METHODS FOR DETECTING INACTIVATION OF THE HOMOLOGOUS RECOMBINATION PATHWAY (BRCA1/2) IN HUMAN TUMORS

Abstract

The invention relates to methods for detecting inactivation of the DNA Homologous Recombination pathway in a patient, and in particular for detecting BRCA1 inactivation.

Claims

1. A method for treating cancer, the method comprising a therapeutically effective amount of a PARP inhibitor and/or an alkylating agent to a human patient identified as having, in a tumor sample obtained from the patient, a number, per genome, of large scale transitions (LSTs) greater than a predetermined threshold number of LSTs, wherein an LST is a breakpoint between two genomic regions of different copy number, each such genomic region greater than or equal to 3 and less than 6 megabases long.

2. The method of claim 1, wherein said PARP inhibitor and/or alkylating agent is selected from the group consisting of iniparib, olaparib, rucaparib, CEP 9722, MK 4827, BMN-673, 3-aminobenzamide, platinum complexes, chlormethine, chlorambucil, melphalan, cyclophosphamide, ifosfamide, estramustine, carmustine, lomustine, fotemustine, streptozocin, busulfan, pipobroman, procarbazine, dacarbazine, thiotepa and temozolomide.

3. The method of claim 1, wherein the cancer is selected from breast cancer, ovary cancer, pancreas cancer, head and neck carcinoma and melanoma.

4. The method of claim 1, wherein the cancer is breast cancer.

5. The method of claim 1, wherein the cancer is basal-like breast cancer.

6. The method of claim 1, wherein the patient is identified by detecting, in the tumor sample, the number of LSTs per genome.

7. The method of claim 6, wherein the number of LSTs per genome is detected by detecting copy number for at least 500 Single Nucleotide Polymorphism (SNP) loci.

8. The method of claim 6, wherein the number of LSTs per genome is detected by detecting copy number for at least 3,000 Single Nucleotide Polymorphism (SNP) loci.

9. The method of claim 6, wherein the number of LSTs per genome is detected by comparative genomic hybridization (CGH) array, Single Nucleotide Polymorphism (SNP) array, or sequencing of polymorphic loci.

Description

FIGURE LEGENDS

[0124] FIG. 1. Chromosome content and BRCA1 status in BLCs. A. Distribution of the chromosome content in the set of BLCs displayed two modes, which evidences 2 populations of tumors with different ploidy status. B. Near-diploid tumors (<50 chromosomes) and over-diploid tumors (>=50 chromosomes) showed different proportions of proven BRCA1-inactivated tumors. WT correspond to non BRCA1.

[0125] FIG. 2. Genomic instability in over-diploid BLCs as estimated by the total number of breaks and by LSTs. LST number clearly discriminated non-BRCA1 BLCs from BLCs with proven BRCA1 inactivation (p-value<0.001, Wilcoxon test). Total number of breaks was less significantly different between non-BRCA1 vs BRCA1 and meBRCA1 comparison (p-value<0.03, Wilcoxon test) and was not discriminative. BRCA1: germline BRCA1 mutation; meBRCA1: BRCA1 promoter methylation; sporadic=non-BRCA1: absence of evidence of BRCA1 inactivation.

[0126] FIG. 3. Tumor ploidy and the number of large-scale transitions (LST) are discriminative of BRCA1 inactivation in the experimental (left) and validation (right) sets. Upper panel: number of LSTs per tumor is indicated in relation to ploidy categories. Near-diploid and near-tetraploid cutoffs are indicated by a bar. Known BRCA1 and BRCA2 statuses are indicated for germline mutations (“BRCA1” and “BRCA2”), methylation of the BRCA1 promoter (“meBRCA1”) and mutations in the tumors (“tumBRCA1”). Tumors without evidence of BRCA1/2 inactivation are referred to as “non-BRCA1”. Fisher's exact tests are indicated below the contingency tables; BRCA1 refers to all proven BRCA1-inactivated BLCs, non-BRCA1 refers to BLCs without evidence of BRCA1 inactivation.

[0127] FIG. 4. Genomic and functional assessments of BRCAness in basal-like cell lines. A. Cell lines with basal-like phenotype display discriminative features of BRCAness similar to primary BLCs. Known status for BRCA1 and BRCA2 are indicated for germline mutations (“BRCA1” and “BRCA2”) and methylation of BRCA1 promoter (“meBRCA1”). Cell lines without evidence of BRCA1/2 inactivation are described as “non-BRCA1/2”. B. RAD51 foci formation 8 hours after 10 Gy irradiation illustrates active homologous recombination (HR) in non-BRCA1 cell lines, and conversely deficient HR in BRCA1 or BRCA2 mutated cell lines. 53BP1 foci in the same experiment are shown as a control for DNA damage response. Scale bars, 20 μm). Number of LST is indicated as well as BRCA1/2 status: mut, mutated; me, methylation of the promoter; wt, wildtype.

[0128] FIG. 5. Survival curves for LST_high and LST_low ovarian tumors. P-value was estimated by log-rank test statistic.

[0129] FIG. 6. Event free survival curves for LST_high and LST_low ovarian tumors. P-value was estimated by log-rank test statistic.

[0130] FIG. 7. LST_10 Mb in tumor cell lines.

