AU2006246241A1 - Gene-based algorithmic cancer prognosis - Google Patents

Description

WO 2006/119593 PCT/BE2006/000051 GENE-BASED ALGORITHMIC CANCER PROGNOSIS 5 Field of the Invention [0001] The present invention is related to new method and tools for improving cancer prognosis. Background of the Invention [0002] Microarray profiling, or the assessment of 10 the mRNA expression levels of hundreds and thousands of genes, has shown that cancer can be divided into distinct molecular subgroups by the expression levels of certain genes. These subgroups seem to have distinct clinical outcomes and also may respond differently to different 15 therapeutic agents used in cancer treatment. But the current understanding of the underlying biology does not permit "individualization" of a particular cancer patients' care. As a result for breast cancer, for example, many women today are given systemic treatments such as 20 chemotherapy or endocrine therapy in an attempt to reduce her risk of the breast cancer recurring after initial diagnosis. Unfortunately, this systemic treatment only benefits a minority of women who will relapse, hence exposing many women to unnecessary and potentially toxic 25 treatment. New prognostic tools developed using microarray technology show potential in allowing us to facilitate tailored treatment of breast cancer patients (Paik et al, New England Journal of Medicine 351:27(2004); Van de Vijver et al, New England Journal of Medicine 347:199(2002); Wang 30 et al, Lancet 365: 671 (2005)). These genomic tools may be WO 2006/119593 PCT/BE2006/000051 2 a much needed improvement over currently used clinical methods. [0003] Histological grading of breast carcinomas has long been recognised to provide significant clinical 5 prognostic information (1) . However, despite recommendations by the College of American Pathologists (2) for use of tumor grade as a prognostic factor in breast cancer, the latest Breast Task Force serving the American Joint Committee on Cancer (AJCC) did not include it in its 10 staging criteria, citing insurmountable inconsistencies between institutions and lack of data (3) . This may be in part related to inter-observer variability and the various grading approaches used, resulting in poor reproducibility across institutions. With the advent of standardized 15 methods such as those developed by Elston and Ellis (1), concordance between institutions has been improved. Nevertheless, whilst grade 1 (low risk) and 3 (high risk) are clearly associated with different prognoses, tumors classified as intermediate grade present a difficulty in 20 clinical decision making for treatment because their survival profile is not different from that of the total (non-graded) population and their proportion is large (40% 50%) . A more accurate grading system would allow for better prognostication and improved selection of women for further 25 breast cancer treatment. [0004] The majority of breast cancers diagnosed today are hormone responsive. Tamoxifen is the most common anti-estrogen agent prescribed today in the adjuvant treatment of these patients. Yet up to 40% of these 30 patients will relapse when given tamoxifen in this setting. At present, due to the positive results of several large trials evaluating the use of aromatase inhibitors instead of,- or in combination or sequence with tamoxifen in the adjuvant setting, there are many options available for post WO 2006/119593 PCT/BE2006/000051 3 menopausal women with hormone responsive breast cancer. Furthermore, it is unclear which treatment option is the best especially given that the long term health costs of aromatase inhibitor use are unknown. The ability to 5 identify a group at high risk of relapse when given tamoxifen could aid in identifying patients for whom tamoxifen is probably not the best option. These patients could then be specifically targeted for alternative treatment strategies. 10 [0005] Particularly relating to the issue of predicting relapse for women treated with adjuvant tamoxifen, two publications have been reported claiming gene sets that can predict clinical outcome (Ma et al, Cancer Cell 5:607(2004), Jansen et al. Journal of Clinical 15 Oncology 23:732(2005). These studies involved small numbers of patients and hence are not thoroughly validated to be widely used clinically. [0006] Accordingly need exists for methods and systems that can accurately assess prognosis and hence help 20 oncologists tailor their treatment decisions for the individual cancer patient. In particular, a need exists for methods and systems directed to breast cancer patients. Aims of the Invention [0007] The present invention aims to provide new 25 methods and tools for improving cancer prognosis that do not present the drawbacks of the methods of the state of the art. Summary of the Invention [0008] One embodiment of the invention provides a 30 method, comprising the steps of (a) measuring gene expression in a tumor sample submitted to an analysis and obtained from a mammal subject, WO 2006/119593 PCT/BE2006/000051 4 preferably a human patient; (b) calculating the gene-expression grade index (or genomic grade) (GGI) of the tumor sample using the formula: Z j x 1 5 wherein: x is the gene expression level of mRNA, G1 and G 3 are sets of genes up-regulated in histological grade 1 (HG1) and histological grade 3 (HG3), respectively, and j refers to a probe or probe set. [0009] The tumor sample may be from tissue afflicted 10 by a cancer selected from the group consisting of breast cancer, colon cancer, lung cancer, prostate cancer, hepatocellular cancer, gastric cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, renal 15 cancer, carcinoma, melanoma, or brain cancer. Preferably, the tumor sample is a histological grade HG2 breast tumor sample. [0010] This embodiment may further comprise designating the tumor sample as low risk (GG1) or high risk 20 (GG3) based on the gene expression grade index (GGI) . This embodiment may further comprise providing a breast cancer treatment regimen for a patient consistent with the low risk or high risk designation of the breast tumor sample submitted to the analysis. 25 [0011] The gene expression grade index GGI may include cutoff and scale values chosen so that the mean GGI of the HG1 cases is about -1 and the mean GGI of the HG3 cases is about +1. The cutoff value is required for calibration of the data obtained from different platforms 30 applying different scales: WO 2006/119593 PCT/BE2006/000051 5 GGI =scale[ Zx -Z x,-cutoff jEG3 jEG1 [0012] The Gi gene set may comprise at least one gene selected from the genes in Table 3 designated as "Up regulated in grade 1 tumors". Preferably, the G 1 gene set 5 comprises at least 4 of those genes, and may include the entire set. The G 3 gene set may comprise at least one gene selected from the genes in Table 3 designated as "Up regulated in grade 3 tumors." Preferably, the G3 gene set comprises at least 4 those genes, and may include the 10 entire set. [0013] In another aspect of the invention, the method according to the invention comprises the steps of (a) measuring gene expression in a tumor sample; (b) calculating a relapse score (RS) for the tumor sample 15 using the formula: G Pi "j /sG jsF}. N wherein: G is a gene set that is associated with distant recurrence of cancer, Pi is the probe or probe set, i identifies the specific cluster or group of genes, wi is 20 the weight of the cluster i, j is the specific probe set value, xij is the intensity of the probe set j in cluster i, and ni is the number of probe sets in cluster i. [0014] This embodiment may further comprises the step of classifying the said tumor sample based on the relapse 25 score as low risk or high risk for cancer relapse. The cutoff for distinguishing low risk from high risk may be a relapse score (RS) of from -100 to +100 or a relapse score (RS) of from -10 to +10. The relapse may be relapse after treatment with. tamoxifen or other chemotherapy, endocrine 30 therapy, antibody therapy or any other treatment method WO 2006/119593 PCT/BE2006/000051 6 used by the person skilled in the art. Preferably, the relapse is after treatment with tamoxifen. [0015] The tumor sample may be from tissue afflicted by a cancer selected from the group consisting of breast 5 cancer, colon cancer, lung cancer, prostate cancer, hepatocellular cancer, gastric cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, renal cancer, carcinoma, melanoma, or brain cancer. Preferably, 10 the tumor sample is a breast tumor sample. [0016] The patient's treatment regimen may be adjusted based on the tumor sample's cancer relapse risk status. For example (a) if the patient is classified as low risk, treating the low risk patient sequentially with 15 tamoxifen and sequential aromatase inhibitors (AIs), or (b) if the patient is classified as high risk, treating the high risk patient with an alternative endocrine treatment other than tamoxifen. For a patient classified as high risk, the patient's treatment regimen may be adjusted 20 to chemotherapy treatment or specific molecularly targeted anti-cancer therapies. [0017] The gene set may be generated from an estrogen receptor (or another marker specific of the cancer tissue sample) positive population. The gene set may be 25 generated by a variety of methods and the component genes may vary depending on the patient population and the specific disorder. [0018] Another embodiment of the invention provides a computerized system or diagnostic device (or kit), 30 comprising: (a) a bioassay module, preferably a bioarray, configured for detecting gene expression for a tumor sample based on a gene set; and (b) a processor module configured to calculate GGI or RS of the tumor sample based on the gene expression and to generate a risk assessment for the WO 2006/119593 PCT/BE2006/000051 7 breast tumor sample. The bioassay module may include at least one gene chip (microarray) comprising the gene set. The gene set may include at least one gene, preferably at least 4 genes, selected from the genes in Table 3 5 designated as "Up-regulated in grade 1 tumors" or may include the entire set. The gene may include at least 4 genes selected from the genes in Table 3 designated as "Up regulated in grade 3 tumors" or may include the entire set. Brief Description of the Drawings 10 [0019] Figure 1 is representing heatmaps showing the pattern of gene expression in the training (panel a) and the validation sets (panel b). The horizontal axis corresponds to the tumors sorted first by HG and then by GGI as the secondary criterion. The vertical axis 15 corresponds to the genes. The GGI values of each tumor and the relapse free survival are indicated underneath. Two groups of genes are found: those that are highly expressed in grade 1 (16 probe sets; highlighted in red) and, reciprocally, those highly expressed in grade 3 (112 probe 20 sets) . The GGI values for HG2 tumors cover the range of values for HG1 and HG3, and those with high GGI tend to relapse earlier (red dots). [0020] Figure 2 shows Kaplan-Meier RFS analysis based on the HG (panel a) and the GG (panel b) for data 25 pooled from the validation datasets 2-5 (table 11) . HG1, HG2 and 1G3 can be split further into low and high risk subsets by GG, indicating that GG is an improvement over HG (panel c, d and e respectively). ER status identifies some, but not all, of the patients with poor prognosis (panel f). 30 [0021] Figure 3 shows Kaplan-Meier RFS analysis based on the NPI (a) and the NPI-GG (b) classification. NPI-GG improves the prognostic discrimination in both low (panel c) and high (panel d) risk NPI subsets, but not vice WO 2006/119593 PCT/BE2006/000051 8 versa (panels e and f) . The Sorlie et al. dataset was excluded from this analysis because of incomplete tumor size information. [0022] Figure 4 shows a Forest plot for hazard 5 ratios for HG2 patients split into GG1 and GG3, showing consistent results in different datasets Hazard ratios were estimated with Cox proportional hazard regressions, horizontal lines are 95% confidence intervals for the hazard ratio. P values were determined by the log rank 10 test. [0023] Figure 5 shows distant metastasis free survival (DMFS) analysis based on the 70-gene expression signature (left row, panels a, c and e) and on GGI (right row, panels b, d and f) for data from the Van de Vijver et 15 al. validation study. a) and b) are all patients, c) and d) are node-negative, and e) and f) are node-positive patients. Note that the node-negative subset includes patients used to derive the 70-gene signature. [0024] Figure 6 represents a genomic grade applied 20 to previously reported molecular subtypes. [0025] Figure 7 represents Kaplan Meyer survival curves for distant metastasis free survival for GGI (high vs. low). Detailed Description of the Invention 25 Definitions [0026] Most terms scientific, medical and technical terms are commonly understood to one skilled in the art. [0027] The term "microarray" refers to an ordered arrangement of hybridizable array elements, preferably 30 polynucleotide probes, on a substrate (an insoluble solid support). [0028] The terms "differentially expressed gene", "differential gene expression" and their synonyms, which WO 2006/119593 PCT/BE2006/000051 9 are used interchangeably, refer to a gene whose expression is activated to a higher or lower level in a subject suffering from a disease, specifically cancer, such as breast cancer, relative to its expression in a normal or 5 control subject. The terms also include genes whose expression is activated to a higher or lower level at different stages of the same disease. It is also understood that a differentially expressed gene may be either activated or inhibited at the nucleic acid level or protein 10 level, or may be subject to alternative splicing to result in a different polypeptide product. Such differences may be evidenced by a change in mRNA levels, surface expression, secretion or other partitioning of a polypeptide, for example. Differential gene expression may include a 15 comparison of expression between two or more genes or their gene products, or a comparison of the ratios of the expression between two or more genes or their gene products, or even a comparison of two differently processed products of the same gene, which differ between normal 20 subjects and subjects suffering from a disease, specifically cancer, or between various stages of the same disease. Differential expression includes both quantitative, as well as qualitative, differences in the temporal or cellular expression pattern in a gene or its 25 expression products among, for example, normal and diseased cells, or among cells which have undergone different disease events or disease stages. For the purpose of this invention, "differential gene expression" is considered to be present when there is at least an about two-fold, 30 preferably at least about four-fold, more preferably at least about six-fold, most preferably at least about ten fold difference between the expression of a given gene in normal and diseased subjects, or in various stages of disease development in a diseased subject.

WO 2006/119593 PCT/BE2006/000051 10 [0029] Gene expression profiling: includes all methods of quantification of mRNA and/or protein levels in a biological sample. The term "prognosis" is used herein to refer to the 5 prediction of the likelihood of cancer-attributable death or progression, including recurrence, metastatic spread, and drug resistance, of a neoplastic disease, such as breast cancer. [0030 ] The term "prediction" is used herein to refer 10 to the likelihood that a patient will respond either favorably or unfavorably to a drug or set of drugs, and also the extent of those responses, or that a patient will survive, following surgical removal or the primary tumor and/or chemotherapy for a certain period of time without 15 cancer recurrence. The predictive methods of the present invention are valuable tools in predicting if a patient is likely to respond favorably to a treatment regimen, such as, chemotherapy with a given drug or drug combination, and/or radiation therapy, or whether long-term survival of 20 the patient, following surgery and/or termination of chemotherapy or other treatment modalities is likely. [0031] The term "high risk" means the patient is expected to have a distant relapse in less than 5 years, preferably in less than 3 years. 25 [0032] The term "low risk" means the patient is expected to have a distant relapse after 5 years, preferably in less than 3 years. [0033] The term "tumor," as used herein, refers to all neoplastic cell growth and proliferation, whether 30 malignant or benign, and all pre-cancerous and cancerous cells and tissues. [0034] The terms "cancer" and "cancerous" refer to or describe the physiological condition in mammals that is typically characterized by unregulated cell growth.

