circles is proportional to the number of samples with such a large number of LOH regions. The frequency of HDR deficiency (cell lines with at least 9 such large LOH regions) was highest in breast cancer and esophageal cancer cell lines. HDR deficiency was not observed in colon cancer cell lines. Review of previous findings in ovarian tumors revealed that all BRCA-deficient cell lines were also HDR-deficient.

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

[0127] Figure 12 summarizes the analysis results for various tumors and cell lines. For several tumors and cell lines, the frequency of HDR deficiency was defined as the percentage of samples containing at least nine large LOH regions (longer than 15 Mb but shorter than the total chromosome length). This frequency was high (50%) in ovarian tumors and not observed at all in brain and colon cell lines. Therefore, HDR deficiency is thought to play a significant role in the majority of cancers.

[0128] Example 2 - Chemical Toxicity Reaction In preparing chemical toxicity reaction experiments, all cell lines were placed in a 75 cm³ solution. 2 Cells were grown in tissue culture flasks (VWR International, Inc. catalog number 353136) and recommended growth medium at 37°C + 5% CO2. Before each experiment, each cell line was trypsinized (Invitrogen Corporation catalog number 25200-056), counted, and seeded in Advanced RPMI 1640 (Invitrogen Corporation catalog number 12633-020), 3% FBS, 1% penicillin / streptomycin (Invitrogen Corporation catalog number 15140-122) in 100 μL of medium per well in columns 2-12 of a 96-well polystyrene microplate with a clear bottom (Perkin Elmer catalog number 600518). Column 1 contained only 100 μL of medium per well. The seeded plates were then incubated overnight at 37°C + 5% CO2.

[0129] Two working stocks with different final drug concentrations were prepared. Advanced RPMI 1640 was used as the diluent for the highest concentration when 100% DMSO was required for drug dissolution. For lower concentrations, a predetermined amount of DMSO equal to the total DMSO in the high-concentration working stock (Advanced RPMI 1640 + DMSO) was used. For the lowest concentration, up to 60% DMSO was used. This was done to maintain equal DMSO concentrations per well and prevent nonspecific cell death resulting from DMSO. The lower of the two drug concentrations was placed in rows A-D and column 12 of a 96-well thin-walled PCR cycle plate (Robbins Scientific catalog number 1055-00-0), while the higher concentration was placed in rows E-H and column 12 of the same plate. Serial dilutions of 1:2 or 1:3 were performed in descending order in columns 12-13. Columns 1 and 2 were used as cell-free / drug-free and drug-control-free. This allowed for quadruplet data points for each drug concentration. Once dilution was complete, 5 μL was transferred from the dilution plate to the corresponding wells of the seeded cell plate. The plates containing the drug were then incubated at 37°C + 5% CO2 for 3 or 6 days.

[0130] Following a 3-day or 6-day dosing regimen, the ATPlite assay (Perkin Elmer catalog number 6016941) was performed on each well of each plate according to the ATPlite assay protocol. Luminescence was then read using a FUSION instrument and saved as a .CSV file. For each cell line and drug combination, the average of four replicates of the drug-free control was divided by 100 to create a "normalization factor" used to calculate normalized percentage survival. The normalized percentage survival of the drug-free control was 100%. For each drug concentration, the average of four replicates of the cell + drug wells was divided by the normalization factor. Using the percentage survival for each drug concentration, starting from a concentration equal to 0, an IC (Identification of Concessions) was performed using software that allows for ownership claims. 50 The result was calculated.

[0131] Figure 13 shows the response of breast cancer and ovarian cancer cell lines to chemotherapy. The y-axis represents the logarithm of various chemotherapy drugs (camptothecin, and the mean of results for platinum compounds (oxaliplatin, cisplatin, and carboplatin), or anthracyclines (doxorubicin and epirubicin)) when exposed to 29 breast cancer cell lines. 10 (I C 50 ) value, and the Log of paclitaxel when exposed to 27 ovarian cancer cell lines 10 (I C 50 The values ​​were displayed. The x-axis showed the number of large LOH regions that were longer than 15 Mb and shorter than the entire chromosome for these cell lines. A dashed line indicated a threshold of 9.

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

[0133] Example 3 - Further validation of the HR deficiency assay method material and method Ovarian tumor sample Three independent human ovarian cancer cohorts were used: 1) 152 unselected ovarian cancer samples; 2) 53 high-grade serous ovarian tumors; and 3) publicly available data from 435 serous ovarian cancer samples with complete information, 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. Various sample sizes were used in the assay methods described (Table 3).