[0131] Calculated ploidy is indicated (2N pseudo-diploid, 4N pseudo-tetraploid). Triangle: wild-type or unknown BRCA1/2 status; square: BRCA2 mutated cell lines.

EXAMPLES

Example 1

[0132] Materials and Methods

[0133] Patients and Tumors

[0134] A series of undifferentiated grade 3 BLCs was assembled from patients who had surgery at the Institut Curie. According to French regulations patients were informed of research and did not express opposition. High quality biological material was available at Institut Curie biobank for 85 tumor samples (some samples were described previously)..sup.28-30 This series was enriched for tumors arisen in patients carrying deleterious BRCA1 mutations (35 tumors).

[0135] Immunohistochemistry

[0136] Immunostaining was performed on 4 μm tissue sections as described previously:.sup.28,29 ER, PR and ERBB2 (Novocastra), EGFR and KRT8/18 (Zymed, Invitrogen), KRT5/6 (Dako) and KRT14 (Biogenex). Positivity for each marker was determined according to standardized guidelines. Negativity was defined as total absence of staining for expression of ER and PR, and as less than 2+ staining for ERBB2.

[0137] The basal-like phenotype was defined according to morphological, phenotypic and/or molecular criteria including i) high grade (Elston-Ellis grading) and pushing margins, ii) triple-negative phenotype and expression of either KRT5/6/14/17 or EGFR assessed by immunohistochemistry..sup.32

[0138] Methylation Status of BRCA1 Promoter

[0139] Methylation of the promoter of BRCA1 was assessed by methyl-specific PCR (MSP) after bisulfite conversion as described previously,.sup.33 with minor modifications (primer sequences and protocols are available upon request).

[0140] BRCA1 Mutation Status

[0141] Pre-screen for mutations of the BRCA1 gene was performed using Enhanced Mismatch Mutation Analysis (EMMA, Fluigent.sup.34; EMMALYS software P/N: 5331254102). For abnormal EMMA profiles, the concerned BRCA1 exons were sequenced with dideoxynucleotides (BigDye Terminator V1.1, Applied Biosystems, Foster City, Calif.), according to standard protocols (primer sequences and protocols are available upon request). Sequences were examined with the Seqscape V2.5 (Applied Biosystems).

[0142] Analysis of Transcriptomic Data

[0143] Transcriptomic data was obtained on the Affymetrix U133plus2 platform in the Institut Curie according to the standard protocol. Normalization was performed with BrainArray algorithm.sup.35. Unsupervised clustering was performed based on the intrinsic signature.sup.36.

[0144] Processing the Genomic Profiles

[0145] Genomic profiling of 85 BLCs was performed using two platforms: Illumina (Illumina SNP HapMap 300K Duo, 33 cases) and Affymetrix (Affymetrix SNP Chip 6.0, 52 cases).

[0146] Illumina platform: Genomic profiling of the tumor samples was performed by a service provider (Integragen, Evry, France) on 300K Illumina SNP-arrays (Human Hap300-Duo). Raw data files were processed by BeadStudio 3.3 in standard settings using supporting data provided by Illumina (HumanHap300v2_A). Allele specific signals (X and Y in BeadStudio notation) were processed into Log R ratio and B allele frequency by tQN algorithm..sup.37

[0147] Affymetrix platform: Hybridization was performed at Institut Curie on Affymetrix SNPChip6.0 array. Cell files were processed by Genotyping Console 3.0.2. Log 2Ratio and Allele Difference profiles resulted from Copy Number and LOH analysis performed with the reference model file HapMap270 (GenomeWideSNP_6.hapmap270.na29) provided by Affymetrix.

[0148] Quality control: 20 SNP arrays were discarded due to: low hybridization quality (3 arrays); low tumor content and/or ambiguous profile interpretation (17 arrays).

[0149] Segmental copy number and genotype detection: Both Illumina and Affymetrix SNP array data were mined using the GAP method described and validated previously: segmental copy numbers, allelic contents (major allele counts) and normal cell contamination were detected; segmentations were optimized with respect to the genomic status detected..sup.27

[0150] Recognition of absolute copy number ranged from 0 to 8 copies with all segments exceeding 8-copy level been ascribed 8-copy status. Thus, 22 possible segmental genotypes were discriminated (copy number/major allele count): 1 copy A (or 1/1); 2 copies AA (2/2) and AB (2/1); 3 copies AAA (3/3), AAB (3/2); 4 copies AAAA (4/4), AAAB (4/3), AABB (4/2), etc. . . .

[0151] Chromosome number: Number of chromosomes was estimated by the sum of the copy numbers detected at the pericentric regions. The status of the pericentric region of each chromosome arm was defined by the corresponding juxta-centromeric segment when the latter contained 500 SNPs or more. When not measurable, missing values were substituted by the modal copy number of the considered chromosome arm (3.4±2.2 out of 41 chromosome arms per genome were substituted in the series). Chromosome counting procedure was validated by comparing estimated chromosome numbers versus available numbers from karyotype or SKY data for 25 breast cancer cell lines {http://www.lgcstandards-atcc.org/}. Error rate was less than 2 chromosomes per sample (1.58±2.3).