WO 2006/119593 PCT/BE2006/000051 11 Examples of cancer include but are not limited to, breast cancer, colon cancer, lung cancer, prostate cancer, hepatocellular cancer, gastric cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder 5 cancer, cancer of the urinary tract, thyroid cancer, renal cancer, carcinoma, melanoma, and brain cancer. [0035] Raw "GGI" (Gene expression grade index) is the sum of the log expression (or log ratio) of all genes high-in-HG3 - sum of the log expression (or log ratio) of 10 all genes high-in-HG1 and can be written as: jEG 3 jeG 1 wherein: x is the gene expression level of mRNA, [0036] Gi and G 3 are sets of genes up-regulated in 15 HG1 and HG3, respectively, and j refers to a probe or probe set. [0037] GGI may include cutoff and scale values chosen so that the mean GGI of the HG1 cases is about -1 and the mean GGI of the HG3 cases is about +1: GGI =scale[ Zx, - Xx/-cutoff ] 20 jEG3 jEOS The cutoff in GGI is 0 and corresponds to the mean of means. GGI ranges in value from -4 to +4. 25 Example 1 Material and Methods for development of grade index (GGI) Patient demographics [0038] Six datasets of primary breast cancer were used, four of which were publicly available (Table 11) (4, 30 5, 10, 11). No patient received adjuvant chemotherapy and WO 2006/119593 PCT/BE2006/000051 12 some had received adjuvant tamoxifen treatment. Histological grade (HG) was based on the Elston-Ellis grading system. Each institutional ethics board approved the use of the tissue material. 5 Table 1: Microarray datasets used in this study Identifier Institution N Microarray Systemic Reference Platform Treatment 1. Training set Karolinska 24 Affymetrix yes this paper (KJX64) U133A (tamoxifen John Radcliffe 40 only) 2. Validation set Karolinska 68 Affymetrix No this paper (KJ129) U133A John Radcliffe 61 3. Sotiriou et al. John Radcliffe 99 cDNA Yes 10 (NCI) (NCI) 4. Sorlie et al. Stanford 80 cDNA Yes 11 (STNO) (Stanford) 5. van't Veer et al. Netherlands 97 Agilent No 4 (NKI) Cancer Institute 6. Van de Vijver et Netherlands 295 Agilent No 5 al. Cancer Institute [61 (NKI2) also in 5)] Total 703 WO 2006/119593 PCT/BE2006/000051 13 [0039] The samples from Oxford were processed at the Jules Bordet Institute in Brussels, Belgium, and those from Sweden at the Genome Institute of Singapore in Singapore. RNA extraction, amplification, hybridization and scanning 5 were done according to standard Affymetrix protocols. Affymetrix U133A Genechips (Affymetrix, Santa Clara, CA). Gene expression values from the CEL files were normalized using RMA (12). [0040] The default options (with background 10 correction and quantile normalization) were used. The output were in logarithmic scale. [0041] The normalizations were done separately for .CEL files from different institutions and batch of measurements. In subsequent analysis, the expression data 15 matrices were treated as if they were "blocks" of separate studies. The training set KJX64 consisted of two blocks (corresponding to two different institutions), and so did the validation set KJ129. [0042] STNO The Stanford/Norway dataset (Sorlie et 20 al., 2001) was downloaded from http: //genome www.stanford.edu/breast.cancer/mopo.clinical/data.shtml [0043] It consists of 85 arrays, with several different chip designs. Only the probes that are common to 25 all were used. The gene expression values used are from the column LOG RAT2N MEAN in the array data files. No further transformation is applied prior to computing the GGI. When more than one spot corresponds* to a probe, their average was used. 30 [0044] All 85 patients were used in the heatmap, but only those with non missing and non zero follow up time were used in survival analysis. This dataset was excluded from analysis involving tumor size, since this information WO 2006/119593 PCT/BE2006/000051 14 was not available (Only TNM category was given, but the conversion to tumor size is not straightforward, particularly when one is concerned with what is appropriate for the NPI formula). 5 [0045] NKI/NKI2 The data set NKI (van't Veer et al., 2002) and NKI2 (van de Vijver et al., 2002) were downloaded from Rosetta website www.rii.com. The log ratio was used without further transformation. For NKI2, flagged expression values were considered missing. Age, tumor size, and 10 histological grade were not available for NKI2. [0046] The field 'conservFlag' in the clinical data table were used to stratify the dataset into two groups. Each group had its own threshold for deciding 'good' vs 'poor' prognosis, as was done for in the original results 15 in van de Vijver et al. (2002). [0047] NCI This dataset from Sotiriou et al. (2003) was downloaded from the PNAS web site http://www.pnas.org/cgi/content/full/1732912100/DC1. The expression values were not modified. 20 Statistical analysis [0048] Gene selection was done only on the KJX64 dataset, which are all estrogen receptor (ER) -positive and either HG1 or HG3. Dataset KJ129 (43 ER-negative, all node negative, no systemic treatment) was used as the validation 25 set, along with other previously published data (see table 11) . ER-positive tumor s were used for the training set, because ER-status and grade were not independent, with very few ER-negative, HG1 tumor s. Using all HG1 and HG3 tumor s regardless of the ER status would have resulted in spurious 30 associations. [0049] The standardized mean difference of Hedges and Olkin (13), was used to rank genes based on their differential expression with respect to HG1 or HG3. This WO 2006/119593 PCT/BE2006/000051 15 meta-analytical score is similar to the t-statistic, but better suited for our training set which consisted of array data originating from two different centres. [0050] To control for multiple testing, the maxT 5 algorithm of Westfall and Young (14), with an extension proposed by Korn et al. (15), was applied to compute false discovery counts (FDC). All 22,283 probe sets were considered. Probe sets having a family-wise error rate p value lower than 0.05 with FDC>2 were identified. Mapping 10 of probes between platforms was done through Unigene (build #180), according to the method in Praz et al. (16). [0051] The gene-expression grade index (GGI) is defined as: GGI = scale[ V x -V x-cutoff ] jeC 3 fec' 15 where x is the logarithmic gene expression measure, and G, and G3 are the sets of genes up-regulated in HG3 and HG1, respectively. These sets differed across platforms. For convenience, the cutoff and the scale were chosen so that the mean GGI of the HG1 cases was -1 and that of the HG3 20 cases was +1. This rescaling was done separately for each data source. [0052] The Nottingham Prognostic Index (NPI) was calculated according to Todd et al. (17): NPI= 0.2 x size [cm] + lymph node status + histological 25 grade. [0053] An index called NPI/GG was defined, where HG was replaced by GG. Cases with NPI 3.4 to be high risk in both NPI and NPI/GG were considered. Survival data were visualized using Kaplan-Meier plot. The hazard ratios (HR) 30 were estimated using Cox regression, stratified by the data source. Assumption-free comparisons were done using the stratified log rank test.

WO 2006/119593 PCT/BE2006/000051 16 Heat maps [0054] For visualization, the values used in the heatmaps for each probe were meancentered across patients. No genespecific scaling (standardization) was done, in order 5 to keep the information about the relative signal strength of all probes. The color tone were calibrated such that saturated red and green were reached at the values three times the standard deviation of the expression values of the entire matrix. Note that the scaled GGI values were not 10 affected by genespecific centering. Survival Analysis [0055] The survival package for R was used by Terry Therneau and a custom program for the KaplanMeier plots, which was checked against the output of the survival 15 package for correctness. Mapping across microarray platforms [0056] The approach of CleanEx database (http://www.cleanex.isbsib.ch), described in Praz et al. (2004) was used. Probe identifies were first mapped into 20 sequence accession number. Unigene (build 180) were then used to map the correspondence between platforms. For Affymetrix chips, probesets which contain oligos that were ambiguously mapped to more than one Unigene id were excluded. 25 Results Differentially expressed genes between high and low grade subsets [0057] 242 Probe sets corresponding to 183 unique genes with FDC>2 at family-wise error rate p-value of 0.05, 30 corresponding to a low false discovery proportion of 0.008 were identified (Table 3). Of these, a list of 128 probe WO 2006/119593 PCT/BE2006/000051 17 sets (97 genes) based 'on a more conservative criterion (FDC>0 at p-value of 0.05) was used in all subsequent analyses, except for checking common genes with signatures published by others, where we used the 183-gene list. 5 [0058] Figure la shows two strong and reciprocal patterns of expression clearly associated with HG1 and HG3. Many genes up-regulated in HG3 were mostly associated with cell cycle progression and proliferation (Table 3). The same gene selection algorithm to contrast HG2 tumor s with 10 a pool combining HG1 and HG3 tumor s were applied. This yielded no differentially expressed genes. Thus, the HG2 population as a whole has no peculiar characteristics of its own that are independent from the HG1 and HG3 distinction. 15 [0059] The list of 128 probe sets was then applied to untreated breast cancer patients (dataset KJ129) . As shown in Fig. 1b, visual inspection revealed an expression pattern for HG1 and HG3 similar to that which was observed on the training set (Fig.la). The GEP of the grade 2 20 population looked like a mixture of grade 1 and grade 3 cases, rather than intermediate between the two. To make this observation more objective, the GGI (which essentially summarizes the differences in the GEP of the reporting genes by averaging their expression levels) was defined. As 25 shown under the heat maps in Fig. 1, the GGI distribution of HG2 covered the range of the GGI values of HG1 and HG3, confirming the visual impression. A similar observation was made on the three previously published datasets, despite differences in the clinical populations and microarray 30 platforms (see figures Ga, b, and c).

WO 2006/119593 PCT/BE2006/000051 18 Histological grade, gene-expression grade (GG) and prognosis [00603 These findings lead to showing that intermediate histological grade can be replaced by low and 5 high grade based on gene expression. Gene-expression grade (GG) based on the GGI score was defined. Patients were classified as GG1 (low grade) if their GGI value was negative or as GG3 (high grade) otherwise. Note that the GGI score of zero corresponds to the midpoint between the 10 average GGI values of HG1 and HG3 (see methods). This choice might not be clinically optimal and could be improved based on the trade-off between the cost of treatment and risk, but it would be sufficient for evaluating the prognostic value of GGI. 15 [00613 For this purpose, breast cancer samples derived from a pool of our own validation population (KJ129) and additional datasets STNO, NCI and NKI (table 11) were used. In figure 2a, the association between histological grade and relapse-free survival (RFS) was 20 examined. As expected, HG3 tumor s had significantly worse RFS than HG1, while HG2 tumor s had an intermediate risk and constituted 38% of the population. In figure 2b, GG1 and GG3 subgroups showed distinct RFS, similar to the RFS of HG1 and HG3 tumor s, respectively. To examine how the 25 discordance between GG and HG are related to prognosis, GG was split for each of the histological categories (figure 2c, 2d and 2e) . The most striking result was that GG split HG2 into two groups, namely HG2/GG1 and HG2/GG3, whose RFS were also respectively similar to those of HG1 and HG3 30 (Fig. 2d). The log rank test failed to reveal any significant difference in survival between HG1 and HG2/GG1, as well as between HG3 and HG2/GG3 (see figure 7). For comparison, ER status also had prognostic power in HG2 tumor s (Fig, 2f), although the hazard ratio was less than WO 2006/119593 PCT/BE2006/000051 19 that of GG (Fig. 2d). Notably, the ER-positive group showed similar RFS as the total population. [0062] While GG was better than HG by classifying some patients with poor prognosis in the HG1 population 5 (fig. 2c), the reverse seems to be the case in HG3 population: it classified some patients as low-risk despite their poor prognosis (fig. 2d). Thus, in the case of discordance involving low and high grade categories, neither GG nor HG were consistently outperform the other. 10 It seemed that whichever decided to classify as high grade tended to be more accurate prognostically. This suggests that for both HG and GG, correctly detecting any indication of high grade was easier than accurately declaring it absent. If this observation is confirmed by future studies, 15 corrections should be done in clinical practice, for example by using a rule which substitutes HG1 and HG2, but not HG3, by GG. However, the frequency of this type of discordance in the data used here was relatively small and such modifications were not used in this study, which aims 20 to characterize GG purely on its own.

WO 2006/119593 PCT/BE2006/000051 20 Table 12: Multivariate analysis of breast cancer prognostic factors (N = 302) Univariate analysis Multivariate analysis Hazard ratio p Hazard ratio p (95%CI) (95%CI) Gene-Expression Grade GG3 vs GG1 2.97 (2.03 - 4.37) 0.0001 2.29 (1.44 - 3.63) 0.0004 Histological Grade 2 + 3 vs 1 1.93 (1.15 - 3.28) 0.0150 0.85 (0.46- 1.57) 0.61 3 vs 1 + 2 2.03 (1.41 -2.92) 0.0001 1.25 (0.80 - 1.94) 0.33 Estrogen Receptor Negative vs Positive 1.76 (1.24 -2.49) 0.0016 1.19 (0.81 - 1.76) 0.38 Nodal Status Positive vs Negative 2.53 (1.34 - 4.78) 0.0040 1.95 (1.01 - 3.73) 0.045 Tumor Size > 2cm vs 2cm 2.06 (1.41 - 3.03) 0.0002 1.63 (1.10 - 2.43) 0.015 Age (years) :60 vs >50 0.99 (0.69 - 1.42) 0.97 1.13 (0.78 - 1.63) 0.53 WO 2006/119593 PCT/BE2006/000051 21 Prognostic value of GG in multivariate model [0063] Almost all clinicopathological variables were significantly associated with clinical outcome in univariate analysis (Table 12). GG and HG status had the 5 strongest effect. However, in multivariate analysis, only GG, nodal status and tumor size kept their significance, with GG having the largest hazard ratio. In accordance with figure 2, GG replaced HG when both were considered, and GG considerably reduced the prognostic impact of ER. 10 GG and the Nottingham Prognostic Index [0064] The independence of GG, nodal status and tumor size in explaining the disease outcome mirrored the Nottingham Prognostic Index (NPI), which combines HG, nodal status and size. -To test whether GG can be used to improve 15 this well-characterized risk score, we propose a score called NPI/GG, which is analogous to NPI except that HG is replaced by GG, with only two possible values (either 1 or 3). As shown in Fig. 3a and 3b, NPI/GG was significantly more discriminative than classical NPI. Moreover, NPI/GG 20 was able to split both the NPI low and high risk groups into subgroups with significantly different clinical outcome (Fig. 3c, 3d), while the reverse was not true (Fig. 3e, 3f). Example 2 25 Consistent prognostic value of GG in different populations and microarray platforms [0065] The results of the pooled analysis above were consistently present in the individual datasets, as shown by the forest plot of hazard ratios in Figure 4. More 30 complete results are shown in figure 8. Figure 4 shows that in each independent validation dataset, GG divided the grade 2 populations into two distinct groups with WO 2006/119593 PCT/BE2006/000051 22 statistically different clinical outcomes. There was no significant heterogeneity between the hazard ratios, even though the different datasets included heterogeneous patient populations, were graded by various pathologists 5 and used different microarray platforms. Relationship with the 70-gene signature [00661 In their pioneering work, van't Veer et al. identified a 70-gene expression signature significantly correlated with distant metastasis in node negative breast 10 cancer patients (5) . The present list of 97 genes (128 probe sets) could be mapped to 93 genes (113 probes) in their Agilent arrays. To allow comparison under the same trade-off between risk and the cost of treatment as the Netherlands Cancer Institute (NKI) classification, cutoffs 15 for GGI that gave the same numbers of patients in high- and low-risk groups were selected (see methods) . Figure 5 shows the comparisons between the NKI prognostic signature and the GGI on distant-metastases-free survival for the overall population (fig.5a, b), as well as for the node negative 20 (fig.5c, d) and positive subgroups (fig.5e, f) . Despite the fact that our probes were selected without using clinical outcome and had to be mapped across platforms, the results were strikingly close. Similar results were found when considering overall survival (see figures 9) . Data were 25 unavailable to compare relapse-free survival. [0067] Low and high grade breast cancers were unexpectedly associated with many differentially expressed genes, the majority being involved in cell cycle and proliferation. For these genes, HG2 tumor s had 30 heterogeneous transcriptional profiles that covered the range of variation of HG1 and HG3 tumor s. A similar observation was made in at least one previous report (18). Here, the clinical implications of this finding and WO 2006/119593 PCT/BE2006/000051 23 discovered that the grade-related GEPs were also correlated with disease outcome are investigated. [0068] As demonstrated by Figure 4 improvements by GG were consistent across the different datasets which 5 would have not been the case if the grading quality differed significantly between these studies. Similarly, figure 2a shows good prognostic separation between HG1 and HG3, indicating that the histological grading was of high quality. Furthermore, central pathologist review would 10 still result in a significant portion of tumor s being classified as HG2. Finally, these results were more reflective of clinical reality, since grading by a central pathologist is rarely done in practice. [0069] The approach in identifying GEP associated 15 with prognosis is quite different from that used by other investigators. Instead of selecting the prognostic genes directly through their correlation with survival, one may identify them indirectly through histological grade, a well-established prognostic factor rooted in cell biology. 20 This may explain the robustness and reproducibility of GGI across independent and heterogeneous validation sets and different microarray platforms. Furthermore, since the GGI can be interpreted as "molecular grade", it can be integrated easily into existing prognostic systems which 25 uses histological grade, such as the NPI. [0070] This gene selection process was not meant to define a specific set of genes to be used as a prognostic "signature". The present invention aims to build a comprehensive "catalogue" where different sets of 30 signatures could be chosen from. This was illustrated by the cross-platform applicability of the catalogue. Although the actual sets of probes used in various platforms differed in numbers and gene compositions, the results were still reproducible. It is remarkable to obtain good WO 2006/119593 PCT/BE2006/000051 24 prognostic discrimination in very different datasets with a linear classifier where the weights of the genes were simply +1 or -1, based on their association with grade on a training set of 64 patients. Thus, the "grade signal" 5 identified was not bound to a particular set of genes nor to any special combination of their expression levels, since the genes were highly correlated and the GGI effectively behaves as a single prognostic factor. It is still beneficial to use many genes, if only to provide 10 redundancy against noise. The consequence for the development of practical diagnostic systems is that arbitrary subsets of the "grade gene catalogue" of the invention might be used, constrained only by technical considerations. 15 [0071] Jenssen and Hovig (19) recently discussed two issues regarding the use of gene-expression signatures for prognosis. These were 1) the lack of agreement between genes included in different signatures and 2) the difficulty in understanding the biological basis of the 20 correlation between the signatures and survival. The present gene catalogue is rich in genes with likely roles in cell cycle progression and proliferation. This class of genes is one important -if not the most important component of any existing profile-based risk prediction 25 method for breast cancer. In Paik et al. (7), the "proliferation set", whose five genes are all in our 183 gene catalogue (Table 3), was the one that had the largest hazard ratios in their extensive training and validation sets and has the highest weight in the "recurrence score" 30 formula. The application to the NKI data in figure 5 also lends support to the idea that grade-related genes may constitute a significant portion of the prognostic power of the NKI 70-gene signature. When compared against our 183 gene catalogue, the following numbers of genes in common WO 2006/119593 PCT/BE2006/000051 25 with other prognostic signatures: 11/70 and 30/231 genes (van't Veer et al.), 5/15 (Paik et al) and 7/76 (Wang et al.)(4, 7, 8) were found. [0072] In summary, gene-expression based grading 5 could significantly improve current grading systems for the prognostic assessment of cancer, in particular breast cancer. [0073] Reproduction of these findings across multiple independent datasets and across different 10 platforms suggests our conclusions are robust. The GGI score does not require a specific set of genes nor is it bound to a particular detection platform. Grading based on the GGI can be incorporated into existing prognostic systems, by substituting HG with GG. Refined grading based 15 on gene expression measurements could have important clinical application for breast cancer management in the future. Example 3 Definition of clinically distinct subtypes within Estrogen 20 receptor positive breast carcinoma Materials and methods Tumor samples [0074] Three hundred and thirty five early-stage breast carcinoma samples comprised our own dataset. Eighty 25 six of these samples have been previously used in another study and the raw data are available at the Gee Expression Omnibus repository database (http://www.ncbi.nlm.nih.gov/geo), with accession code GSE2990. These samples had received no adjuvant systemic 30 therapy. Two hundred and forty-nine samples, previously unpublished, had received adjuvant tamoxifen only (tam treated dataset) . All samples were required to be ER positive by protein ligand binding assay.