[0134] (Table 2) Patient and cancer characteristics TIFF0007844597000003.tif176156TIFF0007844597000004.tif175156

[0135] (Table 3) Number of samples used in each assay method TIFF0007844597000005.tif126161

[0136] cell line Sixty-seven cancer cell lines were analyzed (29 ovarian cancer cell lines, 34 breast cancer cell lines, 3 colon cancer cell lines, and 1 pancreatic cancer cell line). Three breast cancer cell lines were obtained from DSMZ (Braunschweig, Germany). The colon cancer cell lines, pancreatic cancer cell lines, and the remaining breast cancer cell lines were obtained from ATCC (Manassas, VA). The cancer cell lines were placed in T75 flasks and incubated in RPMI + 10% FBS + 1% penicillin / streptomycin medium at 37°C for approximately 5 × 10⁻¹⁴°C. 6 Cells were grown to cell density. Exceptions were cell lines requiring non-standard media, L-glutamine, or insulin. Cells grown in suspension were placed in a 1.5 mL centrifuge tube and centrifuged at 1700 rpm for 5 minutes, and the supernatant was discarded. The medium was removed from cells grown in monolayers by aspirate, washed with PBS, and trypsin solution was added. After detaching the cells, they were collected in medium, transferred to a 1.5 mL microcentrifuge tube, and centrifuged at 1700 rpm for 5 minutes. The supernatant was discarded. Isolated cells were resuspended in 200 μL of PBS.

[0137] Extraction of genomic DNA and total RNA from frozen tumors and cell lines 10 μm frozen sections were cut and microdisected. The tissue was homogenized after adding QIAzol lysis reagent (TissueRuptor (Qiagen)), and RNA was subsequently isolated using the Qiagen miRNAeasy Mini Kit according to the manufacturer's protocol. DNA was isolated using the QIAamp DNA Mini Kit (Qiagen) by overnight lysis incubation at 56°C and RNase A treatment according to the manufacturer's protocol.

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

[0139] Promoter methylation qPCR assay method Methylation levels were quantified using the Methyl-Profiler DNA Methylation PCR Array System (SABiosciences) according to the manufacturer's recommended protocol. Unmethylated or methylated genomic DNA was selectively digested using DNA methylation-sensitive restriction enzymes and DNA methylation-dependent restriction enzymes, respectively. The digested DNA was quantified by real-time PCR using primers adjacent to the region of interest. The relative concentrations of differentially methylated DNA were determined by comparing the amount of each digest with the amount of a mock digest.

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

[0141] Expression assay method for BRCA1 and cell cycle progression signatures RNA was treated with Amplification Grade Deoxyribonuclease I (Sigma-Aldrich Inc.) for a long incubation period of 30 minutes, 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] Repeated preamplification was independently performed using the Taqman PreAmp Master Mix Kit (Applied Biosystems) protocol at a reaction volume of 5 μL. For cell cycle gene assays, preamplification was performed for 8 and 18 cycles, respectively. Only the BRCA1 assay underwent three preamplification repeats over 18 cycles. The amplified products were diluted 1:5 with low-EDTA Tris-EDTA (TE). Quantitative polymerase chain reaction (qPCR) was then performed and evaluated using Gene Expression M48 Dynamic Arrays (Fluidigm, South San Francisco, CA) according to the manufacturer's protocol. The comparative cycle threshold (C) was measured. T Relative gene expression was calculated using the ) method. C from pre-amplification of different cycle numbers. T s was centered by the average of the genes in the repeats, which is common to all repeats. The resulting value was first the average C of the housekeeper gene. T s, then, the normalized C of each assay for all samples derived from the first cohort T Normalized by the mean of s, ΔΔC T The following was obtained: CCP score and relative BRCA1 expression were measured using the ΔΔC of the cell cycle gene and BRCA1 assay methods, respectively. T It was calculated as the average of the negative values ​​of s.

[0143] Identification of samples in which BRCA1 expression is lost: Samples in which CCP expression and BRCA1 expression were anticorrelated 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 regression-estimated against the CCP score using repeatedly 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 Patent Application No. PCT / US2011 / 054369 to Timms et al.

