ES2351327A1 - Method of obtaining useful data for the prediction of clinical response to a treatment of a patient with cancer. (Machine-translation by Google Translate, not legally binding) - Google Patents

Abstract

Method for the prediction of clinical response to a treatment of a patient with cancer. The present invention is within the field of molecular biology and medicine. Specifically, it relates to a method for predicting response to a treatment comprising a platinum derivative and a taxane of a cancer patient, by analyzing the expression profile of the genes cav1, cxcl1, edn3, egf, epor, fgf2, nrp1 and timp2. Preferably, this method allows to predict the response to a treatment comprising carboplatin and paclitaxel from a patient with ovarian cancer. (Machine-translation by Google Translate, not legally binding)

Description

Method for predicting clinical response to A treatment of a cancer patient.

The present invention is within the field of molecular biology and medicine. Specifically, it refers to a method for predicting a response to a treatment comprising a platinum derivative and a taxane of a cancer patient, by analyzing the expression profile of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2 genes , NRP1 and TIMP2 . Preferably, this method allows predicting the response to a treatment comprising carboplatin and paclitaxel of a patient with ovarian cancer.

Prior art

Epithelial ovarian cancer, although relatively rare (around 2000 new cases annually in Spain) is the most lethal tumor of the female genital tract (around 1100 deaths / year in Spain) (National Epidemiology Center, Carlos III Health Institute , November 2002), due not only to its intrinsic aggressiveness, but also to the difficulty of early diagnosis, which means that almost two thirds of the women diagnosed are already in advanced stages of the disease. Long-term survival rates are for patients with stages I and II close to 80%, while patients with advanced disease have survival rates of 10 to 30% (Heintz et al , 2003. Int J Gynaecol Obstet. 83 Suppl 1: 135-
66).

Currently, the treatment of choice for ovarian carcinoma, except for those in stage I, consists of surgery followed by a chemotherapy protocol based on the combination of paclitaxel and platinum derivatives. Tubal carcinoma and primary peritoneal carcinoma have biological, histological characteristics and clinical behavior similar to ovarian carcinoma, so the recommended treatment for these tumors is the same (Barda 2004 Am J Obstet Gynecol; 190 (4): 1039-45). Several retrospective studies have shown that the amount of residual tumor after the first surgery is a factor with prognostic value (Cannistra, 2004. N Engl J Med. 9; 351 (24): 2519-29). In the work carried out by the Gynecologic Oncology Group (GOG) it was shown that a residual tumor smaller than 1 cm in diameter implies greater survival. From that moment on, the most favorable modifiable prognostic factor is the presence of less than 1 cm of residual tumor after the intervention. The median overall survival for patients with a suboptimal surgical intervention is 37 months. On the contrary, patients with optimal surgery have a median progression-free survival close to 22 months and a global survival of 52 months (Cannistra, 2004. N Engl J Med. 9; 351 (24):
2519-29).

Most patients with ovarian cancer require chemotherapy to eradicate the possible disease residual. Although more than 50% of patients with stages III or IV get a complete remission after surgery and chemotherapy, the most of them will suffer a relapse of the disease during the two years later. In addition, most of these patients develops chemoresistance and dies as a result of its disease, so it is necessary to identify those patients They will not respond to treatment.

There are many works that can be found in the literature that analyze the expression of a single marker as a prognostic or predictive response factor. But in recent years we have assisted in the development of high performance gene screening techniques. These techniques applied to tumor analysis have resulted in promising advances in the knowledge of the biological and clinical behavior of tumors. One of the first works in this line is from a Dutch group (Van't Veer et al , 2002. Nature. 31; 415 (6871): 530-6), in which a gene expression profile was identified by microarray with prognostic ability in breast cancer patients. The use of this predictor is approved by the FDA ( Food and Drug Administration ) for patients with early breast cancer diagnosed with positive hormonal receptors and nodes.
affected.

There is another predictor called Oncotype Dx. The main difference with the previous one is that it determines gene expression through the use of microfluidic cards and on samples fixed in formalin and included in paraffin. With the expression results obtained from 21 genes, an estimated risk index of tumor recurrence is constructed (Cronin et al , 2004. Am J Pathol. 164 (1): 35-42; Paik et al 2004. N Engl J Med. 30; 351 (27): 2817-26). The use of this trial is recommended in the review by the American Society of Medical Oncology (ASCO) on tumor markers in breast cancer (Harris et al ., 2007. J Clin Oncol. 20; 25 (33): 5287-312) for patients diagnosed early, with positive hormonal receptors and no affected nodes. This test seeks to incorporate this molecular test into the daily clinic to assess whether genes that are frequently associated with the risk of recurrence among women with early breast cancer can be used to assign patients the most appropriate and effective treatment.