WO 2006/119593 PCT/BE2006/000051 26 [0075] Microarray analysis was performed with Affymetrix" Ul13A Genechips* (Affymetrix, Santa Clara, CA). This dataset contained samples from the John Radcliffe Hospital, Oxford, U.K., Guys Hospital, London, U.K. and 5 Uppsala University Hospital, Uppsala, Sweden. Samples from Oxford and London were processed at the Jules Bordet Institute in Brussels, Belgium. For the samples from Uppsala, RNA was extracted at the Karolinska Institute and hybridized at the Genome Institute of Singapore in 10 Singapore. The quality of the RNA obtained from each tumour sample was assessed via the RNA profile generated by the Agilent bioanalyzer. RNA extraction, amplification, hybridization, and scanning were done according to standard Affymetrix protocols. Gene expression values from the CEL 15 were normalized by use of RMA 1. Each population was normalised separately. Each hospital's institutional ethics board approved the use of the tissue material and written informed consent was obtained. The raw data for the tam treated dataset are available at the Gene Expression 20 Omnibus repository database (http://www.ncbi.nlm.nih.gov/geo/), with accession code GSE XXX. [0076] The inventors also used four other publically available datasets, described in recent publications: van 25 de Vijvers (n=295) , Wang 8 (n=286) , Sotiriou 0 (n=99), Sorlie" (n=78), in the analysis. For the survival analysis, we used tumors classified as ER-positive only (van de Vijver 5 (n=122), Wang 8 (n=209)). For the survival analysis involving patients who had received no systemic 30 adjuvant treatment, patients from the van de Vijver et al.5, Wang et al." and previously published dataset were combined (n=417 ER-positive patients, hereby referred to as the "untreated" dataset) WO 2006/119593 PCT/BE2006/000051 27 All clinical data are shown in Table S1 of the Supplementary Information. Data Analysis Estrogen (ER) and Progesterone receptor (PgR) level 5 [0077] Patients were initially selected at their institutions according to a positive ER status which was determined by protein ligand-binding assay. The inventors subsequently confirmed a positive ER level by using the microarray data. The ER level was measured by probe set (a 10 30-mer oligonucleotide) on our human Affymetrix" GeneChip® U133 A&B microarray. The inventors have used the probe set "205225_at" for ER. PgR was represented by the probe set "208305_at". The immunohistochemical measurement of ER is known to correlate with mRNA levels of ER 4. Tumours with 15 any positive expression level of ER and PgR were considered. Histological grade [0078] Histological grade was based on the Elston Ellis grading system. A central pathologist reviewed the 20 histological grade and ER status for all samples from Uppsala, Sweden, Guys Hospital, London, UK and the Van de Vijver et al. dataset 5. An index based on the expression of proliferation-related 25 genes to quantify genomic grade: gene expression grade index (GGI) [0079] "Gene expression grade index" (GGI) is a linear combination of the expression of 128 probe sets (97 genes) that were found to be differentially expressed 30 between histological grade 1 and 3 (see definitions) . The index is effectively, a quantification of the degree of similarity between the tumourexpression profile and tumour WO 2006/119593 PCT/BE2006/000051 28 grade. A high gene-expression grade index corresponds to a high grade and vice versa. This index was used to divide each data set into high and low grade sub-groups. [0080] Mapping of probes between microarray 5 platforms was done through Unigene (build #180), according to the method in Praz et al.16. Hierarchical clustering The "Cluster" program was used to perform 10 average linkage hierarchical cluster analysis 28 after median centering of each gene using an uncentered Pearson correlation as similarity measurement. The cluster results were viewed using "TreeView". Expression data was downloaded and extracted from datasets Sorlie et al." and 15 Sotiriou et al. 10. The samples were ordered according to subtype as in the original publications '0, 11 to investigate the relation between the expression of the genes in the GGI and the subtypes. 20 Statistical Analysis [0081] In order to assess the relation between survival and some continuous variable, a variant of a method introduced to compute the expected survival for individual was used: "Rate of distant recurrence" plots 29 25 (ref: Terry M. Therneau and Patricia M. grambsch, 2000, "Modeling Survival Data: Extending the Cox Model", chapter 10) . The expected proportion of distant metastasis with respect to the GGI, ER and PgR was plotted using a Cox model fitted with only the variable under study. 30 [0082] Survival curves were visualized using Kaplan Meier plots and compared using log-rank tests. The univariate and multivariate hazard ratios (HR) were estimated using Cox regression analysis. All statistical WO 2006/119593 PCT/BE2006/000051 29 tests were two-sided. Statistical analysis was performed using SPSS statistical software package, version 11.5. RESULTS 5 Applying genomic grade to the previously reported molecular subtypes [0083] To investigate the expression of the gene expression grade index (GGI) in -relation to the subtypes, expression data were extracted from data sets Sorlie and 10 Sotiriou et al., the original and confirmatory publications respectively 11, 13 The genes were clustered using average linkage clustering and the samples were ordered according to the subtypes as presented in the published manuscripts . Applying genomic grade to the previously reported 15 molecular subtypes (6a: Sorlie et al.; 6b: Sotiriou et al.) Subtypes are ordered the same as in the original publications. The heatmap of GGI genes is placed below the dendrogram. Boxplots of the GGI score (median and range) are placed below each subtype. High grade is indicated by a 20 GGI score >1 and vice versa. [0084] Figure 6 shows the results of this analysis. In general, the ER-negative subtypes, the basal and the erbB2 subtypes, had high expression of GGI, or were of high grade. However, the ER-positive subtypes showed a diverse 25 range of GGI levels, particularly the luminal C or 3 subtype both highly expressing these proliferation associated genes, whereas luminal A or 1, and the normal like were mostly negative for the expression of the GGI, or low grade. This confirmed the hypothesis that there are 30 varying degrees of contribution of cell cycle genes to the biological makeup of ER-positive tumours, whereas ER negative tumours seem to consistently have over-expression of these genes. It is interesting to note the similarity in WO 2006/119593 PCT/BE2006/000051 30 expression profiles of the GGI genes between the high grade ER-positive subtype and the ER-negative subtypes. Clinical relevance of ER-positive luminal subtypes as 5 defined by genomic grade [0085] Genomic grade could distinguish clinically subtypes within the ER-positive tumours and the prognostic value of these genomic grade defined subtypes were an improvement over current traditional methods, such as that 10 based on quantitative levels of estrogen and progesterone receptor levels. A Kaplan-Meier survival analysis was performed comparing classes of ER-positive tumours according to GGI score (high vs. low grade) and expression levels of estrogen and progesterone receptor (rich vs. poor 15 expression) with, respect to time to distant metastasis (TDM), which is often used as a surrogate for breast cancer specific survival (Figure 7- KM and Cox) . Kaplan Meier survival curves for distant metastasis free survival for GGI (high vs. low), ER expression levels and PgR expression 20 levels (rich vs. poor). Figure 7a displays the results for the untreated dataset (n=417). Figure 7b for the tamoxifen treated dataset (n=249) . For the untreated dataset, results shown were combined from multiple datasets involving 417 ER-positive samples hybridized using two popular 25 commercially available oligonucleotide microarray platforms- Affymetrix" and Agilent" (see methods) . As shown, for both untreated and tamoxifen-treated populations, the expression levels of the ER did not have any prognostic value (p=0.74 and 0.51 respectively) . In 30 contrast, both the GGI and expression levels of the PgR had prognostic value (untreated: p<0.0001 for both GGI and PgR; tam-treated: GGI p<0.0001, PgR p=0.0058) . The luminal low grade subtype had a much better 10-year estimate of TDM compared with the luminal high grade subtype.

WO 2006/119593 PCT/BE2006/000051 31 [0086] Table 13 shows the univariate and multivariate analysis with other standard prognostic covariates of age, grade, tumour size as well as genomic grade. In the multivariate Cox regression analysis, only 5 the GGI retained significant prognostic value (untreated: HR 2.3 (95%CI: 1.2-4.3; p=0.008; tam-treated: HR 2.14 (95%CI: 1.04-4.02; p=0.0038), subsuming those factors that were significant at the univariate level, including the progesterone receptor expression levels (p=0.3) . For the 10 untreated population, tumour size also retained significance in the multivariate model (HR 2.2 (95% CI:1.2 3.8, p=0.0068). This suggests that genomic grade, as measured by the GGI, can distinguish clinically distinct groups of patients within those that express positive 15 levels of estrogen receptor. Furthermore, the GGI had highly significant prognostic value, suggesting a better ability to discriminate clinical outcome over these traditional factors. The ER-positive high grade subgroup's worse disease outcome in the tamoxifen-treated dataset 20 seems to suggest that adjuvant tamoxifen does not alter this subtype's natural disease history despite having a positive ER status. This could potentially flag a group of tumours worthy of further investigation from both a biological and therapeutic standpoint. 25 [0087] As further demonstration of the GGI's prognostic value in ER-positive tumours, the inventors generated figures displaying the rate of distant recurrence as continuous function of the GGI and compared this to continuous levels of ER and PgR for both untreated and tam 30 treated populations. [00883 Two subtypes of tumours can be distinguished within patients whose breast cancers express at least some level of estrogen receptor. In patients whose tumours express a high level of the genes that comprise the GGI, WO 2006/119593 PCT/BE2006/000051 32 i.e. corresponding to high genomic grade, their disease outcome was clearly different, with a higher incidence of relapses compared with tumours of low genomic grade. Furthermore, their worse disease outcome seemed unchanged 5 even when given adjuvant tamoxifen, suggesting that this group of women do not seem to benefit from adjuvant tamoxifen despite their positive estrogen receptor values. Note that none of the patients in this study had received adjuvant chemotherapy, so it is unclear if chemotherapy can 10 alter this group's natural disease history. The potential clinical significance of this finding is also underscored by the similarities between the high grade ER-positive group and the high grade ER-negative tumours (basal and erbB2), further suggesting that high levels of expression 15 of the genes associated with high genomic grade is associated with a poor prognosis. The GGI can consistently identify these two groups across multiple datasets which were hybridized using several microarray platforms, involving 666 ER-positive samples, suggesting our 20 conclusions are robust and highly reproducible than that produced previously by hierarchical cluster analysis 1,3 [0089] The genes present in the GGI are associated with cell cycle progression and proliferation: among the top 20 overexpressed genes were UBE2C, KPNA2, TPX2, FOXM1, 25 STK6, CCNA2, BIRC5, and MYBL2; see Supplemental Table 14) . For ER-positive tumours, genomic grade was associated with differing relapse-free survival, but for ER-negative tumours, as almost all are associated with high genomic grade, the GGI had no prognostic value. Therefore, cell 30 cycle related genes seem to have prognostic value only in breast cancer patients with positive expression of ER. Within this group, the incidence of distant metastases seems to be predominantly driven by this set of proliferation and grade-derived genes. However, in ER- WO 2006/119593 PCT/BE2006/000051 33 negative tumours, there may be further factors driving the underlying biology of metastasis besides cell-cycle associated genes. The prognostic ability of a "cell proliferation signature" in a subset of patients has been 5 reported previously in women who express relatively high estrogen receptor expression for their age 5. The analysis of the ER-positive subgroups was divided by genomic grade to the previously described luminal subgroups and this concept was validated in over 650 patients. Furthermore, 10 genomic grade remains the strongest variable in univariate and multivariate analysis (Table 4) that takes clinical prognostic factors into consideration. [0090] Currently there are several molecular signatures derived from microarray technology that claim to 15 be able to predict prognosis in breast cancer patients 8, 4, 7, 24 Some of these gene signatures reported can predict clinical outcome in ER-positive tumours treated with adjuvant tamoxifen , 24, 30 In the recurrence score developed by Paik et al. 7 the proliferation set of five 20 genes had the largest hazard ratios in their large training and validation sets and the highest "weight" or coefficient in their recurrence score formula indicating their high importance in deriving a prognosis classification for women with early stage breast cancer treated with adjuvant 25 tamoxifen. Proliferation-related genes appear to be an important-if not the most important-component of many existing prognostic gene signatures for breast cancer that are based on gene-expression profiles. By using the 11 genes in common between the GGI and a 70-gene prognostic 30 gene classifier for women with early stage breast cancer under the age of 55 4, similar survival curves to the validation publication 5 were obtained, suggesting that grade-related genes constitute a significant amount of the prognostic power of this signature. The subgroups achieved WO 2006/119593 PCT/BE2006/000051 34 by these prognostic signatures and that obtained by the classification of ER-positive tumours by genomic grade overlap significantly because of a strong dependence on cell-cycle genes to drive metastasis and relapse. The 5 advantage of this approach is that the biological mechanism that is responsible for the poor outcome is obvious, rather than a gene set that likely represents a variety of molecular functions and biological processes 8' 4. Because antiestrogens such as tamoxifen have a cell cycle-specific 10 action on breast cancer cells and influence the expression and activity of several cell cycle-regulatory molecules, the development of aberrant cell cycle control mechanisms is an obvious mechanism by which cells might develop resistance to antiestrogens. It is currently incompletely 15 understood why up to 30-40% of ER-positive breast cancers develop resistance to tamoxifen when positive expression of the ER is the best predictor predictors of tamoxifen response in the clinical setting 31. Over-expression of cyclin D1, a critical controller of the cell cycle, has 20 been associated with tamoxifen resistance and can reverse the growth-inhibitory effect of antiestrogens in estrogen receptor-positive breast cancer cells 3. Further investigation into the oncogenic pathways that drive the cell cycle machinery will be beneficial in developing new 25 agents to treat the high grade subgroup. [0091] Definition of clinically relevant tumour subclasses within ER-positive breast cancers is of great importance to the treating oncologist today. The emergence of new strategies of adjuvant anti-estrogen therapy 3* as 30 well as new chemotherapeutic and biological agents has made treatment decision making for women with early stage breast cancer sometimes a difficult task. Previously, tamoxifen was the mainstay of anti-estrogen therapy, with significant reductions in the risk or relapse, death and contralateral WO 2006/119593 PCT/BE2006/000051 35 breast cancer for women with early stage, ER-positive breast cancer 38. However, since the advent of aromatase inhibitors and the reporting of several trials finding them to be more effective than tamoxifen in postmenopausal 5 women, the American Society of Clinical Oncology has recommended that an aromatase inhibitor be included in the therapy of postmenopausal women with early stage hormone responsive breast cancers ". However, it is still unclear the best combination and sequencing of aromatase inhibitors 10 and tamoxifen, and whether all women with ER-positive tumours derive the same or differing benefit from these agents. The elucidation of clinically relevant and biological distinct hormone responsive breast tumour phenotypes can help facilitate the optimization of such 15 therapy as they may require different therapeutic strategies. [0092] In conclusion, the use of genomic grade can distinguish two subtypes with ER-positive breast cancers in a reproducible manner across multiple datasets and 20 microarray platforms. This is validated ept in over 650 ER positive breast cancer samples. These subgroups have statistically distinct clinical outcome in both systemically untreated and tamoxifen-only treated populations. Stratification by subtype in clinical trials 25 may provide important information on the potentially diverse effect of endocrine therapies, chemotherapies and biological agents on these subgroups. A focussed biological investigation into these distinct phenotypes may result in identification of separate and different therapeutic 30 targets. [0093] The genes identified herein may be used to generate a model capable of predicting the breast cancer grade of an unknown breast cell sample based on the expression of the identified genes in the sample. Such a WO 2006/119593 PCT/BE2006/000051 36 model may be generated by any of the algorithms described herein or otherwise known in the art as well as those recognized as equivalent in the art using gene(s) (and subsets thereof) disclosed herein for the identification of 5 whether an unknown or suspicious breast cancer sample is normal or is in one or more stages and/or grades of breast cancer. The model provides a means for comparing expression profiles of gene (s) of the subset from the sample against the profiles of reference data used to build the model. The 10 model can compare the sample profile against each of the reference profiles or against model defining delineations made based upon the reference profiles. Additionally, relative values from the sample profile may be used in comparison with the model or reference profiles. 15 [00941 In a preferred embodiment of the invention, breast cell samples identified as normal and non-normal and/or atypical from the same subject may be analyzed for their expression profiles of the genes used to generate the model. This provides an advantageous means of identifying 20 the stage of the abnormal sample based on relative differences from the expression profile of the normal sample. These differences can then be used in comparison to differences between normal and individual abnormal reference data which was also used to generate the model. 25 The detection of gene expression from the samples may be by use of a single microarray able to assay gene expression. One method of analyzing such data would be from all pairwise comparisons disclosed herein for convenience and accuracy. 30 [0095] Other uses of the present invention include providing the ability to identify breast cancer cell samples as being those of a particular stage and/or grade of cancer for further research or study. This provides a particular advantage in many contexts requiring the WO 2006/119593 PCT/BE2006/000051 37 identification of breast cancer stage and/or grade based on objective genetic or molecular criteria rather than cytological observation. It is of particular utility to distinguish different grades of a particular breast cancer 5 stage for further study, research or characterization. [0096] The materials for use in the methods of the present invention are ideally suited for preparation of kits produced in accordance with well known procedures. The invention thus provides kits comprising agents for the 10 detection of expression of the disclosed genes for identifying breast cancer stage. Such kits optionally comprising the agent with an identifying description or label or instructions relating to their use in the methods of the present invention, is provided. Such a kit may 15 comprise containers, each with one or more of the various reagents (typically in concentrated form) utilized in the methods, including, for example, pre-fabricated microarrays, buffers, the appropriate nucleotide triphosphates (e.g., dATP, dCTP, dGTP and dTTP; or rATP, 20 rCTP, rGTP and UTP), reverse transcriptase, DNA polymerase, RNA polymerase, and one or more primer complexes of the present invention (e.g., appropriate length poly(T) or random primers linked to a promoter reactive with the RNA polymerase). A set of instructions will also typically be 25 included. [0097] The methods provided by the present invention may also be automated in whole or in part. All aspects of the present invention may also be practiced such that they consist essentially of a subset of the disclosed genes to 30 the exclusion of material irrelevant to the identification of breast cancer stages in a cell containing sample. [0098] An exemplary system for implementing the overall system or portions of the invention might include a general purpose computing device in the form of a computer, WO 2006/119593 PCT/BE2006/000051 38 including a processing unit, a system memory, and a system bus that couples various system components including the system memory to the processing unit. The system memory may include read only memory (ROM) and random access memory 5 (RAM) . The computer may also include a magnetic hard disk drive for reading from and writing to a magnetic hard disk, a magnetic disk drive for reading from or writing to a removable magnetic disk, and an optical disk drive for reading from or writing to a removable optical disk such as 10 a CD ROM or other optical media. The drives and their associated machine-readable media provide nonvolatile storage of machine-executable instructions, data structures, program modules and other data for the computer. 15 [00993 Embodiments of the present invention may be practiced in a networked environment using logical connections to one or more remote computers having processors. Logical connections may include a local area network (LAN) and a wide area network (WAN) that are 20 presented here by way of example and not limitation. Such networking environments are commonplace in office-wide or enterprise-wide computer networks, intranets and the Internet and may use a wide variety of different communication protocols. Those skilled in the art will 25 appreciate that such network computing environments will typically encompass many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, network PCs, 30 minicomputers, mainframe computers, and the like. Embodiments of the invention may also be practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked (either by hardwired links, wireless links, or by a WO 2006/119593 PCT/BE2006/000051 39 combination of hardwired or wireless links) through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices. 5 [0100] Different embodiments of the present invention have been described according to the present invention. Many modifications and variations may be made to the techniques and structures described and illustrated herein without departing from the spirit and scope of the 10 invention. Accordingly, it should be understood that the apparatuses described herein are illustrative only and are not limiting upon the scope of the invention.