[0152] Breakpoint counts: Number of breakpoints in each genomic profile was estimated based on the resulting interpretable copy number profile and after filtering less than 50 SNPs variation. Small interstitial alterations were defined as <3 Mb alterations surrounded by the segments with identical status for genotype and copy number. They were removed when estimating total number of breakpoints. Large-scale State Transitions (LSTs) were calculated after smoothing and filtering of variation less than 3 Mb in size.

[0153] Compilation of Validation Sets

[0154] The validation series comprises 55 samples including TNBC from a cohort of young women with breast cancer (17 cases); BLCs with medullary features (8 cases) and one BLC arisen in a BRCA2 mutation carrier; BRCA1 BLCs from GEO GSE19177 (12 cases).sup.38; basal-like tumors from GEO GSE32530 (4 cases).sup.39. BRCA1 BLCs from Institut Bergonid (5 cases).

[0155] Basal-like cell lines with available SNP array profile comprised 17 cases (15 cases hybridized in Institute Curie and 2 cases were obtained from the Wellcome Trust Sanger Institute Cancer Genome Project web site.

[0156] Results

[0157] BRCA1 Status of Basal-Like Carcinomas (BLCs)

[0158] A series of 65 well characterized basal-like breast carcinomas included 23 tumors arisen in patients carrying deleterious BRCA1 mutations (herein called “BRCA1 BLCs”) and 42 BLCs arisen in patients without evidence of familial predisposition of breast/ovarian cancer or tested negative for BRCA1/2 mutations (herein called “sporadic BLCs”). Sporadic BLCs were tested for the methylation of the BRCA1 promoter and nearly 25% were found positive (11 out of 41 tested, herein called “meBRCA1 BLCs”). No evidence of methylation in the remaining 31 cases was found. BRCA1 status was confirmed by the gene expression in 35 out of 36 tested cases with available transcriptomic data. BRCA1 and meBRCA1 BLCs comprise the group of tumors with proven BRCA1 inactivation (34 cases), which were further compared to the group of presumably non-BRCA1 BLCs (31 cases).

[0159] Near-Diploidy in BLCs has 75% Positive Predictive Value of BRCA1 Inactivation

[0160] In order to get insight into the specific genomic alterations of BLCs, genomic profiling was performed using SNP-arrays, which provide two complementary measurements: copy number variation and allelic imbalance. Genome Alteration Print (GAP) methodology for mining SNP arrays.sup.27 allowed us to obtain the segmental genotype profiles (i.e. exact copy numbers and allelic contents: A, AB, AA, AAB, AAA, . . . ) for each sample. General genomic characteristics such as number of chromosomes, DNA index, number of chromosome breaks, and proportions of genome in each genomic state were inferred from the segmental genotype profiles.

[0161] Estimated chromosome counts per genome showed a bimodal distribution (FIG. 1, top panel) similar to those demonstrated for the genomes in various types of cancers.sup.40. Tumor genomes carrying less than 50 chromosomes and with the DNA index close to 1 were considered to have ploidy of two and were thereafter called “near-diploid genomes” (23 cases). Following the hypothesis of the whole genome duplication during cancer progression explaining the second mode in chromosome distribution.sup.40 tumor genomes carrying more than 50 chromosomes and DNA index higher than 1.2 were considered to have a ploidy of four and were thereafter called “over-diploid genomes” (42 cases).

[0162] Interestingly, the 23 near-diploid tumors almost consistently carried germline mutation or epigenetic inactivation of BRCA1 (20/23) in contrast to the over-diploid tumors, which were slightly enriched in non-BRCA1 BLCs (28/42) (FIG. 1, bottom panel). Taking into account the fact that BRCA1 germline mutation is responsible for near 10% of basal-like carcinomas.sup.41 positive predictive value of genomic near-diploid status was estimated to be 75%.

[0163] Large-Scale Chromosomal Rearrangements Discriminate BRCA1 and Non-BRCA1 Basal-Like Carcinomas

[0164] Total number of breakpoints detected in the cancer genome characterizes the level of genomic instability. However, overall comparison of BRCA1 versus non-BRCA1 tumors did not show any significant difference (p-value=0.28). In the subgroup of 42 over-diploid BLCs, 14 BRCA1-inactivated tumors displayed elevated total number of breakpoints (range [57-224], 140.6±45.7), while 28 non-BRCA1 tumors showed significant heterogeneity (range [8-213], 101.2±50.6) and were enriched in the low values compared to BRCA1 tumors (p<0.017, Wilcoxon rank test). However, large overlap in the breakpoint numbers precluded accurate demarcation.

[0165] In order to get a robust and discriminative estimation of the genomic instability we evaluated the number of Large-scale State Transitions (LSTs) by calculating chromosomal breaks between adjacent regions of at least 10 Mb (comprising ˜3000 SNPs in Affymetrix SNP6.0).