In the specific case of ovarian cancer, we can also find papers that analyze expression profiles using high performance techniques. Most of these studies aim to identify diagnostic markers by comparing cancer samples against normal tissue samples, or established cell line studies (reviewed in Crijns et al , 2006. Int J Gynecol Cancer. 16 Suppl 1: 152-65; Olivier et al , 2006. Eur J Cancer. 42 (17): 2930-8). There are fewer studies on the prediction of response to treatment, although studies with preliminary data in studies with arrays begin to appear in small tumor series (Helleman et al , 2006 Int J Cancer. 15; 118 (8): 1963-71 ; Spentzos et al , 2005. J Clin Oncol. 1; 23 (31): 7911-8; Dressman et al , 2007. J Clin Oncol. 10; 25 (5): 517-25), however, gene profiles described to date to predict the response to a certain ovarian cancer treatment still require validation tests.

Angiogenesis, initially described by Judah Folkman in 1969, is the process by which new ones are formed blood vessels from the preexisting vascular bed. The neovascularization can be induced anywhere in the organism by different stimuli such as hypoxia, ischemia, physical trauma (wounds, fractures) or inflammation, both in situation Physiological as pathological. Carcinogenesis is a process. angiodependent, the formation of new vessels being necessary to uncontrolled growth, invasion and metastasis, associated to tumor development.

Angiogenesis has a relevant role in the specific case of ovarian cancer, due to its own biology. The angiogenesis, controlled by a balance between pro and antiangiogenic agents, particularly the proteins of the family of VEGF, plays a fundamental role in the physiology of the ovary not tumor. On the other hand, we must also bear in mind that the Ovarian cancer is a very vascularized tumor.

There are a large number of studies on the involvement of angiogenic factors in ovarian carcinogenesis (Rasila et al ., 2005. Int J Gynecol Cancer. 15 (5): 710-26). Models in mice indicate that VEGF expression is associated with ascites formation. As in other tumors, VEGF expression is proportional to tumor growth and correlates with vascular microdensity. We can find several works in which it has been shown that VEGF expression in ovarian cancer samples, both of the related subtypes With angiogenesis such as those related to lymphangiogenesis, it correlates with the prognosis and aggressiveness of this disease, and is proportional to vascular microdensity (Nakanishi et al ., 1997. Int J Gynecol Pathol. 16 (3): 256-62; Yoneda et al ., 1998. J Natl Cancer Inst. 18; 90 (6): 447-54; Sood et al ., 2005. Gynecol Oncol. 97 (3):
916-23).

In recent years, new therapies have been developed whose purpose is the inhibition of the formation of new vessels in the tumor environment, as well as the maintenance of the pre-existing vasculature. Undoubtedly, the greatest advance in the field of antiangiogenic therapy occurred when the FDA ( Food and Drug Administration ) approved the use of the first antiangiogenic agent, bevacizumab, for the treatment of human cancer. This humanized antibody specifically binds to VEGF, thus preventing its binding to its receptors. By neutralizing the biological activity of VEGF, tumor vascularization is reduced and therefore its growth is inhibited. Being the first approved anti-angiogenic for use, it is the most available data. Several trials are currently underway in which the efficacy of the addition of bevacizumab to standard chemotherapy in patients with ovarian cancer is evaluated. In the case of a recurrent tumor, the final results of two studies have shown that although the treatment seems effective, with response rates between 15 and 21%, this treatment is associated with a high rate of gastrointestinal perforations as a side effect. In addition to the role as a chemotherapy enhancer, both bevacizumab and other antiangiogenic agents may also have a role as a single agent or as part of a combination regimen that does not include classical chemotherapeutics, and whose clinical trials are also ongoing (Kaye, 2007 J Clin Oncol. 20; 25 (33): 5150-2).

Although the results with antiangiogenic are encouraging, the toxicity issue remains a major pitfall importance and does not seem therefore that in the short term the treatment standard will be modified. It is therefore of great importance development of tools that help the clinician decide when Administer this treatment with a high probability of success.

Description of the invention

The standard treatment of choice in carcinoma Advanced ovary consists of the combination of paclitaxel and a platinum derivative after cytoreductive surgery. This treatment has some associated limitations such as the acute and long-term toxicity. But more important is the fact that, although most patients will respond so initial treatment, only 50% of them will have a Survival greater than 5 years after the end of chemotherapy. There is therefore a need to find a method of analysis that allows to predict the response to treatment, and preferably at the standard treatment of choice (paclitaxel plus carboplatin) in ovarian carcinoma. This way, you can identify, on the one hand, the patients who will benefit from standard treatment And on the other hand, and more importantly, to those patients with resistant tumors, unnecessarily subjected to toxicity associated with treatment without any benefit, and in which also the administration of others is fatally delayed Alternative treatments possibly more effective.

Example 1 of this patent document shows how the authors of the present invention have identified a set of genes related to angiogenesis that predict the likelihood of the response to a chemotherapeutic treatment consisting of paclitaxel plus carboplatin for ovarian carcinoma in samples. fixed in formalin and included in paraffin. This gene expression profile developed by the inventors allows us to predict the response to a treatment comprising a platinum derivative and a taxane. Some types of cancer in which this chemotherapeutic treatment is currently being used, comprising a platinum derivative and a taxane are, for example, but not limited to ovarian cancer, tubal, peritoneum, cervix, endometrium, lung or tumors of origin
unknown.