WO 2006/119593 PCT/BE2006/000051 40 Table 3: Up-regulat.ed in grade 3 tumor No. D PD>O FD>1 F0>2 probesat gene symbol description 1 2.1102 D.0001 0.0001 0.0001 2020540at UBE20 ubiquitin-conjugating enzyne 2 1.0037 0.0001 0.0001 0.0001 222077-.Eat RACGAPI ac GTPuse activating protein 1 3 1.7292 0.0001 0,0001 0.0001 201088..at KPNA2 karyoplerin alpha 2 (RAG cohort 1, importing al pha 1)* 4 1.7264 0.0001 0.0001 0,0001 218542-at ClDorf3 chromosome 20 open reading frame 3 5 L,7259 0,0001 0.0001 0.0001 203554Lx-at PTTGI piltary tumor-trunsforring 1 6 1.7053 0.0001 0.0001 0.0001 218355.at K1F4A kinesin family member 4K 7 1.6600 0.0001 0.0001 0.0001 210052-s-at TPX2 TPX2, microtubuL-aanciated protein honolog (Xenopus leevia) 8 1.6598 0,0001 0,0001 0,0001 202580.x.at FOXM1 forehead box MI 9 1654B 0.0001 0.0001 0.0001 208079....at STKC6 sarine/threonine kinaae 8 10 1.6513 0.0001 0.0001 0.0001 204092-Bsat STK6 aerine/threonine linahe 8 11 2.f495 0.0001 0.0001 0.0001 218755.at Ki1F20A kinein family member 20A 12 1.6387 0.0001 0,0001 0.0001 201584-..at DDX39 DEAD (Aep-Glu-AlAap) boy. polypeptde 30 13 1.6347 0.0001 0.0001 0.0001 20376A..at DLG7 discs, large homolog 7 (DroaophIla) 14 1.6223 0.0001 0.0001 0.0001 2048215-at MELY, maternal embryonic lencine zipper kinase 16 1.6213 0.0001 0,0001 0.0001 203418..at CONA2 cycling A2 16 1.095 0.0001 0.0001 0.0001 204766-B-at NUDT1 nudix (nucleonide diphoeplate linked moiety X) type Motif 1 17 1.0057 0.0001 0.0001 0.0001 206102..at KIAA0186 K IAAQI86 gene product 18 1.5986 0.0001~ 0.0001 0,0001 202005..-at BIRC5 baculovira) IA? repeat-containing 5 (survivin) 19 1.5957 0.0001 0.0001 0.0001 201710..at MYBL2 Y-nyb myeloblaetosis viral oncogene homolog (avian)-like 2 20 1.56879 0.0002 0.0001 0,0001 211762,...at KPNA2 karyopherin alpha 2 (RAG cohort 1, importing &I pha. 1) 21 1.6816 0.0002 0.0001 0.0001 209680..a..at KIFC1 Kinesin family member C1 22 1.5785 0.0002 0.0001 0,0001 209408-at KIF12C kinesi family member 20 23 1.5671 0.0002 0.0001 0.0001 219918-s..at ASPM Sp (abnormal opindle)-like, microcephaly aasoci ated (Drosophila) 24 1.5650 0.0003 0.0001 0,0001 203145.at SPAG5 sperm associated antigen 5 25 1.5595 0.0003 0.0001 0.0001 204962-..at CENPA centromere protein A, l7kDa 26 1.6551 0.0003 0.0001 .0001 202870-a..at CDC20 CDC20 cell division cycle 20 homolog (S. care visiae) 27 1,5446 0.0003 0.0001 0.001 38168-at E3SPLI xtra spindle pois like I (S. cerevisiae) 28 1.5376 0.0003 0,0001 0,0001 202107..a..at MCM2 MCM2 minichromosome maintenance deficient 2, mitotin (S. cerevisae) 29 1.5236 0.0004 0.0001 0.0001 204767..-at FENI flap structure-specific edonucleese 2 30 1.5226 0.0004 0.0001 0,0001 203046.s..at TIMELESS timeless homolog (Drosophila) 31 1.5221 0.0004 0.0001 0.0001 221677-B-at DONSON downstream neighbor of SON 32 1.5134 0.0005 0.0001 0.0001 210559..-at CDC2 cell civieion cycle 2, G1 to S and 02 to M 33 1.5047 0.0006 0.0001 0.0001 221520..B.at CDCA8 cell division cycle eseociated 8 34 1.5017 0.0007 0.0001 0,0001 214710-..at CCNB1 cycling Bl 35 1.4945 0.0007 0.0001 0.0001 209714.e..at CDKN3 cyclin-dependent kinase inhibitor 3 (CDX2 associated dual specificity phosphatase) 36 14933 0.0008 0.0001 0.0001 204444-at KIF11 kinslo family member 11 37 1.4927 0.0008 0.0001 0.0001 210821...at CENPA centromere protein A, I7kDa 38 1,4915 0.0008 0.0001 0.0001 218726..at DKFZp762E1312 hypothetical protein DICPZp762E1312 39 1.4895 0.0000 0.0001 0.0001 220651.-at MCM10 lvlC mniebTomesome maintenance deficient 10 (S. cerevislee) 40 1.4865 0.0010 0.0001 0,0001 201475.x..at MARS methionine-tNA eyntmtae 41 1.4715 0.0014 0.0001 0.0001 204033..at TRIP13 thyroid hormone receptor interactor 13 42 1.4672 0.0014 0.0001 8.0001 202705..at CCNB2 cycling B2 43 1.4624 0.0014 0.0001 0.0001 204649.at TROAP trophinin associated protein (tantin) 44 1.4603 0.0014 0.0001 0.0001 220060....at PL320641 hypothetical protein FL220641 46 1,4534 0.0018 0,0001 0.0001 209836..x..at LATI-3TM LATI.3TM protein 46 1,4533 0.0016 0.0001 0.0001 203276-at LMNBI lamin B 47 1.4471 0.0018 0,0001 0.0001 205034..al CNE2 cycling B2 48 1,4455 0.0018 0.0001 0.0001 203213..at 002 coil division Cycle 2, 01 to S and 2 to l 49 1,4384 0.0019 0.0001 0.0001 20D464..at AURKB aurora IinaaL B 50 1,4381 0,0019 0.0001 0.0001 2051046..at CENPEI; cntromere protein E, 312kD 51 1.4373 0.0019 0.0001 0,0001 203755..at BUB1B BUBl beddig uninbibited by banzimidazolen I liomulog beti, (yeast,) 52 1,4336 0.0010 0.0001 0.0001 203214.,x-atl 0002 ceil division cycle 2, Cl to S and 02 to M 53 1.4236 0.0022 0.0002 0.0001 214804..at FSIPRHI FS primary repone (LRPRl lomolog, rat) 1 54 1.4107 0.0026 0.0003 0.0001 2120,40-al BRtNI barren homlog (Droophila) 55 1-.4134 0.0027 0.0003 11.0001 2043111.at GTSB1 0-2 adR GTPd e aexpretinged I WO 2006/119593 PCT/BE2006/000051 41 Table 3: . . continued p-valuies No. D FD>0 FD>l PD>2 probeset gene symbol description 56 14105 0.0030 0.0003 0.0001 207165..at HiMMR hyaluronan-mediated motility receptor (RH A MM) 57 1.4079 0.0031 0.0003 0.0001 212022..sat MK67 antigen identified by ronoclonal antibody Ki-67 58 1.4051 0.0031 0.0003 0.000) 213226..at CCNA2 cyclin A2 59 1.3931 0.0041 0.0004 0.0301 219510..at POLQ poly'merase (DNA directed), theta 60 1.3890 0.0044 0.0004 0.0001 204026..n..at ZWINT ZW10 interactor 61 .3890 0.0044 0.0004 0.000) 203432.at TMPO thymopoictin B2 1.3872 0.0046 0.0004 0,001 204768..at FENI flap structure-specific endonuclease 1 63 1.3855 0,0047 0.0004 0.0001 209773..s-at RRM2 ribonucleotide reductnae M2 polypeptide 64 1.3847 0,0047 0.0004 0.0001 214431-at GIMPS guanine monphosphate synthetase 65 1.3842 0.0048 0.0004 0.0001 212023..e..at MKI67 antigen identified by monoclonal antibody KM-67 66 1.3752 0.0052 0.0004 0.0002 218883..U..-at ILP1IP MLFI interacting protein 67 1.3541 0.0077 0.0006 0,0003 211519..sat KIP20 kinesin family member 20 68 1.3503 0,0083 0.0006 0.0003 202240.at PLKI polo-like kinase 1 (DroeOphlla) 69 1.3460 0.0089 0.0007 0.0003 205733..-at BLM Bloom syndrome 70 1.3457 0.0092 0.0008 0.003 222039.-at LOC146909 hypothetica) protein LOC146909 71 1.3443 0.0096 0,0008 0.0003 209642-at BUBI BUB1 budding uninhibited by benzimidazoles I homolog (yeast) ' 72 1.3376 0.0102 0.0010 0.0003 21359.-at OIPb Opa-interacting protein 6 73 1.3372 0.0102 0.0010 0.0003 214096....at SHMT2 serine hydroxymethyltransferase 2 (mitochon drial) 74 1.3348 0.0105 0.0012 0,0003 211072..x.-at K-ALPHA-1 tubulin, alpha, ubiquitous 76 1.3237 0.0130 0.0017 0.0004 202779.-a-at UBE2S ubiquitin-conjugating enzyme E2S 76 1.3226 0.0133 0.0017 0,004 218447-at DC13 DC13 protein 77 1.3215 0.0138 0.0017 0.0004 213911.a-at H2AFZ H2A histone family, member Z 78 1.3211 0.0138 0.0017 0.0004 212141-at MCM4 MlCM4 minichromosome maintenance deficient. 4 (S. cerevisiae) 79 1.3156 0.0153 0.0019 0.0005 221591.-a-t FLJ10156 hypothetical protein FL310156 80 1.3139 0.0162 0.0019 0.0005 204822..at TTK TTK protein kinase 81 1.3121 0.0165 0.0020 0.0005 209251.x.at TUBA6 tubulin alpha 6 82 1.3086 0.0173 0.0023 0.0006 217835..x.at C20orf24 chromosome 20 open reading frame 24 83 1.3081 0.0176 0.0023 0.0006 201890.at RRM2 ribonucleotide reductaae M2 polypeptide 84 1.3009 0.0184 0,0024 0.0006 213671.a.at MARS imethionine-tRNA eynthetase 85 1.3063 0.0185 0.0024 0.0006 218009.a..at PRCI protein regulator of cytokinesi 1 86 1.3010 0.0197 0.0026 0.0007 207828...at' CENPF centromere protein F, 350/400ka (mitonin) 87 1.3002 0.0198 0.0026 0.0007 219555..at SM039 uncharacterized bone marrow protein BM039 88 1.2969 0.0206 0.0026 0.0007 204695-at CDC25A cell division cycle 25A 89 1.2953 0.0214 0.0026 0,0009 212021..a..at MK167 antigen identified by monoclona) antibody Ki-67 90 1.2898 0.0229 0,0028 0.0009 201090.x..at K-ALPHA-1 tubulin, alpha, ubiquitous 91 1.2885 0.0233 0.0033 0.0010 218039..at NUSAPI nucleolar and spindle associated protein 1 92 1.2851 0.0246 0.0034 0.0012 204603.at EXO1 exonuclease 1 93 1.2846 0.0248 0.0034 0.0012 203362.a..at MAD2L3 MAD2 mitotic arrest deficient-like 1 (yeast) 94 1.2845 0.0248 0.0034 0.0012 202094.at BIRC baculoviral 1AP repeat-containing 5 (eurvivin) 95 1.2840 0.0249 0.0034 0.0012 204162..at KNTC2 kinetochore associated 2 96 1.2825 0.0254 0.0037 0.0012 222036...-at MCM4 MCM4 minichromosome maintenance deficient 4 (S. cerevielae) 97 1.2780 0.0272 0.0039 0.0014 204252.at CDK2 cyelin-dependent kinase 2 98 1.2775 0.0274 0.0039 0.0014 219000..e..at DCC1 defective in sister chromatid cohesion homolog 1 (S. cerevisiae) 99 1.2772 0.0277 0.0041 0.0014 201524.x-at UBE2N ubiquitin-conjugating enzyme E2N (UBC13 ho.