[0166] Number of LSTs in the subgroup of over-diploid tumors had a bimodal distribution with a clear gap between two modes (12.5±4.9 and 35.5±6.7) separating 18 non-BRCA1 BLCs from the mixture containing 14 BRCA1-inactivated tumors and 10 tumors with neither BRCA1 germline mutation nor BRCA1 promoter methylation (FIG. 2). In the subgroup of 23 near-diploid BLCs, which mainly contained BRCA1 tumors, LSTs had unimodal distribution (28.0±6.5) with two non-BRCA1 tumors within one standard deviation (24 and 28 LSTs) and one non-BRCA1 BLC below two standard deviations from the average (12 LSTs). Interestingly, all tumors with low LSTs had no evidence of BRCA1 inactivation and displayed either few chromosomal breaks and a high level of aneuploidy (3 samples) or firestorm-like alterations (16 samples).

[0167] To conclude, LSTs reflected well the overall genomic patterns of the tumors, contrary to the total number of breakpoints, and provided the discriminative values for BRCA1 status prediction.

[0168] A Two-Step Decision Rule Consistently Detects BRCA1 Inactivation in BLCs.

[0169] Based on the LSTs distributions described above, two thresholds for BRCAness prediction were applied, more than 15 LSTs per genome in the near-diploid cases and more than 20 LSTs in the over-diploid cases, predicting BRCAness with 100% sensitivity (p-value=4*10.sup.−5, Fisher test).

[0170] Moreover, all “False Positive” cases (thereafter called “BRCA1-looking” BLCs) had similar high number of LSTs as the “True Positive” cases (with proven BRCA1-inactivated status), which actually questioned their false positive status and might evidence other mechanisms of homologous recombination defect including BRCA1 or BRCA2 mutations. Such mutations were searched in 28 sporadic BLCs with available material including 13 cases with the BRCA1-looking pattern. Deleterious BRCA1 mutations were found in six cases all belonging to BRCA1-looking tumors (p-value=0.02). Deleterious BRCA2 mutations were found in three cases all belonging to BRCA1-looking tumors. With these findings specificity reached 89% (p-value=1.4*10.sup.−11, Fisher test) in the considered experimental set of BLCs (FIG. 3A).

[0171] A validation series of 55 BLC/TNBC was assembled, including 15 cases with BRCA1 germline mutations, 15 cases with BRCA1 promoter methylation, 1 case with a BRCA2 germline mutation, and 24 sporadic cases. SNP array data were processed using the same workflow. Prediction of the BRCA1 inactivation displayed sensitivity of 100% (all 30 BRCA1 inactivated cases were predicted to be BRCA1-looking) and specificity of 80% (11 cases were predicted to be BRCA1-looking with yet no evidence of BRCA1 inactivation) (FIG. 3B; p-value=1.7*10.sup.−6, Fisher test). Noteworthy, the BRCA2 mutated tumor was near-diploid with a high LST number, thus clearly following a BRCA1-looking pattern.

[0172] Model Systems Supported the Discriminative Features Observed in the Primary Tumors A series of 17 basal-like cell lines was analyzed, including MDA-MB-436 and HCC1937 bearing BRCA1 mutations.sup.42 and HCC38 with BRCA1 promoter methylation.sup.43. The obtained results followed the trend found in primary tumors: firstly the only near-diploid cell line found was the BRCA1 mutated MDA-MB-436; secondly among over-diploid cell lines, HCC1937 and HCC38 carried the highest number of large-scale chromosomal breaks, which is again consistent with their BRCA1-inactivated status. Nevertheless, and as expected considering cell line establishment and long term maintenance in culture, the cutoff separating non-BRCA1 cell lines was found shifted to 23 LSTs (FIG. 4). One cell line HCC1599 had LST number very close to BRCA1 inactivated cell lines, whereas not associated with BRCA1/2 mutation.sup.44. To clarify the BRCA1 function and more precisely the homologous recombination pathway, RAD51 foci were measured 8 hours after ionizing radiations in BLC cell lines. All cell lines without BRCA1 looking pattern had the expected RAD51 foci accumulation, whereas no foci were observed in cell lines with BRCA1 looking pattern, including HCC1599 (data not shown).

[0173] In conclusion, the inventors have shown that it is possible to predict tumor deficiency in the DNA homologous recombination (HR) pathway in a patient suffering from cancer, by quantifying the number of rearrangements in the genomic DNA of a tumor sample obtained from said patient, wherein the number of rearrangements corresponds to the number, per genome, of breakpoints resulting in segments of at least 10 megabases.

[0174] Similar results were obtained by using a cutoff value between 3 megabases and 20 megabases for the definition of Large Scale Transitions.

Example 2—Performance of LST Number Predicting BRCAness in all Types of Breast Carcinomas

[0175] The series of 426 breast tumors (invasive ductal carcinomas including HER2-positive tumors, luminal (eg expressing receptors for estrogen or progesterone), triple negative/basal-like breast carcinoma (eg expressing no hormone receptors and not overexpressing HER2) as well as rare subtypes such as medullary carcinomas or micropapillary carcinomas from Institut Curie) was considered.

[0176] The series was enriched with BRCA1 and BRCA2 mutated tumors. The cut-offs on the LST number predicting BRCAness were inferred based on this series (Table 1). False Positive and True Positive Rates (FPR and TPR) show the quality of LST based predictor of BRCAness.