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

[0145] CN analysis and LOH analysis of SNP microarray data This algorithm is designed to determine the most likely allele-specific copy number at each SNP site. The corresponding probabilities explicitly take into account contamination of cancer DNA samples with non-cancerous stromal cell DNA. A similar algorithm for CN analysis is detailed in international application number PCT / US2011 / 026098 (publication number WO / 2011 / 106541) to Abkevich et al. The algorithm used in this document was performed 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). The latter array contains numerous probes for non-polymorphic sites across the entire human genome, in addition to SNP probes. These probes are useful for CN analysis but not directly useful for LOH analysis.

[0146] statistical analysis In this document, p-values ​​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 if BRCA1 was methylated or its mRNA was low. Of the 53 samples from the second cohort, 14 were carriers of BRCA1 and / or BRCA2 mutations, and of the 435 samples from the third cohort, 83 were carriers of BRCA1 and / or BRCA2 mutations (two of which were excluded from further analysis; see below). The mutations are summarized in Table 4.

[0148] (Table 4) Abnormalities of BRCA1, BRCA2, and RAD51C detected in the study cohort TIFF0007844597000006.tif53158 1 Since one copy of BRCA2 remained intact, two of these mutations were excluded from the analysis.

[0149] Methylation levels were measured for both BRCA1 and BRCA2 promoter CpG islands. Methylation was observed in multiple samples for BRCA1, but not for BRCA2. Of the 126 samples from the first cohort, 11 were defined as HR deficiency due to high levels of BRCA1 promoter methylation. Of the 34 samples from the second cohort, 3 were defined as HR deficiency due to high levels of BRCA1 promoter methylation. No harmful BRCA1 or BRCA2 mutations were observed in any of these samples, with the exception of one sample from the third cohort.

[0150] Low mRNA expression of BRCA1 or BRCA2 may also lead to HR deficiency, and this 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 deficiency due to BRCA1 promoter methylation. No abnormally low expression was observed for BRCA2.

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

[0152] Distribution of LOH region lengths The initial hypothesis was that regions with LOH of varying lengths may appear in the cancer genome via different pathways, and therefore, the relationship between LOH and HR deletion may be determined by the length of the LOH region. Figure 16 shows the distribution of LOH region lengths corrected for the length of the chromosome arms in which these LOH regions were observed. Chromosomes 13, 14, 15, and 22 were excluded because SNPs are not available on the short arms of these chromosomes. Three distinctive features were observed in this distribution. First, there are many short LOH regions. Second, there are long, flattened LOH region tails that are the length of a single chromosome arm. There are very few LOH regions that are longer than a single chromosome arm but shorter than the entire chromosome. Finally, there is a high peak corresponding to LOH spanning the entire chromosome. The observed distribution is completely different from a similar distribution obtained for CN variability (Beroukhim et al. 2010). This suggests that CN variability and LOH regions may arise through 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 the entire chromosome 17 in almost all samples, the LOH region on this chromosome was excluded from the analysis, presumably because genes important for ovarian cancer progression are thought to be located on this chromosome. The inventors investigated the correlation between HR deficiency and the number of short LOH regions (<15Mb), the number of long LOH regions (>15Mb but shorter than the entire chromosome), and the number of LOH regions covering the entire chromosome. Various different LOH region length cutoffs can be used. The effect of these cutoffs on the detection of HR deficiency was investigated in Figure 19 and its accompanying discussion, and 15Mb was found to be generally preferred. There was no significant correlation between the number of short LOH regions and HR deficiency. The number of LOH regions covering the entire chromosome was significantly higher in tumors with intact BRCA1 or BRCA2 (p=4×10⁻⁶). -5The number of long LOH regions (referred to as the "HRD score" after this example 3 and throughout this document) was significantly higher in tumors with BRCA1 or BRCA2 deficiency (p=9×10). -11 (Figure 17a).

[0154] The results obtained for the first cohort were validated using the second and third cohorts. The correlation between HR deletion and the number of LOH regions covering the entire chromosome was not validated in the second cohort, probably due to the small sample size, but was significantly larger in the third cohort among tumors with intact BRCA1 and BRCA2 (p=3×10⁻¹⁰). -11 A strong and significant correlation was observed between HRD scores and HR absence in both cohorts (p=2×10 for each). -7 and p=9×10 -30 ), the HRD score was significantly lower among ovarian tumors with intact BRCA1 and BRCA2 (Figures 17b and 17c).