Therefore, a first aspect of the invention relates to the use of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes to predict the clinical response to a treatment comprising a platinum derivative and a taxane of a Cancer patient

A preferred embodiment of this first aspect of the invention relates to the use of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 to predict the clinical response to a treatment comprising a platinum derivative and a taxane of a cancer patient, where the platinum derivative is selected from the list comprising: carboplatin, oxaliplatin or cisplatin. In a more preferred embodiment, the platinum derivative is carboplatin.

A preferred embodiment of this first aspect of the invention relates to the use of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 to predict the clinical response to a treatment comprising a platinum derivative and a taxane of a cancer patient, where the taxane is selected from the list comprising: paclitaxel or docetaxel. In a more preferred embodiment, the taxane is paclitaxel.

A more preferred embodiment of this first aspect of the invention relates to the use of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 to predict the clinical response to a treatment comprising carboplatin and paclitaxel of a patient with cancer

An even more preferred embodiment of this first aspect of the invention relates to the use of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 to predict the clinical response to a treatment consisting of carboplatin and paclitaxel of A patient with cancer.

In a preferred embodiment of this first aspect of the invention, the cancer is from the list comprising: ovary, tube, peritoneum, cervix, endometrium, lung or tumor of unknown origin. In a more preferred embodiment the cancer is of ovary. In an even more preferred embodiment, ovarian cancer is an ovarian carcinoma, preferably in stage IC, II, III or IV and, more preferably, in stage III or IV.

A second aspect of the invention relates to a method of obtaining useful data to predict the response clinical to a treatment comprising a platinum derivative and a taxane or survival associated with a treatment comprising a derived from platinum and a taxane from a cancer patient, which understands:

to.

obtain an isolated biological sample comprising a patient's cells, and

b.

Detect the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes in the biological sample from step (a).

This method allows, therefore, to obtain data useful for determining the probability of a patient's response to a standard chemotherapeutic treatment, comprising a derivative of platinum and a taxane, and for example, help the clinician in taking decision concerning the administration of another treatment alternative in those patients for whom it was obtained a low probability of response to treatment.

A preferred embodiment of this second aspect of the invention, refers to a method of obtaining useful data to predict the clinical response to a treatment that comprises a platinum derivative and a taxane or survival associated with a treatment comprising a platinum derivative and a taxane from a cancer patient, where the platinum derivative is select from the list comprising: carboplatin, oxaliplatin or cisplatin In a more preferred embodiment, the derivative of Platinum is carboplatin.

A preferred embodiment of this second aspect of the invention, refers to a method of obtaining useful data to predict the clinical response to a treatment that comprises a platinum derivative and a taxane or survival associated with a treatment comprising a platinum derivative and a taxane from a cancer patient, where the taxane is selected from the list that includes: paclitaxel or docetaxel. In one embodiment more preferred, the taxane is paclitaxel.

A more preferred embodiment of this second aspect of the invention, refers to a method of obtaining useful data to predict the clinical response to a treatment that comprises carboplatin and paclitaxel or the survival associated with a treatment comprising carboplatin and paclitaxel of a patient with cancer

A more preferred embodiment of this second aspect of the invention, refers to a method of obtaining useful data to predict the clinical response to a treatment that consists of carboplatin and paclitaxel or the survival associated with a treatment consisting of carboplatin and paclitaxel of a Cancer patient

In a preferred embodiment of this second aspect of the invention, the cancer is from the list comprising: ovary, tube, peritoneum, cervix, endometrium, lung or tumor of unknown origin. In a more preferred embodiment the cancer is of ovary. In an even more preferred embodiment, ovarian cancer is an ovarian carcinoma, preferably in stage IC, II, III or IV and, more preferably, in stage III or IV.

       \ newpage
    

A third aspect of the invention relates to a method to predict the clinical response to a treatment that it comprises a platinum derivative and a taxane of a patient with cancer comprising:

to.

obtain an isolated biological sample which comprises cells of a patient,

b.

Detect the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes in the biological sample from step (a), and

C.

assign the patient a probability of clinical response to treatment, where said clinical response depends on the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes in the sample obtained in (a), in relation to the amount of expression detected for said genes in a reference patient population of known clinical response.

This method allows, therefore, to determine the probability of a patient's response to a treatment chemotherapeutic comprising a platinum derivative and a taxane, and for example, to help the clinician in the decision making regarding the administration of an alternative treatment in those patients with a low probability of response to this treatment Chemotherapeutic

Steps (b) and / or (c) of the methods described previously they can be totally or partially automated, by example, by means of a robotic sensor device for the detection of the amount of product of gene expression in step (b) or the computerized comparison in step (c). In addition to the steps specified above may comprise other steps additional, for example related to the pre-treatment of the sample or evaluation of Results obtained through these methods. For example, the method to predict the clinical response to a treatment that comprises a derived from platinum and a taxane from a cancer patient, you can include an additional step consisting of generating a report on the results of the patient's gene profile, which includes the probability of response to the treatment of said patient diagnosed with cancer