TABLE-US-00001 TABLE 1 Cut-offs for breast cancer BRCAness prediction based on the LST number LST_S Ploidy 2: (P = 68, N = 182) Ploidy 4: (P = 53, N = 123) Mb, S Cut-Off* FPR TPR Cut-Off FPR TPR 6 19 (17) 0.04 0.99 32 (32) 0.10 1 7 17 (15) 0.05 0.99 29 (27) 0.07 0.98 8 14 (14) 0.06 1 26 (26) 0.08 1 9 14 (11) 0.04 0.99 25 (19) 0.07 0.98 10 11 (11) 0.07 1 22 (18) 0.06 0.98 *Cut-offs correspond to max(TPR-FPR); cut-offs in parenthesis correspond to 100 sensitivity. P: Number of positives, i.e. BRCA1/2 mutated tumors; N: Number of negatives, i.e. number of tumors with BRCA1/2 wild-type or status not available; TPR: True positive rate; FPR: False positive rate.

Example 3—the Number of LSTs is a Good Predictor of Response to Treatment

[0177] Two publically available data sets from clinical trial of Cisplatin treatment of patients with triple-negative breast tumors [GSE28330 GEO database][59] were processed and the number of LST_10 Mb was calculated for each tumor with good quality of measured profile. Genomic profiles were measured by two types of chip: Affymetrix Oncoscan 70K (Dataset 2) and Oncoscan 300K (Dataset 1). Information about mutational status of BRCA1/2 was available for some tumors. Response to treatment was measured by Miller-Payne score, where 4 and 5 were considered as “positive response”, while scores<4 were considered as “no response” [59] Case by case and summary results are presented in Table 2 and Tables 3-5 (statistical comparisons were performed by the Fisher exact test). To conclude, (i) almost all known BRCA1/2 inactivated cases (17/18) and 15 tumors with wild-type or unknown BRCA1/2 status were classified as LST_high (Table 3); (ii) BRCA1/2 inactivation does not always mean response to Cisplatin (Table 4); (iii) LST_10Mb is a better cisplatin response predictor than the BRCA1/2 status (Table 4-5).

TABLE-US-00002 TABLE 2 Individual results Miller- Data Recognition Payne set ID Quality BRCA1/2 response LST Response 1 DFHCC_06.202_45R good 5 High Yes 1 DFHCC_06.202_15 good mut 5 High Yes 1 DFHCC_06.202_41 good 5 High Yes 1 DFHCC_06.202_7 good mut 5 High Yes 1 DFHCC_06.202_17 good 5 High Yes 2 DFHCC_04.183_9T good non 5 High Yes 2 DFHCC_04.183_18T good mut 5 High Yes 2 DFHCC_04.183_3T good non 5 High Yes 2 DFHCC_04.183_29T good non 5 High Yes 2 DFHCC_04.183_5T good mut 5 High Yes 2 DFHCC_04.183_17T good met 5 High Yes 1 DFHCC_06.202_6 good met 4 High Yes 1 DFHCC_06.202_48 good met 4 High Yes 2 DFHCC_04.183_7T good met 4 High Yes 2 DFHCC_04.183_8T good met 4 High Yes 1 DFHCC_06.202_40 good 4 High Yes 2 DFHCC_04.183_10T good non 4 High Yes 1 DFHCC_06.202_3 good 4 High Yes 1 DFHCC_06.202_27 good 4 Low Yes 1 DFHCC_06.202_13 good met 3 High No 1 DFHCC_06.202_5 good 3 Low No 1 DFHCC_06.202_4 good met 3 High No 2 DFHCC_04.183_23T good met 3 High No 2 DFHCC_04.183_11T good non 3 High No 2 DFHCC_04.183_25T good met 3 High No 2 DFHCC_04.183_1T good met 3 High No 1 DFHCC_06.202_37 good 3 Low No 1 DFHCC_06.202_20 good mut 2 High No 1 DFHCC_06.202_42 good mut 2 High No 1 DFHCC_06.202_21 good 2 High No 2 DFHCC_04.183_14T good non 2 High No 2 DFHCC_04.183_24T good non 2 Low No 2 DFHCC_04.183_22T good non 2 Low No 2 DFHCC_04.183_28T good non 2 Low No 1 DFHCC_06.202_24 good 2 Low No 1 DFHCC_06.202_10 good 1 Low No 1 DFHCC_06.202_32 good 1 Low No 1 DFHCC_06.202_35 good 1 Low No 1 DFHCC_06.202_46 good 1 Low No 2 DFHCC_04.183_13T good non 1 Low No 1 DFHCC_06.202_34 good 1 High No 1 DFHCC_06.202_29 good 1 High No 1 DFHCC_06.202_45L good 1 High No 2 DFHCC_04.183_4T good non 1 High No 2 DFHCC_04.183_12T good non 1 Low No 1 DFHCC_06.202_18 good 1 Low No 1 DFHCC_06.202_9 good 1 Low No 2 DFHCC_04.183_16T good non 1 Low No 1 DFHCC_06.202_14 good 1 Low No 2 DFHCC_04.183_6T good 1 Low No 1 DFHCC_06.202_28 good 0 Low No 2 DFHCC_04.183_21T good non 0 High No 2 DFHCC_04.183_27T good non 0 Low No 2 DFHCC_04.183_26T good met 0 Low No 2 DFHCC_04.183_15T bad met 0 No 2 DFHCC_04.183_20T bad non 2 No 2 DFHCC_06.202_33 good NA 2 DFHCC_06.202_43 good NA 2 DFHCC_06.202_50 good NA 2 DFHCC_06.202_39 bad 2 No 2 DFHCC_06.202_39 bad 2 No