[0155] Changes in RAD51C and other HR pathway genes Available data suggest that BRCA1 and BRCA2 are the main genes responsible for HR deficiency 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 identified as predisposing genes for ovarian cancer. In the first cohort, methylation levels were measured for promoter CpG islands of eight additional genes involved in the HR pathway (Table 5). Only RAD51C showed high levels of promoter methylation (3 out of 89 samples). In the third cohort, 11 out of 435 samples showed RAD51C promoter methylation. All RAD51C methylation-positive samples from both cohorts were homozygous at the RAD51C locus for LOH. To test whether HRD scores were higher in samples with RAD51C promoter methylation, these samples from both cohorts were compared to intact BRCA samples without RAD51C methylation. Consistent with our observations regarding the BRCA1 and BRCA2 genes, HRD scores were significantly higher in samples with RAD51C methylation (p=0.0003).

[0156] (Table 5) Promoter methylation assay method used (SABiosciences) TIFF0007844597000007.tif94158

[0157] In the third cohort, harmful mutations and methylation of HR pathway genes were reported (TCGA, 2011). Mutations were investigated and analyzed, limiting the analysis to abnormalities likely to be harmful (e.g., nonsense mutations and frameshift mutations). As a result, a total of eight harmful mutations were found in six genes (ATM, ATR, FANCA, FANCD2, FANCM, and PALB2). Methylation of HR pathway genes was found in five additional samples. Loss of the second allele was detected in only one of the thirteen samples (FANCM nonsense mutation). Since deactivation of both alleles is necessary to eliminate the function of tumor suppressor, most of these thirteen samples are expected to have intact HR. Unsurprisingly, the HRD scores were not high in the majority of these samples.

[0158] Analysis of combined data Figure 17d shows the correlation between HRD scores and HR deficiency (defined as BRCA1 deficiency, BRCA2 deficiency, or RAD51C deficiency) in all three cohorts. A highly significant relationship was observed (p=2×10⁻¹⁰). -54 ).

[0159] A key question is whether the distribution of HRD scores is the same for HR deficiencies caused by different genomic loci. To answer this, the distribution of HRD scores for BRCA1-deficient tumors, BRCA2-deficient tumors, and RAD51C-deficient tumors was analyzed separately (Figure 21). A significant difference was observed (p=7×10⁻⁶). -5 The mean HRD score (16.1; SD=4.3) of BRCA1-deficient samples was higher than that of BRCA2-deficient samples (13.0; SD=3.9). The difference in HRD scores between either BRCA1 or BRCA2 and RAD51C (14.5; SD=5.1) was not statistically significant.

[0160] Normal tissue samples were available from some samples from the first two cohorts and from all samples from the third cohort. These were used to determine whether the BRCA1 and BRCA2 mutations were germline or somatic mutations. There was no significant difference between somatic and germline HRD score distributions for either BRCA1 or BRCA2 deficiency (Figure 20).

[0161] HRD scores in BRCA1-deficient and BRCA2-deficient cell lines Unselected mammary gland cell lines (n=34) and ovarian cell lines (n=29) were obtained from multiple sources. Three additional colon cell lines and one pancreatic cell line obtained from NCI60, whose BRCA1 and BRCA2 statuses were publicly available, were analyzed. Of these 67 cell lines, seven had homozygous harmful mutations or BRCA1 promoter methylation, two had homozygous mutations with obvious reversal mutations, 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. There was no significant difference in the distribution of HRD scores between wild-type ovarian tumors and wild-type cancer cell lines. The distribution of HRD scores among cancer cell lines with heterozygous mutations was similar to that of wild-type cancer cell lines. This is likely because cells become HR deficient only when both copies of BRCA1 or BRCA2 become non-functional. Cancer cell lines with functional loss of both copies of either BRCA1 or BRCA2 showed high HRD scores similar to those observed in ovarian tumors with BRCA1-deficient, BRCA2-deficient, or RAD51C-deficient genes. Cancer cell lines with BRCA1 and BRCA2 reverse mutations also showed high HRD scores. This supports the initial hypothesis that HR deficiency leads to irreversible LOH changes. The difference in HRD score distribution between wild-type or heterozygous mutant cell lines and homozygous mutant cell lines (with or without reverse mutations) or BRCA1 promoter methylation is extremely significant (p=10). -5Importantly, after removing ovarian cancer cell lines from the dataset, there was a significant correlation between HRD scores and BRCA1 and BRCA2 deficiencies (p=0.01). This suggests that the relationship between HRD scores and HR deficiencies is not limited to ovarian cancer.