A preferred embodiment of this third aspect of the invention relates to the amount of expression product of each gene is normalized relative to the amount of expression product of a series of reference genes. For example, the reference genes can be, but not limited to, 18S, ACTB, B2M, GAPDH, GUSB, HMBS, HPRT1, IPO8, PGK1, PPIA, RPLP0, TBP, TFRC and UBC . In a preferred embodiment of this third aspect of the invention, the amount of the expression product of each gene is normalized relative to the amount of the expression product of at least one reference gene that is selected from the list that It comprises: 18S, ACTB, B2M, GAPDH, GUSB, HMBS, HPRT1, IPO8, PGK1, PPIA, RPLP0, TBP, TFRC and UBC. In a more preferred embodiment of this third aspect of the invention, the amount of the expression product of each gene is normalized relative to the amount of the expression product of the reference genes 18S, ACTB, B2M, GAPDH, GUSB, HMBS, HPRT1, IPO8, PGK1, PPIA, RPLP0, TBP, TFRC and UBC. In another preferred embodiment of this aspect of the invention, the amount of the expression product of each gene can be normalized relative to the average of the amount of the expression product of the chosen genes or their products.
expression.

In example 1 of this patent document, a model generated from the data obtained from the analysis of the gene expression of samples from a reference patient population of known clinical response is described. Based on patients with a positive response and no response to treatment, a specific weight has been assigned to each of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes , based on which the predictor, whose final result is a probability of response to treatment in said model. Thus, using a method called Stepwise-AIC-Best Subsets approach , a combination of genes whose expression is related to a probability value of clinical response to treatment has been obtained. This value can be analyzed continuously, or by establishing cut-off points according to the established analysis, as can be seen in Table 5 of Example 1 of the patent.

In a preferred embodiment of this third aspect of the invention, the probability of clinical response is Assign using the formula:

one

where Z = independent term + \ sumPE_ {gen} x Normalized expression value_ {gen}; the independent term being a value between 2.5228 and 4.8228 and the specific weight of each gene (PE_) a value between -2.6332 and -0.3332 for PE_ {CAV1}, between -5 , 2609 and -2,9609 for PE_ {CXCL1}, between -8.3516 and -6.0516 for PE_ {EDN3}, between
-3.4534 and -1.1534 for PE_ {EGF}, between -2.8429 and -0.5429 for PE_ {EPOR}, between 2.0392 and 4.3392 for PE_ {FGF2}, between 2.0149 and 4.3149 for PE NPR1 and between 1.4703 and 3.7703 for PE TIMP2. These values that take the coefficients correspond to 75% of the confidence interval.

In a more preferred embodiment of this third aspect of the invention, the probability of clinical response is assigned by the formula described above, where the independent term is a value between 2.0228 and 5.3228 and the specific weight of each gene ( PE_ {gen}) a value between -3,1332 and 0,1668 for PE_ {CAV1}, between -5,7609 and -2,4609 for PE_ {CXCL1}, between -5,5516 and -8,8516 for PE_ {EDN3}, between -3.9534 and -0.6534 for PE_ {EGF}, between -3.3429 and -0.0429 for PE_ {EPOR}, between 1.5392 and 4.8392 for PE_ {FGF2} , between 1,5149 and 4,8149 for PE_ {NRP1} and between 0.9703 and 4.2703 for PE_ {TIMP2}. These values that take the coefficients correspond to 90% of the confidence interval. In an even more preferred embodiment of this third aspect of the invention, the probability of clinical response is assigned by the formula described above, where the independent term is a value between 0.299 and 7.0466 and the specific weight of each gene (PE_ {gen}) a value between -3.7578 and 0.7914 for PE_ {CAV1}, between -7.3895 and -0.8323 for PE_ {CXCL1}, between -13.4446 and -0.9586 for PE_ {EDN3}, between -4.2346 and -0.3722 for PE_ {EGF}, between -3.1511 and -0.2347 for PE_ {EPOR}, between -0.7812 and 7.1596 for PE_ {FGF2} , between -0.5405 and 6.8703 for PE_ {NRP1} and between -0.2571 and 5.4977 for PE_ {TIMP2}. These values that take the coefficients correspond to 95% of the range of
trust.

In an even more preferred embodiment of this third aspect of the invention, the probability of clinical response is assigned by the formula described above, where the independent term is a value between 1.1020 and 6.2428 and the specific weight of each gene (PE_ {gen}) a value between -4.0532 and 1.0868 for PE_ {CAV1}, between -6.6809 and -1.5409 for PE_ {CXCL1}, between -9.7716 and -4.6316 for PE_ {EDN3}, between -4.8734 and -0.2666 for PE_ {EGF}, between -4.2629 and 0.8771 for PE_ {EPOR}, between 0.6192 and 5.7592 for PE_ {FGF2} , between 0.5949 and 5.7349 for PE_ {NRP1} and between 0.0503 and 5.19903 for PE_ {TIMP2}. These values that take the coefficients correspond to 99% of the range of
trust.