molog, yeast) 100 1.2694 0.0294 0,0044 0.0018 204817.at ESPLI extra spindle poles like I (S. cerevisiae) 101 1.2657 0.0313 0.0046 0.0019 218662.a..at HCAP-G chromosome condensation protein G 102 1.2620 0,0324 0.0052 0.0022 206364..at KIF14 Kineein family member 14 103 1.2612 0.0329 0.0054 0.0022 221436..a..at CDCA3 cell division cycle aasociated 3 104 1.2609 0.0331 0.0054 0.0022 202 10.a-at SLC7A5 eolute carrier family 7 (cationic amino acid trans porter, y+ nyntem), member b 105 1.2002 0,0332 0.0055 0.0022 208696-at CCTb chaperonin containing TCP1, subunit 5 (epsilon) 206 1.2553 0.0358 0.059 0.0023 218556.al ORMIDL2 ORM3-.likee 2 (S. cerevisiao) 107 1.2476 0.0400 0.0069 0.0024 2)1058.x..at K-ALPHA-I tubulin, alpha, ubiquitous 108 1.2468 0.0407 0.0072 0.0024 212723..at PTDSR phosphatidyleerine receptor 309 1.2447 0.0)414 0.0076 0.0024 200022.at INASEH2A ribonuc)eanse 12, large subunit 110 1.2402 0.0450 0.0O84 0.0029 2)0334.-.x-at BIRC5 baculoviral lAP repeat-containing 5) (nurvivin) 111 1.2383 0.0461 0.0090 0.0031 203744..at NMGB3 high-mobility group box 3 112 1.2360 0.0478 0.0095 0.0034 2219306at KNSL7 kineein-lIke 7 113 1.2332 0.0509 (,0102 0.0038 220865.irat TPRT trane-prenyltransferae 114 1.2331 0.0510 0.0103 0,0038 204641..at NEK2 NIMA (never in miteoie gene a)-rclated kinane 2 WO 2006/119593 PCT/BE2006/000051 42 Table 3: . .. continued p-values No. D FD>0 FD>1 FI>2 probeset gene symbol description 11. 1.2319 0.0618 0.0104 0.0038 203358.s-at EZH2 enhancer of zeste homolog 2 (Drosophila) 116 1.2249 0.0581 0.0119 0.0044 213088.-sat DNAJC9 DnaJ (Hsp4O) homolog, subfamily C, member 9 117 1.2233 0.0593 0.0123 00047 21451 6.at H]ST1IH4B histone 1, -l4b 118 1.2226 0.0602 0.0124 0.0049 202110..at COX7B cytochrome c oxidase subunit VIb 110 1.2214 0.0610 0.0128 0.0050 218982.s..at MRPS17 mitochondrial ribosomal proLein S17 120 L.2207 0.0615 0.0129 0,0061 205339..at SIL TALI (SCL) interrupting locus 121 1.2206 0.0615 0.0129 0.0051 201342-at SNRPC small nuclear ribonucleoprotein polypeptide C 122 1.2199 0.0619 0.0131 0.0051 201678-a...at DC12 DC12 protein 123 .2192 0.0622 0.0132 0,01062 218875-E.at FBXO5 F-box protein 0 124 1.2160 0.0650 0.0139 0.0055 218663..L H CAP-G chromosome condensation protein G 120 1.2165 00652 0.0141 0,0055 212020.a..at MIK167 antigen identifed by monoclonal antibody Ki-67 126 1.2065 0.0738 0,0163 0.0065 217755...at HN1 hematological and neurological expressed 1 127 1.2028 0.0777 0,0176 0.0067 202635...at POLR2K polymerase (RNA) 11 (DNA directed) polypep tide K, 7.OkDa 128 1.2003 0.0802 0.0183 0.0069 202397..at NUTP2 nuclear transport factor 2 129 1.1993 0.0811 0.0183 0.0071 201930.at MCM6 MCM6 minichromosome maintenance deficient 6 (MIS5 homolog, S. pomsbe) (S. cerevisiae) 130 1.1960 0,0849 0,0194 0.0073 222037.at MCM4 MCM4 minichromosorne maintenance deficient 4 (S. cerevisiao) 131 1.1914 0.0900 0.0210 0.0083 205024..a..at RAD51 RAD51 homolog (RecA homolog, E. coll) (S. cere Visiae) 132 1.1884 0.0932 0.0217 0.0090 211750..xat TUBA6 tubulin alpha 6 133 1.1880 0.0938 0.0221 0.0092 203856-at VRK1 vaccinin related kinase 1 134 1.1861 0.0961 0,0229 0.0093 204267-.at PKMYT1 membrane-associated tyrosine- and threonine specific cdc2-inhibitory kinase 135 1.1807 0.1036 0.0256 0.0107 219787..a-at ECT2 epithelial cell transforming sequence 2 oncogene 136 1.1800 0.1045 0.0258 0.0109 219494..at RAD54B RAD54 homolog B (S. cerevisias) 137 1.1790 0.1050 0.0202 0,0110 219990..at FL323311 FL323311 protein 138 1.1770 0.1078 0.0268 0.0114 219061..a..at DXS9879E DNA segment on chromosome X (unique) 9879 expressed sequence 139 1.1767 0.1083 0.0269 0.0114 203832..at SNRPF small nuclear ribonucleoprotein polypeptide F 140 1.1757 0.1094 0.0274 0.0116 213646..x..at K-ALPHA-1 tubulin-alpha, ubiquitous 141 1.1749 0.1107 0.0277 0.0117 201619.at TOMM70A translocase of outer rnitochondrial membrane 70 homolog A (yeast) 142 1.1728 0.1131 0.0287 0.0121 202824..a..at TCEB1 transcription elongation factor B (SI1), polypep tide 1 (15kDa, eiongin C) 143 1.1726 0.1134 0.0290 0.0122 222029..x.at HKE2 HLA class 11 region expressed gene KE2 144 1.1714 0.1142 0.0296 0.0127 205644..a-at SNRPG small nuclear ribonucleoprotein polypeptide C 145 1.1664 0.1222 0.0317 0.0146 204170..s..at CKS2 CDC28 protein kinase regulatory subunit 2 146 1.1658 0.122B 0.0321 0.0147 205394.at CHK1 CHK1 checkpoint honolog (S. pombe) 147 1.1630 0.1270 0.0340 0,0155 204023.at RF04 replication factor C (activator 1) 4, 37kDa 148 1.1619 0.1289 0.0345 0.0155 218151..x..at GPR172A G protein-coupled receptor 172A 149 1.1616 0.1290 0.0345 0,0156 202352-s..at PSMD12 proteasome (prosome, macrops.in) 26S subunit, non-ATPase, 12 150 1.1597 0.1322 0.0362 0.0158 202188..at NUP93 nucleoporin 93kDa 151 1.1548 0.1420 0,0390 0.0175 201291..s..at TDP2A topoisomerase (DNA) 11 alpha 17OkDa 152 1.1528 0,146D 0.0404 0.0179 219978..s..at NUSAP1 nucleolar and spindle associated protein 1 153 1.1525 0.1462 0.0405 0.0182 201266.at TXNRD1 thioredoxin reductase 1 184 1.1514 0.1487 0.0415 0,0186 204126..-at CDC45L CDC45 cell division cycle 45-like (S. cerevisiae) 155 1.1608 0.1497 0.0418 0.0189 209709-s-at HMMR byaluronan-mediated motility receptor (RHA MM) 186 1.1501 0.1613 0.0421 D0189 219512..at C20orfl72 chromosome 20 open reading frame 172 157 1.1466 0.1883 0,0446 0.0204 218408..at TIMM10 translocase of inner mitochondrial membrane 10 homolog (yeast) 158 1.1444 0.1616 0.0457 0.0216 20155-at MCM3 MCM3 minichromosome maintenance deficient 3 (S, cerevisiao) 159 1,1413 0.1670 0.0479 0.0223 218239-a-at CTPBP4 GTP binding protein 4 160 1.1412 0.1674 0,0470 0,0223 200783.ui-at STMN1 stathmin 1/oncoprotein 18 161 1.1389 0.1729 0.0408 0.0228 214095.at S-M1AT2 nerine liydroxymethyltrannferase 2 (mltochon drial) 162 1.1380 0.1736 0.0803 0.0231 200853..at F2AFZ 112A histone family, member Z 163 L1346 0.1818 0.05r46 0.0248 203931--at MRPL12 mitochondrial ribosomal protein L12 164 1.1332 0.1840 (.0554 0.0254 20U744...at ITCH itchy hosMoIDu 3 ubiqUitin protein ligase (mouse) 165 1.1320 0.1846 0.0860 0.0256 212639.-x.at TUBA3 tubulin, alpha 3 166 1.1316 0.1873 0,0576 0,0259 204044..at QPRT quinlolinate phosphorlbosyltrannfeoras (nicotinate-nucleot.ide pyrophouphorylsase (car boxylating)) WO 2006/119593 PCT/BE2006/000051 43 Table 3: co entinued No1. D I'D >[ PD> I PD>2 probenet gene symobol deacription 167 1.1254 0.2017 0,0638 0,0296 20886soi-ij TXN thiorodoxin 168 1.1233 0.2063 0.0661 0.030 201114.xjnt PSivA7 proteenorne (prosome, racropain) aubunit, alphai type, 7 160 1. 122B 0.2073 0,0666 0.311 209372..s-al CISNPF centromere protein F, 360/400kiA (mitosin) 170 1,1224 0.2000 0,0672 0,0314 201577.-it NMIAE non-roetestetlc cell 1, protein (10023A) ex preened in 173 1.1204 0.212D 0.0699 0.0324 213330..s..t STIP I streenf-lnduced-p)hosphioprotein 1 (Hlsp70/Hep9O orga~nizing protein) 172 1.1197 0.21242 0.0701 0.0331 228230..at (CTPBP4 GTP binding protein 4 173 1.1102 0.2155 0.0708 0.0335 224437-B-a.t SIHMT2 serine liydroxymetLhyltrariafereae 2 (mitochon drial) 274 1,1181 0.2180 0.0726 0.0343 218027.nt MARPL16 mitochondrial ribosorma) protein LIS 176 1.1178 0.2103 0.0728 0.0346 203612-a.t BYSL ~ byBtin-lice 276 2.2 173 0.2200 0.07393 0.0147 202487-saet 1H2ArV 112A lietone- family, member V 277 2.1000 0.2410 0.081b 0.0300 218308..at TACC3 transforming, acidic coiled-coil containing protein 3 178 1,2080 0.2440 0,0023 8.0408 208511-at PTTG 3 pituitary tumnor-transforming 3 170 1.1070 0.2500 0.0840 0.0421 212160-at XPOT exportin, LIINA (nuclear export receptor for tR. NAB) 180) 1.1061 0.2541 0.0863 0.0429 2028..nt 162P1 22F transcription factor 1 181l 1.1037 0.2608 0.0000 0.0449 203746..n..t 1-100 holoicytochrornt c synthane cytochromee c hoe lyUBe) 182 1.1018 0.2555 0.0034 0.0464 219004-s-.at C2) orf45 chronnonone 21 open reading frame 45 183 1.1010 0.2681 0.0940 0.0473 206632-.e.at APOBEC313 apolipoprotein B mTRNA editing enzyme, cat alytic polypeptid,-lke 3B 184 2.2000 0.2691 0.0946 0.0478 219688..&.at IVTB more than blood bemolog 185 1.2000 0.2706 0.0053 0.0483 20393-sa.at CH-E3(1 CH1(1 checkpoint inomelog (S. pombq) WO 2006/119593 PCT/BE2006/000051 44 Table 3: Up-regulated in grade I tumors p-Vdues No. D PD>0 FD>1 PD>2 probeset gene symbol description 1 -1.4739 0.0014 0,0001 0.0001 233103-at STAfRD13 START domain containing 13 2 -1.4647 0.0014 0.0001 0,0001 204703-at TTC10 tetraricopeptide repeat domain 10 3 -3.4196 0.0024 0.0002 0.0001 218346.s-at SESNi estrin 1 4 -1.4084 0.0031 0.0003 0.0001 218471-s-at BBS) Burdet-Bied) syndrome 1 fi -1.3840 0.0048 0.0004 0,0001 205898-at CX30R1 clmokina (C-X3-C motif) receptor 3 6 -1.3482 0.0084 0.0007 0.0003 204072-s-at 13CDNA73 hypothetical protein CGO03 7 -1.323h 0.0131 0.0017 0.0004 219455.-at PLJ21062 hypothetical protein FL321062 8 -1.2840 0.0249 0.0034 0,0012 217889.-s-at CYBRDI cytochrome b inductee I 9 -1.2835 0.0252 0.0034 0.0012 219238-at FLJ20477 hypothetical protein F1,320477 10 -1,2663 0,0312 0.0045 0.0018 216264...at LA MB2 laminin, beta 2 (Iamic S) 11 -1.2656 0.0314 0,0046 0.0019 221562....nt SIRT3 rtuin (silent mating type information regulation 2 homolog) 3 (S. cerevinia.6) 12 -1.2628 0.0322 0.0049 0,0022 216620..e..at TPT1 tumor protein, translationally-conrolled 3 13 -1.2568 0.0351 0.0056 0.0023 220141.-.t FL323554 hypothetical protein FLJ23554 14 -1.2557 0.0356 0.0059 0.023 218483..s-at PLJ21827 hypothetical protein FLJ21827 15 -L,2548 0.0364 0.0060 0.0023 221771..s-at HlSMAPP8 M-pbnsn phosphoprotein, mppB 16 -1.2300 0.0450 0.0084 0.0029 220927.-.-at WDRI9 WD repeat domain 19 17 -1.2259 0,0574 0.0117 0.0044 212696..at CRY2 cryptochrome 2 (photolyaselike) 18 -1.2233 0.0693 0.0124 0.0047 213340..h.at KIAA0496 1AA0496 19 -1.216 0.0652 0,0140 0.006 213444-at KIAA0543 K1AA0643 protein 20 -1.2144 0.0662 0.0141 0.0057 220173-at C14orf45 Chromosome 14 Open reading frame 46 21 -1.2139 0.0666 0.0144 0.0059 201384-..at M17S2 membrane component, chromosome 17, surface marker 2 (ovarian carcinoma antigen CA125) 22 -1.2098 0.0711 0.0152 0.0062 203156..at AKAP11 A kinase (PRI(A) anchor protein 11 23 -1.2064 0.0740 0.0163 0.0065 209407-B-at DEAFi deformed epidermal autoregulatory factor (Drosophila) 24 -1.2017 0.0789 0.0179 0.0067 219469-at DNCH2 dynein, cytoplasmic, heavy polypeptide 2 26 -1.2003 0.0802 0.0183 0,0069 203984-a-at CASP9 ceepase 9, apoptois-related cysteine proteane 26 -1.1973 0.0837 0.0190 0,0071 217844-at CTDSP1 CTD (carboXY-termlnal domain, RNA poly Ita~ 1, polypeptids A) email phoephatese 1 27 -1.1906 0.0914 0.0212 0,0084 213397-x-at RNASE4 ribonuclease, tNms A family, 4 28 -1.1896 0.0919 0.0214 0,0088 206197.-at NME5 non-metaetatic calls 6, protein expressed in (nucleoeide-diphoephnte kinase) 29 -1.1878 0.0941 0.0221 0.0093 219922-a-at LTBP3 latent transforming growth factor beta binding protein 3 30 -1.1829 0.1003 0.0247 0.0102 201383-..at M17S2 membrane component, chromosome 17, surface marker 2 (ovarian carcinoma antigen CA125) 31 -1.1827 0.1007 0.0249 0.0102 206081-at SLC24A1 solute carrier inmIly 24 (sodium/potaseium/cnlcium exchanger), member 1 32 -1.1709 0.1153 0.0296 0.0129 213266-at 76P Gamma tubuli ring complex protein (76p gene) 33 -1.1707 0.1156 0.0297 0.0129 209189-at FOS v-foe PB. marine oeoearcoma viral oncogene homolog 34 -1.1679 0.1201 0.0307 0,0142 234829-at AASS aminoadipa-emialdehyde synthase 36 -1.1633 0.1268 0.0337 0.0153 221123-x.-at ZNP395 zinc finger protein 39 36 -1.1625 0.1279 0.0344 0.0165 200810-a-at CIRBP cold inducible RNA binding protein 37 -1,1612 0.1298 0.0364 0.0157 210365..at RUNX1 runt-related transcription factor 1 (acute myeloid leukemia 1; amli oncogene) 38 -1,1602 0.1314 0.0368 0.0168 212842..x..at RANBP2LI RAN binding protein 2-like 1 39 -1.1695 0.1324 0.0362 0.0158 213364-a-at SNXI Sorting nexin 1 40 -1.1686 0.1339 0.0369 0.0161 220911..s..at KIAA1306 XIAA1306 41 -1.1539 0.3444 0.0398 0.0175 201335..s..at ARMGEF12 Rho guanine nucleotide exchange factor (02F) 12 42 -1.1499 0.1619 0.0421 0.0190 221276.s..at SYNCOILIN intermediate filament protein syncoin 43 -1.1463 0.1600 0.0461 0.0200 221824..-at MIR c-mir, cellular modulator of immune recognition 44 -1.1437 0,1026 0.0461 0.0217 211943.-x.-at TPTI tumor protein, tranlationally-controlled 2 46 -1.1408 0.1681 0.0481 0.0223 218552-at LJ10948 hypothetcal protein FLJ10948 46 -1.1382 0,1743 0.0510 0.0232 220326-.-t PLJ10357 hypothetical protein PLJ10367 47 -1.12b4 0.2014 0.0638 0.0296 212869y..-at TPTI tumor protein, translntionally-controlld 1 48 -1.1184 0.2172 0,0722 0.0342 218648-at TORC3 transducer of regulated cAMP response element binding protein (CIIEB) 3 49 -1.1171 0.2215 0.0735 0.0348 212549-at STAT5B ignal transducer and activator of transcription 5B 60 -1.1165 0.2233 0.0743 0.0366 219961-a-at C20orfl2 chromoeome 20 open Tending frame 2 51 -1.1147 0.2277 0.0703 0.0364 212678-at NF1 Neurofibromin I (neurofirmtosie, Von Reck iAghuTn dioain, W tiOn di 3a) 02 -1.1113 0.2369 0.0807 0.0386 230852..s-jit AASS iBiodpt-smsdhd ynthase WO 2006/119593 PCT/BE2006/000051 45 Table 3: ... .otnr 1 )-vfl 08 NO. D PD>fl FD>a I D> 2 probeoat gene symobo) decripLion 53 -~ 3112 0.2370 0.0807 0.0387 202962..nt ]CIF13B kiNesin fanfly member 13 54 -1.3 102 0.2404 0.0813 0.0)394 214 724.nL D3XDCl DIN domain contuiming I 58 -J.3089 0.2450 0.0823 0.0408 201342..nat SMARCA2 SWI/SNqF related, matrix amsochted, actin de pendent rt~gu~aL~r of chrometin, subfai)l a, member 2 563 -1.3085 0.2464 0.0827 0.0411 207757..at PLJ21282 hy-potheticel protein FtlJ21628 57 -1.1085 0.2464 0.0827 0.0411 2]1686nL TBClDI7 'IDOl domatin family, member 17 WO 2006/119593 PCT/BE2006/000051 46 Table 3: Up-regulated In grade 3 tumors No. D FD> C PD> PD>2 probeat gene Isymbo) description 2 2.] 62 0,C)000 (),(001 0.0001 2020574.at UBIE2C nb!0uiG11-cojUg,,tjng af2Vn)e E2C 2 1037 (1.0001 0,0001 0.0001 222077..E-at RACGAPI Re GTPaae IctiVng protein I 3 ,7292 (.0002 0.0001 0.0001 201088-1" KCP1A2 hnryopberin alph 2 (RAG cohort 1, inportin ul pbii 1) 4 .7264 (1.0001 0000 0(0001 218542...at C)Dorf3 cbrnome ID open reelog frame 3 5 3.7250 0.0003 .] 0,0001 203514.x..a. PTTG1 pitiry Lumor-trene(orming I 6 3.7053 0, 0.0001 0,0003 (,000] 2) 8355iad KIP4A kinesin (emily member IA 7 1.6600 0.0001 0,0001 0.0001 21 0052..s..al TPX2 TPX2, NIcroLnbolahocieted protin homolog (PenoPue leeviH) 8 2.0598 0,0001 0.0001 0.0001 202580.x...at PoXMi lor1head boy MI 0 1.86.48 0.0001 0.0001 0,0001 208070.-Bat STK6 scrine/01ronine kinabc 6 10 1.6513 0.0003 0,0001 0.0001 20)4092..n..at STK6 sprInL/threonine kine 6 12 2.6495 (.0001 0.0001 0,0003 218755..at 1C1F20A kincein famly marnbar 20A 12 1.6387 0.0001 0.000) 0,0001 201584B.at DDX39 DEAD (Asp-01e-Ala.Aup) box polypeptide 39 13 3 I.347 0.0003 0,0001 0.0001 203764-al DLG7 thee, large homoleg 7 (Drosophlin) 14 3.6223 0.0001 0.0001 0.0001 204825-at MELY maternal embryonic macme zipper kina 15 1.6213 0.0001 0.0003 1.0001 203418.aL CONA2 cyclic A2 16 1.6095 0,0003 0,0001 0.0001 204766.-at NUDT1 nodix (nucleoside diphosphnLe linked moiety type motif 1 17 1,6057 0.0001 0.0001 0,0001 200102..at KIAAO186 X1AAO26 gene product 18 1.5986 0.0001' 0,0001 0.0001 202095.s..a BIRC5 baculoviral IAP repeat-containing 5 (survivin) 19 1.5957 0.0001 0.0001 0.0001 201710,at MYBL2 v-rnyb myeloblnionie viral oncogene homolog j1 (avian)-ime 2 20 1.5879 0.0002 0.0001 0.0001 211762..a..at KPNA2 keryopberin alpha 2 (RAG cohort 1, importing al phE 1) 21 1.5816 0.0002 0.0001 0.0001 200680.n-at KIFC1 Kinenin family member 01 22 1.5785 0,0002 0.0001 0.0001 209408-at KIF2C kinesio family niober 20 23 1.5671 0.0002 0.0001 0,0001 219918.-at ASPM ap (abnormal spindle)-llke, microcaphrjy wsnoci ateri (Droaopblla) 24 1.5650 0.0003 0.0001 0,0001 203145..at SPAG5 sperr aociated antigen 5 25 1.5595 0.0003 0.0001 0,0001 204962.-at CENPA centromere protein A, l7kDa 26 1.5551 0.0003 0,0001 0.0001 202870..a..at CDC20 CD020 cell division Cycle 20 hoMolog (S, cere visae) 27 1.5446 0.0003 0.0(1 0,0001 38158..at ESPLI extra pindle poles like I (S. cereviae) 28 1.5376 0,0003 0.0001 0.0001 202107..at MOCM2 MOM2 oinicbromosore maintenance deficient 2, coitotin (S. caravisae) 29 1.5236 0.0004 0.0002 0.0001 204767..a..at FENI fap atrcture-pecifir endorucleauc I 30 15226 0,0004 0.0001 0.0001 203046..a-at TIMELESS timele boog (Droaopbla.) 31 1.5221 0.0004 0.0002 0,0001 221677...at DONSON downstream neighbor of SON 32 1.5134 0.0005 0,001 0,0001 210559.a..at CD C2 cell division cycle 2, G1 to S and G2 to M 33 1.5047 0.0006 0.0001 0,0001 221520.a-at CDCA8 cell division cycle associated 8 34 1.5017 0.0007 0.0002 0,0001 214710..-at CONB1 cycling BI 35 1.4945 0.0007 0.0001 0.0001 209714.a-at CDKN3 cyclin-dependant kinana inhibitor 3 (003(2 associated dual apecificilty pheaphatase) 36 1.4033 0.0008 0.0001 0.0001 204444..at KIF11 kinaain family member 11 37 1.4927 0,0008 0.0001 0,0002 210821.x..at CENPA centromere protein A, l7kDB 38 1.4915 0.0008 0.0001 0.0001 218726..at DKFZp762E1312 bypotbetical protein DHFZp762E1312 39 1.4895 0.0009 0.0001 0.0001 220651.e..nt MOM10 MOC ID mirichmomoaome maintenance deficient 10 (S. acrevifliac) 40 1,4865 0.0010 0.0001 0,0001 201475..x..at MAILS nietionine-ttNA nyntbotne 41 1.4715 0,0014 0,0001 0.0001 204033.at TRIP13 thyroid hormone receptor interacT 13 42 1.4672 0.0014 0,0001 .3 202705-at CCNBl2 cyclic B2 43 3.4624 0.0014 0.000 0,0001 20464D..at TROAP trophinin meacieted protein (tatin) 44 1.4003 0.0014 0.0001 0.0001 220060...at FL320643 bypolheticul protin FLJ2D641 45 2,4534 0,0016 0,000) 0.000] 200830.x-at LATJ-3TM LATI-3Ti protin 46 1.4533 0.0016 0.0003 0.0003 203276..at LMNB1) Imin Dl 47 1,4471 0,0018 0.0001 0.0002 205034..at CCNE2 cyclic E2 48 1.4455 0.0018 0.0003 0.0001 203213-at CDC2 cell divialon cycle 2, 03 to S end 02 to M 40 1.4384 0.0011 1.0001 0.0001 209464.at AURKB auror line B 50 2,438) 0.0010 0.0001 010001 205046..at CENPE Mntromer- protein E, 1lcDn 63 1.4373 0.0019 0.00] 0.0001 203755..at BUB1B BUD) budding ninlibiwd by beoeimidleolon I honolog beti, (youat) 52 1.4336 (.0010 0.1.00 0.01) 211214xat CDC2 cell division cycle 2, 01 to S imd G2 to M 53 1.4230 0.0022 0.0002 0,13001 214 804-at PS].1Pfl341 PS1 primary ruapenwm (LRPRI hurnolog, rat) 2 54 1.467 0.0026 (,.003 0.1001 2J2940..t D1713N) barren honolog (DrBRopRNlln) 56 1.26.0 .(3 233 n GTS G-2 jid ' R -pcGT a expre aed I kary4heri 0.0pha 2,00 (RGohr0100prtn1 WO 2006/119593 PCT/BE2006/000051 47 Tu3ab 3: , ,, conlinurd j'-yhiuea No. D PD>0 PD> PD>2 probeset gene symnbo) description , 3.4305 0.0030 (.0003 0,D5001 207165..at )II lAIR hvliroiiin Ld ited ruotY receptor 57 1.4079 0.0031 0,00 3 0.0001 21 2022--at M1067 antigen identfied by monoclol antibody M-67 58 1.4051 0.0033 0.0003 0.0001 233226..at CCNA2 cycling A2 50 1.3033 0,0041 0,0004 0.000) 210510-at POLQ polymernee (DNA dicedd, theta 60 1.38010 0.0044 0.0004 0.1M 204026....at ZWINT Z' interactor 61 1,3890 0.0044 0.0004 01,001 203432.-at TIMPO thymopoictin 62 1.3872 0.0046 0.004 0.0001 204768..-at FENI flap 1tructure-spefiic endonucjc I 63 1.3865 0.0047 0.004 0.001 209773...at RRA2 ribonileotide reduame ('2 polypeptide 64 1.3847 .0047 0.0004 0.000) 214431.at GMlPS guanine monPhosplitn iynthotao 66 1.3842 0,0048 0.0004 0,001 212023...at M'4167 antigen identified by monoclonal ntibody (-07 66 3.3752 (1.002 0.0004 (.0002 218883-.at IALFIIP IALPI interacting PLotPin 67 1.3641 0.0077 (.000 0.0003 213 51OJi-at K3IF2C kinouin fily membin 2C 68 1.3503 0.0083 0,0006 ((.0003 202240..at PLKI polo-like Inae I (Drecophila) 69 3.3460 0.0089 0.0007 0.0003 205733..at BLM Bloom syndrome 70 1.3457 0.0092 0.0008 (.013 222030..at LOC346D00 hypothetcal protin L0146909 71 1.3443 0.0096 0.0008 0.0003 209642-at BUBI BUB) budding uninhibited by bezimidazolos I lIoMOlOg (yceat)' 72 1.3376 0.0102 0.0010 0.0003 213599..a( t COPS Opa-itracting protein 73 1.3372 0.0102 0.0010 0.0003 214096ah.at SHMT2 Berine hydroxyrnthyltranefrae 2 (rnlochon drial) 74 1.3348 0.0105 0,0012 0.0003 211072..x..at (-ALPHA-1 tubulin, alphn, ubiquitouu 75 1.3237 0.0130 0.0017 0.0004 202770..eat UBE32S ubiquitin-conjugating enzyme 28 76 1.3226 0.0133 0.0017 0,0004 218447..at DC13 DC13 protein 77 1.3215 0.0138 0.017 0.0004 213911-.at H2AFZ 12A histone family, member Z 78 1.211 0.013B 0.0(17 0.0004 212141..aAt MCM4 IA 4 minichromoome maintenance deficient 4 (S. cereviniae) 79 1.3156 0.0153 0.0010 0.0005 221591.s-at FLJ10156 hypothetical protein PL110166 80 1.3139 0.0162 0.0019 0.0006 204822-at TTK TT( protein kinase 81 1.3121 0.0165 0.0120 0.0005 209251..x-at TUBA6 tubulin alpha 6 82 1.3086 0.0173 0,0023 0.0006 217835.x-at C2Oorf24 chromosome 20 open reading frame 24 83 1.3081 0.0176 0.0023 0.0010 201890-at RRM2 ribonucleotide reducte M2 polypeptide 84 1.3059 0.0184 0,0024 0.0006 213671-a-at MARS methionine-tRNA Hynthetase 85 1.303 0.0186 0.0024 0.0006 216009-a.-at PRCI1 protein regulator of cytokinea 1 86 1.3010 0.0197 0.0026 0.0007 207828-..at CENPF centromere protein F, 350/4OMk (mitoein) 87 1.3002 0.0198 0.0026 0.0007 219555..z..at BM039 uncharacterized bone marrow protein BM03D 88 1.2969 0.0206 0.0026 0.0007 204695..at CDC25A cell division cycle 2(A 89 1.2953 0.0214 0.0026 (.0009 212021..s-at MK167 antigen identified by monoclonal antibody Ki-67 90 1.2898 0.0229 0,0028 (100009 201090.:..at K-ALPHA-1 tuhulin, alpha, ubiquitous 91 1.2885 0.0233 0.0033 0,0010 218039-at NUSAP1 nucleolar and spindle aslaociated protein 1 92 1.2851 0.0246 0.0034 0.0012 204603-at EXO1 exonoclease 1 93 1.2846 0.0248 0.0034 0.0012 203362-a-at MAD2L) MAD2 mitoic arrest deficient-like I (yeast) 94 1.2845 0.0248 0,0034 0.0012 202094-at BIRO5 baculoviral IA? repeat-containing 5 (survivin) 95 1.2840 0.0240 0.0034 0.0012 204162..at KNTC2 kinetochore associated 2 96 1.282(1 0.0254 0.0037 0,0012 222036....at 1MCM4 MCM4 ninicbromoeome maintenance deficient 4 97 1.2780 0.0272 0.0039 0.0014 204252-at CDK2 cydlin-dependent Moms 2 08 1.277 0.0274 0,0039 0.0014 219000..e..at DCO) defective in sister chromatid cohesion hoMOlog 1 (S. cerevisiale) 99 1.2772 0.0277 0.0041 0.0014 2031524..x-at UBE2N ubiquitin-conjugnting enzymu MN (UB033 ho molog, yeast) 100 1.2694 0.0294 0,0044 0.0018 20417-.at ESPL1 extra spindle polen lie 1 (S, cerevielne) 101 1.2067 0.0313 0.0046 0.0019 218 662..sat HCAP-C chromosome condensation protein C 102 1.2020 0.0324 0.002 0,0022 206364.at KIP14 Kiin (emIly member 14 303 1,2612 0.0320 0.0054 0.0022 221436..s.-at CDCA3 call divielon cycle anoclatod 3 104 1.2609 0.0331 1.00(4 0.0022 203 95..n..at SLC7Ab solu carrier famly 7 (cationic amino acid trans porter, y4 system), nwieor 6 105 1,2602 0.0332 0.00 1(10022 208696-at CCTb clinperonin containing TOP1, subuldt 5 (ePellon) 106 1.2553 1,.358 0,.0(5 0,0023 218556..at ORADL2 - 31(A-like 2 (S. erevisiac) 107 3.2476 0.0400 ((.0( 1(,1024 21 J01058...0at E-ALPHA-1 tlin, nipha, ubiquitous 108 1.2408 0.0407 0.0072 0.(024 212723-at PTDSR phosplitidylirie receptor 101) 1.2447 0,0414 0.0076 ((.1(24 203022-at RNASE12A ribonuclinee 312, lrge subunit 310 1.2402 0.0450 0.)084 0.002f 230334.;;-ut BIRC5 heculoviril 1AP e (nurvivin) 111 1.2383 0.0411 .0(00 0,0031 203744.-at HM10CB3 hiIl-nobill.r group box 3 112 1.230 0.0478 0.000 0.0034 21(130Gat HNSL7 lrii)-lli 7 113 3.2332 0.10509 0.0102 0,0138 22016.n..at TPRT 334 1.2333 0(0510 (0.0303 0100-38 21)48414 1 13K2 NMIIIA (nuveL In monit eobunrn n-)-meilitdd kimose 2 WO 2006/119593 PCT/BE2006/000051 48 Table 3: .. . contained No,. D FD>0 PD>3 FD>2 probaset gone symbol description 115 1.2319 0.0518 0.0)[4 0.0038 20335.s-nt EZH2 enhancer of zeste hoiolg 2 (Drosophila) 316 1.2240 0.0583 0.0119 0,0044 2)3088.s.aut DNA3O9 Dna. (Hsp4O) homulog, subfanily C, member 0 117 1.2233 0.0593 0.0123 0.0047 214516-at HJSTIH14B histone 3, H4b 118 1.2226 0.0602 0,0124 0.11049 202110..at COX7B cytochrome c oxidnas subunil VIIb 310 1.2214 0.0610 0.0128 0.000 218082.s..a. IMRPS17 miochondrial ribonornal proLin S17 120 1.2207 0.06151 0.0129 0,0051 205339-at SIL TALl (SCL) interrupting locus 121 3.2206 0.0615 0.0129 0.0051 201 342..at SNRPC Small nuclear ribnucleoprotein polypeptide C 122 1.2190 0.0629 0,0131 0,0051 201 678.i-at DC12 DC12 protein 123 3.2192 0.0622 0.0132 0.0062 218875.na..at PBXO5 F-box protein 5 124 3.210 0.0650 0.0130 0.0056 238663.at H CAP-G chromosome coidensation protein G 126 1.21 60 0.0662 0.0141 0,0066 212020..st MAK167 antigen identifed by monoclonal antibody Ki-67 126 2.2065 0.0738 0.0103 0.0065 217755-at 1-1N1 hmatological and neurological exprennod 3 127 1.2028 0.0777 0,0176 0.0087 202635..nat POL-t2( polymernae (R1NA) 11 (DNA directed) polypep tide X, 7.OkDa 128 1,2003 0.0802 0.0183 0.0069 202397-at NUTP2 nuclear transport factor 2 129 2.1993 0.0811 0.0183 0.0071 201930..aIt MCV6 MCIM6 minichromosome maintenance deficient 6 (M115 hornolog, S. pombe) (S. cereviaiae) 130 L,1960 0.084D 0.D194 0.0073 222037-ast MClCM4 MCM4 minichromosome maintenance defcient 4 (S. cerviniane) 131 1.1914 0.0900 0.0210 0.0083 205024..at RAD5I RAD51 homolog (RecA homolog, E. coll) (S. cere visiae) 132 1.1884 0.0932 0.0217 0.0090 211750..x..at TUBA6 tubulin alpha 6 133 1.1880 0.0938 0.0221 0.0092 203856..at VRK1 vaccinia related kinase 3 134 1.1861 0.0961 0.0229 0.0093 204267.x.-at PICMYTI membrane-associated tyrosine- and threonine specific cdc2-inhlbitory kineae 135 1.1807 0.1036 0.0256 0.0107 219787.s..at E CT2 epithelial cell transforming sequence 2 oncogene 136 1.1800 0.1045 0,0268 0,0109 219494..at RAD54B RAD54 homolog B (S. cereyisias) 137 1.1790 0.1050 0,0262 0.0110 219990..at FL323311 FLJ23311 protein 138 1.1770 0.1078 0.0268 0,0114 219061-a-at DXS9879E DNA segment on chromosome X (unique) 9879 expressed sequence 139 1.1767 0.1083 0,0269 0.0114 203832.at SNRPF small nuclear ribonucleoprotein polypeptide F 140 1.1767 0.1094 0.0274 0.0115 213646.x..at K-ALPHA-1 tubulin-alpha, ubiquitous 141 1.1749 0.1107 0.0277 0.0117 201519.at TOMM70A translocase of outer mitochondrial membrane 70 homolog A (yeast) 142 1.1728 0.1131 0.0287 0.0121 202824"..at TCEBI transcription elongation factor B (SII), polypep tide I (15kDa, elongin C) 143 1.1725 0.1134 0.020 0.0122 222029,.x.at HKE2 HLA class U1 region expressed gene KE2 144 1.1714 0.1142 0.0296 0,0127 205644..a.at SNRPG small nuclear ribonucleoprotein polypeptide C 145 1.164 0.1222 0.0317 o.0146 204170..at CKS2 CDC28 protein kinase regulatory subunit 2 146 1.1658 0.1228 0.0321 0.0147 205394..at CHEK CHK1 checkpoint homolog (S. pombe) 147 1.1630 0.1270 0.0340 0.0155 204023..at RF04 replication factor C (activator 1) 4, 37kDa 148 1.1619 0.1289 0.0345 0.0155 218151..X-at GPR172A G protein-coupled receptor 172A 149 1.1616 0.1290 0.0345 0.0155 202352..a-at PSMD12 proteasome (prosome, ma.cropain) 26S subunit, non-ATPase, 12 150 1.1597 0.1322 0.0302 8.0168 202188..at NUP93 nucleoporin 93kDa 151 1.1548 0.1420 0.0390 0,0175 201291-b-at TOP2A topoisomerass (DNA) II alpha 170kDa 152 1.1528 0.1459 0.0404 0.0179 219978..at NUSAPI nucleolar and spindle associated protein 1 103 1.1625 0.1462 0.0405 0.0182 201266.at TXNRD1 thioredoxin redUctaso 1 164 1.1514 0.1487 0.0415 8.8186 204126.-4at CDC46L CDC45 cell division cycle 45-like (S. cerevisiae) 155 1.1508 0.1497 0,0418 0.0180 209709..B.at HMMR hyaluronan-mediated motility receptor (RHAMM) 156 2.1501 0,1513 0.0421 0,0189 210512.at C20orf172 chromosome 20 open reading frame 172 157 1.1460 0.1583 0.0446 0,0204 218408.at TIMM1O tranniocase of inner mitochondrial membrane 10 Iomolog (yeast) 158 1.3444 0.1616 0.0457 0,0215 201555..at MCM3 MCM3 minichromonone maIntenance deficient 3 (S. cerevisins) 159 3.1413 0.1670 0.047P 0,0223 218230..-at CTPBP4 GTP binding protein 4 200 1,1412 0.1674 0.0470 0.0223 200783-i-at STNINI athmin 1/ancoprotein 18 10) 1.1389 0.1720 0.0408 (,0228 214095-..t SHMT2 nerine hrydroxymethryl-ransforane 2 (mLtochon drial) 102 1,1386 0.1735 1.1500 0.0231 20l83at H 1-12AFZ hione imnily, meoxber Z 303 1.1346 0.1818 10.11545 0.0248 203931-.-.at MRPL2 rMtUchondrial ribaSoMal protein L12 164 1.11132 0.21840 0.4 .,24 209744.x.Jet JTCH itchy hoiolog E ubiquitin protein by,lae (moune) 105 1.1320 0,3846 0.0560 0.0256 212639.-.at TUBA3 tubulin, alpha 3 166 1.13110 0.1873 0.0576 0,0268 2104044-at QPRT quinolinate phosphorlbouyltrannleraae (ni cotinate-ncleotidt pyrophouphorylans (car boxy-lating)) WO 2006/119593 PCT/BE2006/000051 49 No. D JPD>C PD>] I D>2 probWIct 91C~ .YMIJOI doicription 167 31.1254 0.,2017 0.0638 0.01219 208864-s-at TXN thiorecloxiri 18 3.1233 0.2063 0.0661 0.030,4 201 114..x.Jal PSIVIA7 pmote~hrnt (procornL, rnncropuln) nubuit iOlpim type, 7 100 1,1228 0.2073 016 [.0311 2003 72-. -at C2N PF CbnLrornerrt Protein F, 35(1/400kui (mTnO~uin) 170 1.1224 0.2066 0.0(672 0,0014 20307..at INlr4E1, noit-niens itic calls 1, protein (141%23A) fex prLecle in 171 1.1204 0.21281 0.0808 0.,0324 21 3330..ajdt STIP3 te~-fde-loporti (-lrrp7 0/H p 90 orgimizing protein) 172 1.1197 0.2142 0.070 0.0311 23 8238-at G3TPBP4 GT1P bindi protein 4 173 1.1192 0.2166 0.0708 0.0335 23 4437.e6.&L S1-11W2 firine lyryrt'1eneru 2 (moitochon dria)) 174 1,118) 0.218BO 0.0726 0. 63 43 2 180 2 7 .nt IvIRPLIb mitochondrial riboanrnal protein L16 170 1.1 178 0.2103 0.0728 (1,346 2031112..it fl-YSL b.Yatn-l11w 176 1.11731 0.2209 0.0733 0.0347 202487a..li. H2APV M1A luiiitone family, member V 177 1.1099 0.2410 0J.081b (1,0380 21 9308.-at TACC3 transforming, aicidic coledc-coll containing protein 3 178 1.108 0.2449 0,0823 0.0408 208511-at PTTG3 pituitary tUrnor-trunalormirig 3 178 1.1070 0.2500 0.0848 0.0421 2121 Majt XPOT wcportiin, fll4qA (nuclear export receptor for tLR N As) 180 I1.1061 0.2541 0,0863 0,0429 2028..a..at E2F1 B2P transcription factor 1 181 1.1027 0.2608 [1.0900 0.0(449 203746-Bat 1- OGS bolocytoclIromC c synthaeti (cytochronue c henm lyanun) 182 1.1018 0.266 0.01134 0,0464 219004-s.at C21orf4b chnromosnomeu 21 open reading frame 45 183 1.1010 0.2681 0.0840 0.0473 2061132..&.at AP0612C313 apolipoprotein B rnnRNA editing enzyme, cat alyfic p olypeptid a-like 3B 184 1,1006 0.2691 0.0846 0.0478 21988..a.at MTB more than blood hornolog 186 1.1000 0.2706 0.0863 0.0483 205303-a-at CHEMC CHIC] chneckpoint hornolog (S. pombe) WO 2006/119593 PCT/BE2006/000051 50 Table 3! Up-rTguliated in grade I tumore p- vahms]O No. D FD>J PD>] PD>2 pmobaset gene eymbQ1o) description 1 -1.4739 0.0014 0,0001 0.0001 2) 3103-at STARD] 3 START ilumin containing 13 2 -1.4647 0.004 0.0001 0.0001 201703.n.t TTC10 tetriiricopvptdepeat domain 20 3 -1.4196 0.0024 0.0002 0.0001 2)8346.h-nt SESN1 sitrin 2 4 -).4084 0.0031 0,0003 0.0001 22 1471..s.at BBS2 BurdOL-Biadl syndrome 2 5 -2.3i40 0.0048 0.0004 0.0002 205808.at CX30R1 cIemnkinC (C.i-C motif) receptor 2 6 -1.3482 0.0084 0.0007 0.0003 204072-at 13CDNA73 hypoheLicid prowin CGO03 7 -1.323b 0.0131 0.0017 0.11004 23 9455-a FLJ 2)D2 hypothetical protein F).052 062 8 -1,2840 0.0249 0.0034 0,0012 2178890s.n-at CYBRDI cytochroie 1) Telicutie 2 9 -1.2835 0.0252 0.0034 0.0012 2]19238-ut FL320477 bypOtieicai Protein FLJ20477 10 -1.2663 0.0312 0.0046 0.0018 236264-..at LAMB2 luminin, bet, 2 (larninin S) 11 -1.2606 0.0314 0.0046 0.0019 221562..s..at SIRT3 nirtuin (aflenl meting type Information regulation 2 liomolog) 3 (S. cerevIiuae1) 22 -1.2628 0.0322 0.0040 0.0022 216520-..a-t TPT3 Luror protein transtionally-controllod 1 13 -2.2668 0.0361 0.0056 0.0023 220141.at PLJ23554 iypotLieca protein FL323554 14 -1.2557 0.0366 0,0059 0.0023 238483..n..at PLJ21827 hypothetical protin PL.21027 15 -1.2548 0.0364 0.0060 0.0023 221771.a..at H1SMiPP8 v-phias phosphoprotein, rppO 16 -1.2399 0.0460 0.0084 0.0029 22007-a-at WDR19 WD repeal domain 2. 17 -2.2250 0,0574 0,0117 0.0044 23269E-at CRY2 cryptoebrome 2 (photolyue-Ike) 18 -2233 0.0503 0.0124 0.0047 213340....at IlAA0495 IAA049b 10 -1.2156 0.0652 0.0140 0.0055 213444-at KC1AA0043 )C1AAM43 protein 20 -1,2244 0.0662 0.0141 0.0057 220173..at C24orf45 chromosome 14 open reading frame 40 21 -1.2139 0.0666 0.0144 0.0059 201384-..at M I17S2 membrane component, chromosome 17 surface merker 2 (ovarian carcinoma antigen CA125) 22 -1.2098 0.0711 0.0152 0.0062 203156..at AK{AP11 A kinase (PRKA) anchor protein 11 23 -1.2064 0.0740 0.0163 0,0065 209407..e..at DEAFI deformed epidermal autorgulatory factor I (Dronophila) 24 -1.2027 0.0789 0.0170 0.0067 219469-at DNCH2 cynein, cytoplasmic, heavy polypeptide 2 25 -1.2003 0.0802 0.0183 0,0069 203984..-at CASP9 caupase 91 apoptosip-reiated cysteine protease 26 -1.1973 0.0837 o.01o 0,0071 217844..at CTDSP1 CTD (cerboxy-termina) domain, RNA poly merase 11, polypeptide A) small phosphatase 2 27 -1.1906 0.0914 0.0212 0.0084 213397..x-at RNASE4 ribonuclase, RNaae A family, 4 28 -1.1896 0.0919 0.0214 0.0088 206197..at NME15 non-metastatic cells 0? protein expressed in (nu cl eosi1de-d) phosphate kcinase) 29 -1878 0.0941 0.0221 0.0093 219922..-at LTBP3 Iatent transforming growth factor beta binding protein 3 30 -1.1829 0.1003 0.0247 0.0102 201383..s-at M17S2 membrane Component, chromosome 17, surface marker 2 (ovarian carcinoma antigen CA125) 31 -1.1827 0.1007 0.0240 0.0102 206081-at SLC24A1 solute carrier family 24 (sodium/pota5Bhum/calciUM exchanger), member 32 -1.1709 0.1153 0.0296 0.0129 213266-at 76P Gamma tubulle ring complex protein (76p gene) 33 -1.1707 0.1166 0.0297 0.0129 209189)-at FOS v-foe FBJ urine osteosarcoma viral oncogene homolog 34 -1.1679 0.1201 0,0307- 0.0142 214829..at AASS aminoadipate-semlahyce synthase 35 -1.1633 0.1268 0,0337 0,0153 221123...at ZNF395 nc finger protein 300 36 -1.2625 0,1279 0.0344 0.0155 200B10-a-at CIRBP cold inducible RNA binding protein 37 -1.1612 0.1298 0.0354 0,0157 210365-at RUNX1 runt-related transcrption factor 2(acute myoid leukemia 1; aLml2 oncogene) 38 -1.1602 0.1314 0.0358 0,0158 232842.x-at RANBP2L] RAN binding protein 2-lke 2 39 -1.1596 0.1324 0.0312 0,0158 223364....at SNX1 Sorting nexin 2 40 -1.1586 '0,1339 0,0369 0,016 220911-s-at KIAA1305 KIAA1306 41 -1.1539 0.1444 0.0398 0.0175 201335-s-at ARHIGEF12 Rho guanine nacleotide exchange factor (CDP) 12 42 -1,34909 0,1519 0.0423 0,0190 221276..a-at SYNCOILIN intermediate filament protein eyncolin 43 -L1463 0.1600 0.0452 0.0209 223824-..aat MIR c-nir, cellular modulator of Immune recognition 44 -3.3437 0.1626 0.0401 0.0217 211943-x..at TPTI tumor protein, traietionaliY-controlled 2 46 -1.2409 0.2681 0,1481 0.0223 218552.t FLJ 10948 hypothetical protein rL.2()48 40 -2.2382 0.2743 0.0510 0.0232 220326.--at FLJ10357 hypothetical protein FL20357 47 -L,3254 0.2014 0,0638 0,1296 232869-e:.at TPT tumor protein, reniationally-controlld 2 48 -1.1184 0.2172 0,0722 0,0342 218648..t TORC3 transducer ef regulitd cAMP ruiponnu element binding protein (COBB) 3 41 -1.1271 0.2215 0,0735 0,0348 2125049at STAT5B signal tranidicur ind activator of trinticription 5B 60 -1.)396 0.2233 0.0743 1.0354 21900L-aat C20orfl12 rhlirononic 21) open mding lerne 2 02 -. 1347 0.2277 0.0763 0,0364 212078a N'1F NeuroNFbroinin ) (neUrom atosis, Von YLack litgilean ipiptie r l, WIr dmin0) 02 -21.2213 021) 0.()807 11,0386 22152 .iint A/t Sa ndpt-ami dhde tjn us WO 2006/119593 PCT/BE2006/000051 51 Tub 71 3; l .. criie N, D PD>O PD>3 PD>2 probciiot golle uyrnboi dowcrip~lo 53 -1 .1112 0.23i70 0).08E07 0,01387 202rG2-iiL ](]F131 161u-61, funlilx inuml,L'r 1,9B i4 -1) .302 0.12404 0.0013 0.0394 21'1724-,nL D]XDCI3 Dl)Y do jn conitiing 1 56 -1.10189 0.2450 0).0823 00A08 20r,542..-it SMARCA2 SVW]/SlqF reliited, mliLriy wociotod, ikcLiri do pendent rogulntor of chromatin, ouibfuili]'y meomber 2 86 -1.)08b 0,24B4 0.0027 0.0411 2077b7.n FLJ 23 28 )IypodhoLiolid proLciii PL.121628 57 -1.1085' 0.2464 0,1027 0,011 21846GLnt 'T13C]D) 7 TBC) dominin furmily, member 17 WO 2006/119593 PCT/BE2006/000051 52 Table 4: Univariate and Multivariate analysis of breast cancer prognostic markers (N=417*) Univariate Analysis Multivariate Analysis Hazard ratio Hazard ratio (95%Cl) p (95%CI) p Age (years) 50 vs >50 1.055 (0.556-2.004) 0.869 0.906 (0.416-1.975) 0.8040 Size >2cm vs 2cm 2.694 (1.618-4.485) 0.0001 2.153 (1.235-3.755) 0.0068 Histological grade 1 vs 2 vs 3 2.102 (1.461-3.024) 0.00006 1.446 (0.963-2.171) 0.0754 Estrogen Receptor Rich vs Poor 0.937 (0.671-1.307) 0.937 1.212 (0.667-2.202) 0.5275 Progesterone Receptor Rich vs Poor 0.536 (0.381-0.754) 0.00034 0.755 (0.430-1.328) 0.3300 Genomic Grade High vs Low 2.610 (1.833-3.717) 0.0000001 2.302 (1.241-4.271) 0.0081 * Only patients with complete information in all variables were included in the multivariate analysis (N=208) Based on Cox regression, stratified according to the datasets WO 2006/119593 PCT/BE2006/000051 53 Table 5: Univariate and Multivariate analysis of breast cancer prognostic markers (N=249*) Univariate Analysis Multivariate Analysis Hazard ratio Hazard ratio (95%Cl) p (95%CI) p Age (years) 50 vs >50 0.926 (0.328-2.612) 0.8840 0.807 (0.223-2.916) 0.7440 Size >2cm vs 2cm 2.002 (1.157-3.463) 0.0130 1.712 (0.897-3.268) 0.1030 Histological grade I vs2 vs 3 1.728 (1.128-2.647) 0.0120 1.071 (0.624-1.839) 0.8040 Nodal status Positive vs Negative 1.444 (0.836-2.493) 0.1870 1.053 (0.554-2.001) 0.8760 Estrogen Receptor Rich vs Poor 0.839 (0.512-1.376) 0.4860 0.982 (0.547-1.764) 0,9530 Progesterone Receptor Rich vs Poor 0.485 (0.291-0.806) 0.0050 0.751 (0.409-1.381) 0.3570 Genomic Grade High vs Low 3.119 (1.861-5.228) <0.000001 2.147 (1.042-4.422) 0.0380 * only patients with complete information in all variables were included in the multivariate analysis I Based on Cox regression, stratified according to the datasets WO 2006/119593 PCT/BE2006/000051 54 References 1. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term 5 follow-up. Histopathology 1991;19(5):403-10. 2. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. C. W. Elston & I. 0. Ellis. Histopathology 1991; 10 19; 403-410. Histopathology 2002;41(3A):151. 3. Galea MH, Blamey RW, Elston CE, Ellis 10. The Nottingham Prognostic Index in primary breast cancer. Breast Cancer Res Treat 1992;22(3):207-19. 4. Paik S, Shak S, Tang G, Kim C, Baker J, Cronin M, et 15 al. A multigene assay to predict recurrence of tamoxifen treated, node-negative breast cancer. N Engl J Med 2004;351(27) :2817-26. 5. Robbins P, Pinder S, de Klerk N, Dawkins H, Harvey J, Sterrett G, et al. Histological grading of breast 20 carcinomas: a study of interobserver agreement. Hum Pathol 1995;26 (8) :873-9. 6. Hopton DS, Thorogood J, Clayden AD, MacKinnon D. Observer variation in histological grading of breast cancer. Eur J Surg Oncol 1989;15(l):21-3. 25 7. Theissig F, Kunze KD, Haroske G, Meyer W. Histological grading of breast cancer. Interobserver, reproducibility and prognostic significance. Pathol Res Pract 1990;186(6):732-6. 8. Fitzgibbons PL, Page DL, Weaver D, Thor AD, Allred DC, 30 Clark GM, et al. Prognostic factors in breast cancer. College of American Pathologists Consensus Statement 1999. Arch Pathol Lab Med 2000;124(7):966-78.