[0162] Correlation between HR deficiency and overall survival (OS) and progression-free survival (PFS) In the third cohort, where survival improved in patients with high HRD scores, PFS (p=0.03) and OS (p=6×10) were observed. -5 A significant correlation was observed between (Figure 18b). The p-value was calculated using the Cox model. This result is consistent with and expands upon 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 HRD scores were validated in two independent ovarian cancer datasets. The HRD scores also reflected mutations causing HR deficiency in mammary and pancreatic cell lines.

[0164] (Table 6) Mean HRD scores and corresponding p-values ​​for tumors lacking BRCA1 and BRCA2, and tumors with intact BRCA1 and BRCA2. TIFF0007844597000008.tif127159

[0165] Intermediate LOH sizes, longer than 15 Mb but shorter than the entire chromosome, were strongly positively correlated with the HR gene deficiency. This suggests that most, though not all, of this type of LOH class exist to incorporate double-strand 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 for whole-chromosome LOH is that it arises through an alternative competitive mechanism that does not involve double-strand DNA breaks.

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

[0167] Among tumors clearly possessing intact BRCA1, BRCA2, and RAD51C, a significant proportion of samples exhibit high HRD scores. One possible explanation is that a considerable proportion of these samples also have abnormalities in other HR pathway genes. Another explanation is that tumor contamination by 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 explain a significant proportion of high-HRD-scoring samples (see supplementary results).

[0168] Published studies have demonstrated that secondary reverse mutations restoring BRCA2 function can occur in BRCA2 mutant cell lines after platinum exposure (Sakai et al., 2009; Sakai et al., 2008; Edwards et al., 2008). Norquist et al., (2011) observed that approximately 28% of recurrent tumors had secondary mutations restoring BRCA function. Reverse mutations were primarily found in individuals pre-exposed to platinum and predicted platinum resistance. The HRD score is due to cumulative abnormalities occurring in the tumor genome. Since DNA-based HR deficiency markers are functionally linked to HR deficiency, they are likely to be strongly associated with HR deficiency. As a result, the HRD score is an extremely robust measure of HR deficiency. However, its persistence suggests that the score may have low sensitivity to reverse mutations. Post-treatment samples were not available from the tumors used in this study. However, the data obtained from the cell lines are consistent with this hypothesis. Failure to detect a reverse mutation results in a false positive. This is likely to affect very small tumors in neoadjuvant or adjuvant settings (Norquist et al., 2011) and is less significant than false negatives, which mistakenly identify individuals as promising non-responders.

[0169] High HRD scores are strongly correlated with HR deficiency. This score can be used to identify patients who are likely to respond to DNA damaging agents and PARP inhibitors (among the active ingredients). Such testing is clearly clinically useful in breast and ovarian cancer and can be used to expand the use of PARPi and platinum salts to other cancers where HR deficiency is not well-characterized.

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

[0171] Fifty-six fresh-frozen mammary gland tumors were analyzed. The median contamination level was 60%. Nine samples were contaminated to at least 90%. Of these tumors, 11 were carriers of BRCA1 adverse mutations and 3 were carriers of BRCA2 adverse mutations. All of these tumors had LOH in the deletion gene. One of the BRCA1 adverse mutation carriers also had a BRCA2 adverse mutation. However, this sample did not have LOH in the BRCA2 gene.

[0172] Thirty samples were obtained from patients who responded to treatment (residual tumor burden of 0 or 1). Of these, 13 were BRCA1 / 2 deficiency. Twenty-six samples were obtained from non-responders (residual tumor burden of 2 or 3). Of these, one was BRCA1 deficiency. Genotyping analysis was performed using Affymetrix with an Affymetrix MIP array (U.S. Patent 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; all of these are incorporated herein by reference). HRD scores were calculated as described above.

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

[0174] There was no significant difference in HRD scores between samples with residual tumor volume of 0 and 1. Similarly, there was no significant difference in HRD scores between samples with residual tumor volume of 2 and 3. There was no significant correlation between treatment response and clinical parameters (disease stage, malignancy grade).

[0175] Other embodiments Although the present invention has been described in conjunction with a detailed description of the invention, it should be understood that the foregoing description is illustrative and does not limit the scope of the invention. The scope of the invention is defined by the appended claims. Other aspects, advantages, and modifications are within the scope of the appended claims.