The expression "expression value normalized_ {gen} "refers to the quantity of the product of expression of each gene normalized relative to the amount of the product of the expression of a series of reference genes, as previously described elsewhere in this document.

In a preferred embodiment of this third aspect of the invention, a greater probability of clinical response It is indicative of greater relapse-free survival at distance and / or greater overall survival.

A preferred embodiment of this third aspect of the invention, refers to a method for predicting the response clinical to a treatment comprising a platinum derivative and a taxane from a cancer patient, where the platinum derivative is select from the list comprising: carboplatin, oxaliplatin or cisplatin In a more preferred embodiment, the derivative of Platinum is carboplatin.

A preferred embodiment of this third aspect of the invention, refers to a method for predicting the response clinical to a treatment comprising a platinum derivative and a taxane from a cancer patient, where the taxane is selected from the list that includes: paclitaxel or docetaxel. In one embodiment more preferred, the taxane is paclitaxel.

A more preferred embodiment of this third aspect of the invention, refers to a method for predicting the clinical response to a treatment comprising carboplatin and Paclitaxel of a cancer patient.

A more preferred embodiment of this third aspect of the invention, refers to a method for predicting the clinical response to a treatment consisting of carboplatin and Paclitaxel of a cancer patient.

In a preferred embodiment of this third aspect of the invention, the cancer is from the list comprising: ovary, tube, peritoneum, cervix, endometrium, lung or tumor of unknown origin. In a more preferred embodiment the cancer is of ovary. In an even more preferred embodiment, ovarian cancer is an ovarian carcinoma, preferably in stage IC, II, III or IV and, more preferably, in stage III or IV.

A fourth aspect of the present invention is refers to a method to identify if a compound is useful for the cancer treatment, which includes:

to.

get cells derived from a Cancer,

b.

detect the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes in the cells of step (a),

C.

administer the compound to the cells of (a),

d.

detecting the amount of expression product of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 , after administration of the compound according to (c), and

and.

compare the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes , before and after administration of the compound.

       \ newpage
    

Differential expression detection indicates that the compound has activity on one or more profile genes and for therefore, said compound could be a potential agent chemotherapeutic, and would be a candidate to be selected for subsequent analyzes.

In a preferred embodiment of this fourth aspect of the invention, the cells are derived from a cancer of the list comprising: ovary, tube, peritoneum, cervix, endometrium, lung or tumor of unknown origin. In a more preferred embodiment the cancer is ovarian. In an even more preferred embodiment, ovarian cancer is an ovarian carcinoma, preferably, in stage IC, II, III or IV and, more preferably, in stage III or
IV.

The term "predict" or "prediction", as used in the present description, it refers to the probability of a patient responding favorably or unfavorably to a treatment or set of treatments, and to the extent of these responses, or that the patient survives, after surgical removal of a primary tumor and / or the administration of a treatment for a period of time without produce cancer recurrence.

The term "clinical response" to a cancer treatment, as used herein description, refers to its radiological evaluation according to RECIST criteria can also take into account the evaluation to biochemical level (through alterations in biomarker levels) and / or the evolution of the symptomatology.

Often associated with a positive clinical response, there is a greater survival free from distant relapse and / or a greater overall survival. The term "free relapse survival", as used in the present description, refers to the time elapsed from the date of surgery to distance relapse or until the last visit. The term "overall survival", as used in the present description, refers to the time elapsed from the date of surgery until the last visit or until the death of the
patient.

The main treatment of cancer is usually cytoreductive surgery or surgical removal of the tumor. Some patients before surgery receive neoadjuvant treatment in order to reduce the size of the tumor, to enable or facilitate surgery. After surgery, many patients receive adjuvant treatment with the goal of preventing cancer relapse at the primary cancer site or elsewhere. Adjuvant or neoadjuvant treatment may consist, for example, but not limited to radiotherapy, chemotherapy or therapy.
biological

The present invention relates to the use of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 , and to methods based on the analysis of the product of the expression of said genes that are useful for predicting the clinical response to a treatment comprising a platinum derivative and a taxane from a cancer patient. Said treatment may be an adjuvant or neoadjuvant treatment.

In addition to the administration of a derivative of platinum and a taxane, the treatment may include other types of therapy such as surgery, radiotherapy, other type of Chemotherapy or biological therapy. Preferably, the treatment comprises the administration of a platinum derivative and a taxane as adjuvant treatment after surgery cytoreducer

Radiation therapy involves the use of a particle current or high energy waves, such as X-rays, gamma rays, electrons or protons, to destroy the tumor or cancer cells.