TABLE-US-00003 TABLE 3 Summary of LST versus BRCA1/2 ALL LST_high LST_Iow BRCA1/2 18 1 p < 0.0001 NON BRCA1/2 or NA 15 20

TABLE-US-00004 TABLE 4 Summary of BRCA1/2 versus Response ALL Responders Non Responders BRCA1/2 9 8 p < 0.06 NON BRCA1/2 or NA 10 27

TABLE-US-00005 TABLE 5 Summary of LST versus Response ALL LST_high LST_Iow Non Responders 15 20 p < 0.0001 Responders 18 1

Example 4—LST in Ovarian Carcinoma

[0178] Series of high grade ovarian carcinoma from Institut Curie were profiled by SNP arrays (Affymetrix CytoScanHD). All patients were treated by chemotherapies including platinum salts. Tumor genomes were annotated as LST_high (50 cases) and LST_low (20 cases) based on the LST_6 Mb with the cutoffs 19 and 32 LSTs for near-diploid and near-tetraploid tumors respectively. Comparison of Overall Survival and Event Free Survival showed better outcome for patients with LST_high tumors, which indicates better response to treatment (FIGS. 5-6).

Example 5—LST in Tumor Cell Lines

[0179] Series of tumor cell lines with known BRCA status and with available SNP-array data were analyzed. LST_10 Mb was calculated and samples with high LST were linked to BRCA2 inactivation in cervix and pancreatic carcinoma cell lines. Two lung cell lines without known BRCA1/2 mutations have a high level of LST, presumably due to BRCA1 methylation described in this disease [60] (FIG. 7).

[0180] This validation of the method in tumor cell lines of various origins and state of differentiation indicates that LST measurement and prediction of the BRCAness can be applied in all types of tumors.