WO 2006/119593 PCT/BE2006/000051 55 9. Singletary SE, Allred C, Ashley P, Bassett LW, Berry D, Bland KI, et al. Revision of the American Joint Committee on Cancer staging system for breast cancer. J Clin Oncol 2002;20 (17) :3628-36. 5 10.Perou CM, Sorlie T, Elsen MB, van de Rijn M, Jeffrey SS, Ross CA, et al. Molecular portraits of human breast tumours. Nature 2000;406:747-52. 11.Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, et al. Gene expression patterns of breast 10 carcinomas distinguish tumour subclasses with clinical implications. Proc Natl Acad Science 2001;98 (19) :10869-74. 12.Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, et al. Repeated observation of breast tumour subtypes in independent gene expression data sets. Proc 15 Natl Acad Science 2003;100(14):8418-23. 13.Sotiriou C, Neo SY, McShane LM, Korn EL, Long PM, Jazaeri A, et al. Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci U S A 20 2003;100(18):10393-8. 14.van de Vijver MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002;347(25) :1999-2009. 25 15.Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, Scherf U, et al. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 2003;4(2):249-64. 16.Hedges L, Olsen I. Statistical methods for meta 30 analysis: Academic Press, London; 1985. 17.Korn EL, Troendle J, McShane LM, Simon R. Controlling the number of false discoveries: application to high dimensional genomic data. J Statist Plann Inference 2004;124:379-398.