Chemotherapy involves the use of compounds or combinations of compounds with the aim of destroying cancer cells Depending on its biological effect, others compounds administered in cancer chemotherapy that they can be administered in addition to the platinum derivative and the taxane, are, for example, but not limited to, alkylating agents, antimetabolites, antimicrotubular agents, inhibitors of topoisomerase or antitumor antibiotics. Some agents alkylating agents are, for example, but not limited to, mechlorethamine, cyclophosphamide, chlorambucil, melphalan, ifosfamide, thiotepa, hexamethylmelamine, busulfan, altretamine, procarbazine, dacarbazine, temozolomide, carmustine, lomustine or streptozocin. Some antimetabolites, are for example, but not limited to, methotrexate, 5-fluoruracil, floxuridine, cytarabine, capecitabine, gemcitabine, 6-mercaptopurine, 6-thioguanine, cladribine, fludarabine, nelarabine or pentostatin Some antimicrotubular agents, such as but not limited, estramustine, vincristine, vinblastine or Vinorelbine. Some topoisomerase inhibitors are, by example, etoposide, etoposide phosphate, teniposide, amsacrine, Irinotecan or Topotecan. Some antitumor antibiotics are for example, but not limited to, doxorubicin, daunorubicin, epirubicin, mitoxantrone, idarubicin, dactinomycin, mitomycin or bleomycin Other chemotherapeutic drugs are, for example, but not limited, hydroxyurea, mitotane, asparaginase, bexarotene, isotretinoin

The term "biological therapy", "biotherapy" or "immunotherapy" as used herein, refers to a treatment used to stop, control or suppress the processes that allow cancer to grow; make cancer cells recognized and destroyed by the immune system more easily; strengthen the destructive power of immune system cells, such as T cells, natural killer cells or macrophages; alter the growth pattern of cancer cells to encourage them to behave like healthy cells; block or reverse the process that causes a normal cell or precancerous cell to become a cancer cell; improve the body's ability to repair or replace normal cells damaged or destroyed by other forms of cancer treatment, such as chemotherapy or radiation; or prevent cancer cells from spreading to other parts of the body. The term "biological therapy" includes, for example, but not limited to interferons (such as, but not limited to, interferon alfa), interleukins (such as, but not limited to, IL-2, IL-6 or IL-12), colony stimulating factors (such as, but not limited to pG-CSF, GM-CSF, IL-11 or erythropoietin), monoclonal antibodies (such as, but not limited to, cetuximab, infliximab, panitumumab, bevacizumab, rituximab or trastuzumab), vaccines, gene therapy or non-specific immunomodulatory agents (such as, but not limited to, the bacillus Calmette-Guérin or
levamisole).

The term "isolated biological sample that comprises patient cells ", as used in the description refers, but is not limited, to tissues and / or fluids biological of a subject, obtained by any method known by an expert in the field that serves this purpose. The biological sample can be a tissue, for example, but without be limited, a biopsy or a fine needle aspirate, or it can be a fluid sample, for example, but not limited, such as blood, plasma, serum, lymph, ascites fluid or urine. Preferably, the cells are tumor cells, and more preferably they come from cytoreduced tumor in surgery, before or after, at administration of an adjuvant or neoadjuvant treatment of Cancer. The biological sample can be, for example, but without be limited, fresh, frozen, fixed or fixed and embedded in paraffin.

The terms "tumor" or "tumor", such and as used in the present description, it refers to cells transformed that show uncontrolled growth. Depending of its possible evolution can be a benign tumor, which it remains in its starting place and does not produce metastasis; or tumor malignant, invasive or metastatic. The term "cancer" or "cancerous" as used in this description, It refers to any malignant tumor.

Examples of cancer include, but are not limited to, ovarian cancer, tubal cancer, peritoneum cancer, uterine cancer, cervical cancer, vaginal cancer, vulvar cancer, breast cancer, colon cancer, prostate cancer, cancer of skin, hepatocellular cancer, lung cancer, esophageal cancer, gastric cancer, pancreatic cancer, genitourinary tract cancer, thyroid cancer, kidney cancer, melanoma, brain cancer, sarcoma, lymphoma or
leukemia.

The present invention relates to the use of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 , and to methods based on the analysis of the product of the expression of said genes that are useful for predicting the clinical response to a treatment comprising a platinum derivative and a taxane from a cancer patient. Some types of cancer in which this chemotherapeutic treatment is currently being used are, for example, but not limited to, ovarian cancer, tubal, peritoneum, cervix, endometrium, lung or tumor of unknown origin, therefore, preferably, cancer It is from the list that includes: ovarian cancer, tube, peritoneum, cervix, endometrium, lung or tumor of unknown origin. More preferably, the cancer is cancer of
ovary.

Ovarian cancer is the accelerated, disordered and uncontrolled proliferation of cells belonging to different tissues of the ovary. Depending on the origin of the cells, ovarian tumors can be, for example, but not limited to, epithelial tumors, germ cell tumors and tumors of the sex-stroma cords. The most common type of malignant tumor derives from the surface epithelium and is ovarian carcinoma. Ovarian carcinomas can be, for example, but not limited to, serous, mucinous, endometrioid, clear or transitional cells. Preferably, the present invention relates to, but is not limited to, a carcinoma of
ovary.