REFERENCES

[0181] 1. Rakha E A, Reis-Filho J S, Ellis I O: Basal-like breast cancer: a critical review. J Clin Oncol 26:2568-81, 2008 [0182] 2. Dawson S J, Provenzano E, Caldas C: Triple negative breast cancers: clinical and prognostic implications. Eur J Cancer 45 Suppl 1:27-40, 2009 [0183] 3. Foulkes W D, Stefansson I M, Chappuis P O, et al: Germline BRCA1 mutations and a basal epithelial phenotype in breast cancer. J Natl Cancer Inst 95:1482-5, 2003 [0184] 4. Bergamaschi A, Kim Y H, Wang P, et al: Distinct patterns of DNA copy number alteration are associated with different clinicopathological features and gene-expression subtypes of breast cancer. Genes Chromosomes Cancer 45:1033-40, 2006 [0185] 5. Melchor L, Honrado E, Garcia M J, et al: Distinct genomic aberration patterns are found in familial breast cancer associated with different immunohistochemical subtypes. Oncogene 27:3165-75, 2008 [0186] 6. Natrajan R, Weigelt B, Mackay A, et al: An integrative genomic and transcriptomic analysis reveals molecular pathways and networks regulated by copy number aberrations in basal-like, HER2 and luminal cancers. Breast Cancer Res Treat, 2009 [0187] 7. Gudmundsdottir K, Ashworth A: The roles of BRCA1 and BRCA2 and associated proteins in the maintenance of genomic stability. Oncogene 25:5864-74, 2006 [0188] 8. Roy R, Chun J, Powell S N: BRCA1 and BRCA2: different roles in a common pathway of genome protection. Nat Rev Cancer 12:68-78, 2012 [0189] 9. Turner N, Tutt A, Ashworth A: Hallmarks of ‘BRCAness’ in sporadic cancers. Nat Rev Cancer 4:814-9, 2004 [0190] 10. Chin S F, Teschendorff A E, Marioni J C, et al: High-resolution aCGH and expression profiling identifies a novel genomic subtype of ER negative breast cancer. Genome Biol 8:R215, 2007 [0191] 11. Stefansson O A, Jonasson J G, Johannsson O T, et al: Genomic profiling of breast tumours in relation to BRCA abnormalities and phenotypes. Breast Cancer Res 11:R47, 2009 [0192] 12. Joosse S A, Brandwijk K I, Mulder L, et al: Genomic signature of BRCA1 deficiency in sporadic basal-like breast tumors. Genes Chromosomes Cancer 50:71-81, 2011 [0193] 13. Farmer H, McCabe N, Lord C J, et al: Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 434:917-21, 2005 [0194] 14. Bryant H E, Schultz N, Thomas H D, et al: Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase. Nature 434:913-7, 2005 [0195] 15. Fong P C, Boss D S, Yap T A, et al: Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med 361:123-34, 2009 [0196] 16. Vollebergh M A, Jonkers J, Linn S C: Genomic instability in breast and ovarian cancers: translation into clinical predictive biomarkers. Cell Mol Life Sci 69:223-45, 2012 [0197] 17. Focken T, Steinemann D, Skawran B, et al: Human BRCA1-Associated Breast Cancer: No Increase in Numerical Chromosomal Instability Compared to Sporadic Tumors. Cytogenet Genome Res 135:84-92, 2011 [0198] 18. Johannsdottir H K, Jonsson G, Johannesdottir G, et al: Chromosome 5 imbalance mapping in breast tumors from BRCA1 and BRCA2 mutation carriers and sporadic breast tumors. Int J Cancer 119:1052-60, 2006 [0199] 19. Tirkkonen M, Johannsson O, Agnarsson B A, et al: Distinct somatic genetic changes associated with tumor progression in carriers of BRCA1 and BRCA2 germ-line mutations. Cancer Res 57:1222-7, 1997 [0200] 20. Melchor L, Alvarez S, Honrado E, et al: The accumulation of specific amplifications characterizes two different genomic pathways of evolution of familial breast tumors. Clin Cancer Res 11:8577-84, 2005 [0201] 21. Wessels L F, van Welsem T, Hart A A, et al: Molecular classification of breast carcinomas by comparative genomic hybridization: a specific somatic genetic profile for BRCA1 tumors. Cancer Res 62:7110-7, 2002 [0202] 22. Waddell N, Arnold J, Cocciardi S, et al: Subtypes of familial breast tumours revealed by expression and copy number profiling. Breast Cancer Res Treat, 2009 [0203] 23. Jonsson G, Staaf J, Vallon-Christersson J, et al: Genomic subtypes of breast cancer identified by array-comparative genomic hybridization display distinct molecular and clinical characteristics. Breast Cancer Res 12:R42, 2010 [0204] 24. Joosse S A, van Beers E H, Tielen I H, et al: Prediction of BRCA1-association in hereditary non-BRCA1/2 breast carcinomas with array-CGH. Breast Cancer Res Treat 116:479-89, 2009 [0205] 25. Lips E H, Mulder L, Hannemann J, et al: Indicators of homologous recombination deficiency in breast cancer and association with response to neoadjuvant chemotherapy. Ann Oncol 22:870-6, 2011 [0206] 26. Vollebergh M A, Lips E H, Nederlof P M, et al: An aCGH classifier derived from BRCA1-mutated breast cancer and benefit of high-dose platinum-based chemotherapy in HER2-negative breast cancer patients. Ann Oncol 22:1561-70, 2011 [0207] 27. Popova T, Manie E, Stoppa-Lyonnet D, et al: Genome Alteration Print (GAP): a tool to visualize and mine complex cancer genomic profiles obtained by SNP arrays. Genome Biol 10:R128, 2009 [0208] 28. Manie E, Vincent-Salomon A, Lehmann-Che J, et al: High frequency of TP53 mutation in BRCA1 and sporadic basal-like carcinomas but not in BRCA1 luminal breast tumors. Cancer Res 69:663-71, 2009 [0209] 29. Vincent-Salomon A, Gruel N, Lucchesi C, et al: Identification of typical medullary breast carcinoma as a genomic sub-group of basal-like carcinomas, a heterogeneous new molecular entity. Breast Cancer Res 9:R24, 2007 [0210] 30. Marty B, Maire V, Gravier E, et al: Frequent PTEN genomic alterations and activated phosphatidylinositol 3-kinase pathway in basal-like breast cancer cells. Breast Cancer Res 10:R101, 2008 [0211] 31. Azoulay S, Lae M, Freneaux P, et al: KIT is highly expressed in adenoid cystic carcinoma of the breast, a basal-like carcinoma associated with a favorable outcome. Mod Pathol 18:1623-31, 