WO 2006/119593 PCT/BE2006/000051 56 18.Praz V, Jagannathan V, Bucher P. CleanEx: a database of heterogeneous gene expression data based on a consistent gene nomenclature. Nucleic Acids Res 2004;32(Database issue) :D542-7. 5 19.Ma XJ, Salunga R, Tuggle JT, Gaudet J, Enright E, McQuary P, et al. Gene expression profiles of human breast cancer progression. Proc Natl Acad Sci U S A 2003;100 (10) :5974-9. 20.van 't Veer LJ, Dai H, van de Vijver MJ, He YD, Hart 10 AA, Mao M, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415(6871):530-6. 21.Wang Y, Klijn J, Zhang Y, Sieuwerts A, Look MP, Atkins D, et al. Gene-expression profiles to predict distant 15 metastasis of lymph-node-negative primary breast cancer. Lancet 2005;365:671-79. 22.Ein-Dor L, Kela I, Getz G, Givol D, Domany E. Outcome signature genes in breast cancer: is there a unique set? Bioinformatics 2004;21(2):171-8. 20 23.Michiels S, Koscielny S, Hill C. Prediction of cancer outcome with microarrays: a multiple random validation strategy. Lancet 2005;365(9458):488-92. 24.Jenssen TK, Hovig E. Gene-expression profiling in breast cancer. Lancet 2005;365(9460):634-5. 25 25. Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, Deng S, O'Brown P, Borresen-Dale AL, Botstein D. Proc Natl Acad Science. 2003;100:8418-23. 26.Dai H, van't Veer L, Lamb J, He YD, Mao M, Fine BM, Bernards R, van de Vijver M, Deutsch P, Sachs A, Stoughton 30 R, Friend S. Cancer Res. 2005;65:4059-66. 27.Sorlie T. Eur J Cancer. 2004;40:2667-75. 28.Eisen MB, Spellman PT, Brown PO, Botstein D. Proc Natl Acad Sci U S A. 1998;95:14863-8.