Ovarian carcinoma can be classified into following stages: stage I (which is subdivided into stage IA, IB and IC), stage II (which is subdivided into stage IIA, IIB and IIC), stage III (which is subdivided into stage IIIA, IIIB and IIIC) and stage IV

Usually in stages IA and IB the standard treatment involves performing surgery with the maximum cytoreductive effort. In stages IC, II, III and IV, the standard treatment consists of performing surgery with the maximum cytoreductive effort followed by a treatment of intravenous chemotherapy with a platinum derivative, usually carboplatin and a taxane, usually paclitaxel. Preferably, The present invention relates to, but is not limited to, a carcinoma. ovarian stage IC, II, III or IV and, more preferably, at a stage III or IV ovarian carcinoma.

The term "clinical response" to a treatment in ovarian cancer, as used in the This description refers to its radiological evaluation according to RECIST criteria, and the biochemical evaluation (by measuring the marker serum CA 125) and / or the evolution of symptoms.

For the radiological evaluation, the RECIST criteria are used, considering " pardal response " to the decrease of at least 30% of the sum of the major diameters of the tumor lesions, and " complete response " to the disappearance of all visible tumor lesions. .

       \ newpage
    

For the biochemical response, the Rustin criteria are used, defining the " 50% response " to the decrease of CA 125 by at least 50% with two previous determinations in which it was high and with a subsequent analysis confirming said response; the " complete response " would be the normalization of CA 125 confirmed with a subsequent analysis.

The term "product of expression of genes ... ", as used in the present description, makes reference to transcription and / or translation products (RNA and / or protein) of said genes, or in any form resulting from processing of said transcription or translation products.

The CAV1 or caveolin 1 gene is also called caveolae protein and is also abbreviated as CAV, VIP21 and MSTP085 . It is located on chromosome 7 (7q31.1). Its reference number in the UniGene database of the National Center for Biotechnology Information (NCBI) is Hs.74034. In the context of the present invention, the term "product of CAV1 gene expression" refers to any product of the expression of SEQ ID NO: 1 or any of its variants. SEQ ID NO: 1 corresponds to the prototypical genomic DNA sequence of CAV1 . Preferably, the term " CAV1 gene expression product" refers to the mRNA encoded by SEQ ID NO: 2 or any of its variants, or to the protein encoded by said mRNA or any of its
variants.

The CXCL1 chemokine (CXC motif) ligand 1 gene is also called melanoma growth stimulating activity, alpha and is also abbreviated as FSP , GRO1, GROa, MGS, NAP-3, SCYB1 and MGSA-a . It is located on chromosome 4 (4q21). Its reference number in the NCBI UniGene database is Hs.789. In the context of the present invention, the term "product of the expression of the CXCL1 gene" refers to any product of the expression of SEQ ID NO: 3 or any of its variants. SEQ ID NO: 3 corresponds to the prototypical genomic DNA sequence of CXCL1 . Preferably, the term "expression product of the CXCL1 gene" refers to the mRNA encoded by SEQ ID NO: 4 or any of its variants, or to the protein encoded by said mRNA or any of its variants.

The EDN3 or endothelin 3 gene is also abbreviated as ET3, MGC15067 and MGC61498 . It is located on chromosome 20 (20q13.2-q13.3). Its reference number in the NCBI UniGene database is Hs.1408). In the context of the present invention, the term "product of the expression of the EDN3 gene" refers to any product of the expression of SEQ ID NO: 5 or any of its variants. SEQ ID NO: 5 corresponds to the prototypical sequence of genomic DNA of EDN3 . Preferably, the term " EDN3 gene expression product " refers to the mRNA encoded by SEQ ID NO: 6 or any of its variants, or to the protein encoded by said mRNA or any of its variants.

The EGF or epidermal growth factor (beta-urogastrone) gene is also abbreviated as URG and HOMG4 . It is located on chromosome 4 (4q25). Your reference number at the base of the NCBI UniGene UniGene data is Hs.419815. In the context of the present invention, the term "product of the expression of the EGF gene" refers to any product of the expression of SEQ ID NO: 7 or any of its variants. SEQ ID NO: 7 corresponds to the prototypical sequence of genomic DNA from EGF . Preferably, the term " EGF gene expression product" refers to the mRNA encoded by SEQ ID NO: 8 or any of its variants, or to the protein encoded by said mRNA or any of its variants.

The EPOR or erythropoietin receptor gene is also abbreviated as MGC138358 and is located on chromosome 19 (19p13.3-p13.2). Your reference number at the base of the NCBI UniGene UniGene data is Hs.631624. In the context of the present invention, the term "product of EPOR gene expression" refers to any product of the expression of SEQ ID NO: 9 or any of its variants. SEQ ID NO: 9 corresponds to the prototypical genomic DNA sequence of EPOR . Preferably, the term " EPOR gene expression product" refers to the mRNA encoded by SEQ ID NO: 10 or any of its variants, or to the protein encoded by said mRNA or any of its variants.