2005 [0212] 32. Nielsen T O, Hsu F D, Jensen K, et al: Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res 10:5367-74, 2004 [0213] 33. Esteller M, Silva J M, Dominguez G, et al: Promoter hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl Cancer Inst 92:564-9, 2000 [0214] 34. Houdayer C, Moncoutier V, Champ J, et al: Enhanced mismatch mutation analysis: simultaneous detection of point mutations and large scale rearrangements by capillary electrophoresis, application to BRCA1 and BRCA2. Methods Mol Biol 653:147-80, 2010 [0215] 35. Dai M, Wang P, Boyd A D, et al: Evolving gene/transcript definitions significantly alter the interpretation of GeneChip data. Nucleic Acids Res 33:e175, 2005 [0216] 36. Sorlie T, Tibshirani R, Parker J, et al: Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci USA 100:8418-23, 2003 [0217] 37. Staaf J, Vallon-Christersson J, Lindgren D, et al: Normalization of Illumina Infinium whole-genome SNP data improves copy number estimates and allelic intensity ratios. BMC Bioinformatics 9:409, 2008 [0218] 38. Waddell N, Arnold J, Cocciardi S, et al: Subtypes of familial breast tumours revealed by expression and copy number profiling. Breast Cancer Res Treat 123:661-77, 2010 [0219] 39. DeRose Y S, Wang G, Lin Y C, et al: Tumor grafts derived from women with breast cancer authentically reflect tumor pathology, growth, metastasis and disease outcomes. Nat Med 17:1514-20, 2011 [0220] 40. Storchova Z, Kuffer C: The consequences of tetraploidy and aneuploidy. J Cell Sci 121:3859-66, 2008 [0221] 41. Young S R, Pilarski R T, Donenberg T, et al: The prevalence of BRCA1 mutations among young women with triple-negative breast cancer. BMC Cancer 9:86, 2009 [0222] 42. Elstrodt F, Hollestelle A, Nagel J H, et al: BRCA1 mutation analysis of 41 human breast cancer cell lines reveals three new deleterious mutants. Cancer Res 66:41-5, 2006 [0223] 43. Xu J, Huo D, Chen Y, et al: CpG island methylation affects accessibility of the proximal BRCA1 promoter to transcription factors. Breast Cancer Res Treat 120:593-601, 2010 [0224] 44. Sjoblom T, Jones S, Wood L D, et al: The consensus coding sequences of human breast and colorectal cancers. Science 314:268-74, 2006 [0225] 45. Garcia Al, Buisson M, Bertrand P, et al: Down-regulation of BRCA1 expression by miR-146a and miR-146b-5p in triple negative sporadic breast cancers. EMBO Mol Med 3:279-90, 2011 [0226] 46. Moskwa P, Buffa F M, Pan Y, et al: miR-182-mediated downregulation of BRCA1 impacts DNA repair and sensitivity to PARP inhibitors. Mol Cell 41:210-20, 2011 [0227] 47. Plo I, Laulier C, Gauthier L, et al: AKT1 inhibits homologous recombination by inducing cytoplasmic retention of BRCA1 and RAD51. Cancer Res 68:9404-12, 2008 [0228] 48. Van Loo P, Nordgard S H, Lingjaerde O C, et al: Allele-specific copy number analysis of tumors. Proc Natl Acad Sci USA 107:16910-5, 2010 [0229] 49. Pujana M A, Han J D, Starita L M, et al: Network modeling links breast cancer susceptibility and centrosome dysfunction. Nat Genet 39:1338-49, 2007 [0230] 50. Xu X, Weaver Z, Linke S P, et al: Centrosome amplification and a defective G2-M cell cycle checkpoint induce genetic instability in BRCA1 exon 11 isoform-deficient cells. Mol Cell 3:389-95, 1999 [0231] 51. Brodie K M, Henderson B R: Characterization of BRCA1 centrosome targeting, dynamics and function: A role for the nuclear export signal, CRM1 and Aurora A kinase. J Biol Chem, 2012 [0232] 52. Kais Z, Parvin J D: Regulation of centrosomes by the BRCA1-dependent ubiquitin ligase. Cancer Biol Ther 7:1540-3, 2008 [0233] 53. Moller P, Hagen Al, Apold J, et al: Genetic epidemiology of BRCA mutations—family history detects less than 50% of the mutation carriers. Eur J Cancer 43:1713-7, 2007 [0234] 54. O'Shaughnessy J, Telli M, Swain S, et al: Phase 3 Study of Iniparib (I) Plus Gemcitabine (G) and Carboplatin (C) in Metastatic Triple-negative Breast Cancer (mTNBC)—Results of an Exploratory Analysis by Prior Therapy European Journal of Cancer 47:S338, 2011 [0235] 55. Miki Y, Swensen J, Shattuck-Eidens D, Futreal P A, Harshman K, Tavtigian S, Liu Q, Cochran C, Bennett L M, Ding W and et al. (1994). A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science, 266, 66-71. [0236] 56. Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, Collins N, Gregory S, Gumbs C and Micklem G. (1995). Identification of the breast cancer susceptibility gene BRCA2. Nature, 378, 789-92. [0237] 57. Esteller M, Silva J M, Dominguez G, Bonilla F, Matias-Guiu X, Lerma E, Bussaglia E, Prat J, Harkes I C, Repasky E A, Gabrielson E, Schutte M, Baylin S B and Herman J G. (2000). Promoter hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl Cancer Inst, 92, 564-9. [0238] 58. Stephens P J, McBride D J, Lin M L, Varela I, Pleasance E D, Simpson J T, Stebbings L A, Leroy C, Edkins S, Mudie L J, Greenman C D, Jia M, Latimer C, Teague J W, Lau K W, Burton J, Quail M A, Swerdlow H, Churcher C, Natrajan R, Sieuwerts A M, Martens J W, Silver D P, Langerod A, Russnes H E, Foekens J A, Reis-Filho J S, van 't Veer L, Richardson A L, Borresen-Dale A L, Campbell P J, Futreal P A and Stratton M R. (2009). Complex landscapes of somatic rearrangement in human breast cancer genomes. Nature, 462, 1005-10. [0239] 59. Birkbak, N.J., Wang, Z. C., Kim, J. Y., Eklund, A. C., Li, Q., Tian, R., Bowman-Colin, C., Li, Y., Greene-Colozzi, A., Iglehart, J. D., et al. (2012). Telomeric allelic imbalance indicates defective DNA repair and sensitivity to DNA-damaging agents. Cancer Discovery 2, 366-375. [0240] 60. Lee, M.-N., Tseng, R.-C., Hsu, H.-S., Chen, J.-Y., Tzao, C., Ho, W. L., and Wang, Y.-C. (2007). Epigenetic inactivation of the chromosomal stability control genes BRCA1, BRCA2, and XRCC5 in non-small cell lung cancer. Clin. Cancer Res. 13, 832-838.