WO 2006/119593 PCT/BE2006/000051 57 29.Therneau TM, Grambasch PM. Modeling Survival Data: Extending the Cox Model". In; 2000. 30.Loi S, Piccart MJ, Desmedt C, Haibe-Kains B, Harris A, Tutt A, Sotiriou C. Prediction of early distant relapses on 5 tamoxifen in early-stage breast cancer (BC) . Proc Am Soc Clin Oncol. 2005;23:6s. 31.Clarke R, Liu MC, Bouker KB, Gu Z, Lee RY, Zhu Y, Skaar TC, Gomez B, O'Brien K, Wang Y, Hilakivi-Clarke LA. Oncogene. 2003;22:7316-39. 10 32.Wilcken NR, Prall OW, Musgrove EA, Sutherland RL. Clin Cancer Res. 1997;3:849-54. 33.Baum M, Buzdar A, Cuzick J, Forbes J, Houghton J, Howell A, Sahmoud T. Cancer. 2003;98:1802-10. 34.Boccardo F, Rubagotti A, Puntoni M, Guglielmini P, 15 Porpiglia M, Mesiti M, Rinaldini M, Paladini G, Distante V, Franchi R. American Society of Clinical Oncology. Orlando, Florida abstract (526); 2005. 35.Goss PE, Ingle JN, Martino S. Proc Am Soc Clin Oncol. 2004;22:88s. 20 36.Coombes RC, Hall E, Gibson LJ, Paridaens R, Jassem J, Delozier T, Jones SE, Alvarez I, Bertelli G, Ortmann 0, Coates AS, Bajetta E, Dodwell D, Coleman RE, Fallowfield LJ, Mickiewicz E, Andersen J, Lonning PE, Cocconi G, Stewart A, Stuart N, Snowdon CF, Carpentieri M, Massimini 25 G, Bliss JM. N Engl J Med. 2004;350:1081-92. 37.Jakesz R, Jonat W, Gnant M, Mittlboeck M, Greil R, Tausch C, Hilfrich J, Kwasny W, Menzel C, Samonigg H, Seifert M, Gademann G, Kaufmann M, Wolfgang J. Lancet. 2005;366:455-62. 30 38.Effects of chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005;365:1687 717.

WO 2006/119593 PCT/BE2006/000051 58 39.Winer EP, Hudis C, Burstein HJ, Wolff AC, Pritchard KI, Ingle JN, Chlebowski RT, Gelber R, Edge SB, Gralow J, Cobleigh MA, Mamounas EP, Goldstein LJ, Whelan TJ, Powles TJ, Bryant J, Perkins C, Perotti J, Braun S, Langer AS, 5 Browman GP, Somerfield MR. J Clin Oncol. 2005;23:619-29.

Claims (34)

1. ,A method, comprising the steps of: (a) measuring gene expression in a tumor sample; 5 (b) calculating the gene-expression grade index (GGI) of the tumor sample using the formula: jeG3 .jeG wherein: x is the gene expression level of mRNA, 10 G1 and G 3 are sets of genes up-regulated in HG1 and HG3, respectively, and j refers to a probe or probe set.

2. The method according to claim 1, wherein the tumor sample is from tissue afflicted by breast cancer, 15 colon cancer, lung cancer, prostate cancer, hepatocellular cancer, gastric cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, renal cancer, carcinoma, melanoma, or brain cancer. 20

3. The method according to claim 2, wherein the tumor sample is a breast tumor sample.

4. The method according to claim 3, wherein the breast tumor sample is histological grade HG2.

5. The method according to claim 3, further 25 comprising the step of designating the breast tumor sample as low risk (GG1) or high risk (GG3) based on the GGI index.

6. The method according to any of the claims 1 to 5, further comprising the step of providing a breast 30 cancer treatment regimen for a patient consistent with the WO 2006/119593 PCT/BE2006/000051 60 low risk or high risk designation of the breast tumor sample.

7. The method according to any of the claims 1 to 6, wherein the GGI includes cutoff and scale values 5 chosen so that the mean GGI of the HG1 cases is about -1 and the mean GGI of the HG3 cases is about +1: GGI = scale[ x -Z x,-cutoff jeG 3 feOi

8. The method according to any of the claims 10 1 to 7, wherein the G, gene set comprises at least one gene selected from the genes in Table 3 designated as "Up regulated in grade 1 tumors."

9. The method according to any of the claims 1 to 8, wherein the G 3 gene set comprises at least one gene 15 selected from the genes in Table 3 designated as "Up regulated in grade 3 tumors"

10.The method according to any of the claims 1 to 9, wherein the G, and G 3 gene sets are generated from an estrogen receptor positive population. 20

11.The method according to any of the preceding claims 1 to 10, which further comprises a step of designating a breast tumor sample as different subtypes within ER-positive tumors.

12.The method according to any of the 25 preceding claims 1 to 11, which further comprises a step of designating a tumor sample as a subtype to be submitted to a different treatment than the other subtype.

13.A method, comprising: (a) measuring gene expression in a tumor sample; 30 (b) calculating a relapse score (RS) for the tumor sample using the formula: WO 2006/119593 PCT/BE2006/000051 61 fEG jE}{ N wherein: G is a gene set that is associated with distant recurrence of cancer, 5 Pi is the probe or probe set, i identifies the specific cluster or group of genes, wi is the weight of the cluster i, j is the specific probe set value, xij is the intensity of the probe set j in cluster i, and 10 ni is the number of probe sets in cluster i.

14.The method according to claim 13, further comprising classifying the tumor sample based on the relapse score as low risk or high risk for cancer relapse by a cutoff value.

15 15.The method according to claim 14, wherein the cutoff value for distinguishing low risk from high risk is an RS of from -100 to +100.

16.The method according to claim 14 or 15, wherein the cutoff value for distinguishing low risk from 20 high risk is an RS of from -10 to +10.

17.The method according to any of the claims 13 to 16, wherein relapse is relapse after treatment with a treatment selected from the group consisting of tamoxifen and/or aromatase inhibitor administration, endocrine 25 therapy, chemotherapy or antibody therapy.

18.The method according to claim 17, wherein relapse is relapse after treatment with tamoxifen.

19.The method according to any of the claims 13 to 18, wherein the tumor sample is selected from the 30 group consisting of breast cancer, colon cancer, lung cancer, prostate cancer, hepatocellular cancer, gastric WO 2006/119593 PCT/BE2006/000051 62 cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, renal cancer, carcinoma, melanoma, or brain cancer. 5

20.The method according to claim 19, wherein the tumor sample is a breast tumor sample.

21.The method according to any of the claims 13 to 20, further comprising adjusting a patient's treatment regimen based on the tumor sample's cancer 10 relapse risk status.

22.The method according to claim 21, wherein the step of adjusting the patient's treatment regimen comprises: (a) if the patient is classified as low risk, treating 15 the low risk patient sequentially with tamoxifen and sequential aromatase inhibitors (AIs), or (b) if the patient is classified as high risk, treating the high risk patient with an alternative endocrine treatment other than tamoxifen. 20

23.The method according to claim 22, wherein the patient is classified as high risk and the patient's treatment regimen is adjusted to chemotherapy treatment or specific molecularly targeted anti-cancer therapies.

24.The method according to any of the claims 25 13 to 23, wherein the gene set is generated from an estrogen receptor positive population.

25.The method according to any of.the claims 13 to 24, wherein the gene set comprises at least four of the genes in Tables 1, 2, and 4. 30

26.The method of claim 25, wherein the gene set comprises the genes PGR, HER2, ESR and MKI-67.

27.A computerized system (preferably a device or kit), comprising: WO 2006/119593 PCT/BE2006/000051 63 (a) a bioassay module, preferably a microarray kit or device including real time PCR analysis, configured for detecting gene expression for a tumor sample based on a gene set, preferably made of sequences bound to a solid 5 support surface as an array; and (b) a processor module configured to calculate GGI or RS based on the gene expression and to generate a risk assessment for the breast tumor sample.

28.The system according to claim 27, wherein 10 the tumor sample is from tissue afflicted by breast cancer, colon cancer, lung cancer, prostate cancer, hepatocellular cancer, gastric cancer, pancreatic cancer, cervical cancer, ovarian cancer, liver cancer, bladder cancer, cancer of the urinary tract, thyroid cancer, renal cancer, carcinoma, 15 melanoma, or brain cancer.

29.The system according to claim 27 or 28, wherein the tumor sample is a breast tumor sample.

30.The system according to any of the claims 27 to 29, wherein the bioassay module comprises at least 20 one gene chip (microarray) comprising the gene set.

31.The system according to claim 30, wherein the gene set comprises' at- least one gene selected from the genes in Table 3 designated as "Up-regulated in grade 1 tumors". 25

32.The system according to claim 30, wherein the gene set comprises at least one gene selected from the genes in Table 3 designated as "Up-regulated in grade 3 tumors".

33.The system according to claim 30, wherein 30 the gene set comprises at least one of the genes in Tables 1, 2, and 4.

34. The method of claims 1 to 12, which is combined to an estrogen receptor and/or progesterone receptor gene expression detection.