Claims

1. A method for predicting mutations in homologous recombination repair (HDR) pathway genes in patient cancer cells, including the following steps: (a) A step of detecting LOH regions in at least one pair of human chromosomes of cancer cells or genomic DNA derived therefrom, wherein the at least one pair of human chromosomes are not a human X / Y sex chromosome pair, and (b) A step of determining the total number of LOH regions in at least one pair of human chromosomes, wherein the LOH region is longer than 5 megabases but shorter than the total length of the chromosome containing the LOH region, and (c) A step in which the total number of LOH regions exceeds a reference number, indicating the possibility of mutations in the HDR pathway genes in the patient's cancer cells.

2. The method according to claim 1, wherein the cancer cells are primary cancer cells or metastatic cancer cells of ovarian cancer, breast cancer, lung cancer, or esophageal cancer.

3. The method according to claim 2, wherein the cancer cells are ovarian cancer cells, and the predicted mutation is a mutation in the BRCA1 gene, the BRCA2 gene, or the RAD51 gene.

4. The method according to claim 3, wherein the RAD51 gene is the RAD51C gene.

5. The method according to claim 1, wherein the cancer cells are breast cancer cells.

6. The method according to any one of claims 1 to 5, wherein the LOH region is determined in at least two pairs, five pairs, ten pairs, or 21 pairs of human chromosomes.

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

8. The method according to any one of claims 1 to 7, wherein the LOH region is not present on human chromosome 17.

9. The method according to any one of claims 1 to 8, further comprising the step of determining, based on the number of LOH regions less than a reference number, that the patient is unlikely to respond to a treatment regimen comprising a DNA damaging agent, anthracycline, a topoisomerase I inhibitor, radiation, and / or a PARP inhibitor.

10. The method according to any one of claims 1 to 8, further comprising the step of determining, based on the number of LOH regions less than a reference number, that the patient is likely to respond to a cancer treatment regime comprising a taxane, a growth factor inhibitor or growth factor receptor inhibitor, or an antimetabolite.

11. The method according to any one of claims 1 to 10, wherein the LOH region is longer than 15 megabases.

12. A method for predicting the status of the BRCA1 or BRCA2 gene in a patient's cancer cells, including the following steps: (a) A step of detecting LOH regions in at least one pair of human chromosomes of cancer cells or genomic DNA derived therefrom, wherein the at least one pair of human chromosomes are not a human X / Y sex chromosome pair, and (b) A step of determining the total number of LOH regions in at least one pair of human chromosomes, wherein the LOH region is longer than 5 megabases but shorter than the total length of the chromosome containing the LOH region, and (c) A step in which the possibility of a deletion in the BRCA1 or BRCA2 gene in the patient's cancer cells is indicated when the total number of LOH regions exceeds the reference number.

13. The method according to claim 12, wherein the cancer cells are primary cancer cells or metastatic cancer cells of ovarian cancer.

14. The method according to claim 12 or 13, further comprising the step of determining, based on the number of LOH regions exceeding a reference number, that the patient is likely to respond to a cancer treatment regime comprising a DNA damaging agent, anthracycline, a topoisomerase I inhibitor, radiation, and / or a PARP inhibitor.

15. (a) The DNA damaging agent is cisplatin, carboplatin, oxalaplatin, or picoplatin; the anthracycline is epirubicin or doxorubicin; the topoisomerase I inhibitor is campotecin, topotecan, or irinotecan; or the PARP inhibitor is olaparib or berapirib. (b) The method according to claim 14, wherein the DNA damaging substance is a platinum-based chemotherapeutic agent.

16. The method according to claim 12 or 13, further comprising the step of determining, based on the number of LOH regions smaller than a reference number, that the patient is likely to respond to a cancer treatment regime comprising a taxane, a growth factor inhibitor or growth factor receptor inhibitor, or an antimetabolite.

17. (a) Whether the taxane is doxetaxel, paclitaxel, or abraxane, (b) The growth factor inhibitor or growth factor receptor inhibitor is erlotinib, gefitinib, lapatinib, sunitinib, bevacizumab, cetuximab, trastuzumab, or panitumumab, (c) The method according to claim 16, wherein the antimetabolite is 5-fluorouracil or methotrexate.

18. The method according to any one of claims 12 to 17, wherein the LOH region is longer than 11 megabases.

19. The method according to any one of claims 1 to 18, wherein the reference number is 5, 6, 7, 8, or 9.

Images