The FGF2 or fibroblast growth factor 2 (basic) gene is also abbreviated as BFGF, FGFB, HBGF-2 and is located on chromosome 4 (4q26-q27). Its reference number in the NCBI UniGene database is Hs. 284244). In the context of the present invention, the term "product of the FGF2 gene expression" refers to any product of the expression of SEQ ID NO: 11 or any of its variants. SEQ ID NO: 11 corresponds to the prototypical genomic DNA sequence of FGF2 . Preferably, the term " FGF2 gene expression product" refers to the mRNA encoded by SEQ ID NO: 12 or any of its variants, or to the protein encoded by said mRNA or any of its variants.

The NRP1 or neuropilin 1 gene is also abbreviated as NP1, NRP, BDCA4, CD304, VEGF165R, DKFZp781F1414 and DKFZp686A03134 . It is located on chromosome 10 (10p12). Its reference number in the NCBI UniGene database is Hs.131704). In the context of the present invention, the term "product of NRP1 gene expression " refers to any product of the expression of SEQ ID NO: 13 or any of its variants. SEQ ID NO: 13 corresponds to the prototypical genomic DNA sequence of NRP1 . Preferably, the term " NPR1 gene expression product" refers to the mRNA encoded by SEQ ID NO: 14 or any of its variants, or to the protein encoded by said mRNA or any of its variants.

       \ newpage
    

The TIMP2 or TIMP metallopeptidase inhibitor 2 gene is also known as Tissue inhibitors of metalloproteinases is also abbreviated as CSC-21K and is located on chromosome 17 (17q25). Your reference number at the base of the NCBI UniGene UniGene data is Hs.633514. In the context of the present invention, the term "product of the expression of the TIMP2 gene" refers to any product of the expression of SEQ ID NO: 15 or any of its variants. SEQ ID NO: 15 corresponds to the prototypical genomic DNA sequence of TIMP2 . Preferably, the term " TIMP2 gene expression product " refers to the mRNA encoded by SEQ ID NO: 16 or any of its variants, or to the protein encoded by said mRNA or by any of its
variants.

In the sense used in this description, it understand that a nucleic acid is a "variant" of another when the nucleotide sequence of the first one has a degree of identity with respect to the nucleotide sequence of the second of at least one 30%, advantageously of at least 50%, preferably of, at less, 70%, more preferably, at least 80%, and more preferably of at least 90%.

The term "identity" as it is used in the present description, refers to the proportion of identical nucleotides between two nucleotide sequences, respectively, which are compared. The percent identity existing between two sequences can be easily identified by a subject matter expert, for example, with the help of a program appropriate computer to compare sequences.

The terms "gene expression profile" or "gene expression profile", as used in The present description, refer to the mRNA quantification and / or protein produced by the genes of interest in a sample biological

The detection of the amount of product of the expression of the genes in the sample obtained, as used in the description, refers to the measure of the quantity or concentration, preferably semi-quantitatively or quantitatively. The measure can be carried out directly or indirectly. Direct measurement refers to the measure of the quantity or concentration of the product of the gene expression based on a signal that is obtained directly from the product of the gene expression and that is directly correlated with the number of molecules of the product of the gene. gene expression present in the sample. Said signal - which we can also refer to as an intensity signal - can be obtained, for example, by measuring an intensity value of a chemical or physical property of the gene expression product. The indirect measurement includes the measurement obtained from a secondary component (for example, a component other than the product of gene expression) or a biological measurement system (for example the measurement of cellular responses, ligands, "labels" or reaction products
enzymatic).

The term "quantity", as used in the description, refers to, but is not limited to, the absolute or relative quantity of the expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes. , as well as any other value or parameter related to them or that may be derived from them. Said values or parameters comprise values of signal intensity obtained from any of the physical or chemical properties of the product of the expression of the genes obtained by direct measurement, for example, intensity values of mass spectroscopy or nuclear magnetic resonance. Additionally, said values or parameters include all those obtained by indirect measurement, for example, any of the measurement systems described elsewhere in this document.

Thus, the "quantity detection" of gene expression product can be carried out by any method of determining the quantity of the product of the gene expression known in the state of the art.

In a preferred embodiment, the detection of the gene expression product is performed by determining the level of mRNA derived from its transcription, where the analysis of the mRNA level of C AV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 can be performed, by way of illustration and without limiting the scope of the invention, by amplification by polymerase chain reaction (PCR), back transcription in combination with polymerase chain reaction (RT-PCR), back transcription combination with the ligase chain reaction (RT-LCR), or any other nucleic acid amplification method; DNA chips made with oligonucleotides deposited by any mechanism; DNA microarrays made with oligonucleotides synthesized in situ by photolithography or by any other mechanism; in situ hybridization using specific probes labeled with any method of marking; by electrophoresis gels; by membrane transfer and hybridization with a specific probe; by nuclear magnetic resonance or any other diagnostic imaging technique using paramagnetic nanoparticles or any other type of detectable nanoparticles functionalized with antibodies or by any other means.

MRNA can be analyzed, for example, but without limitation, in samples of fresh tissue, frozen tissue, fixed tissue or tissue fixed and embedded in paraffin. The use of tissue samples fixed and embedded in paraffin presents important advantages over fresh tissue samples or frozen: they are stable at room temperature, easy to store and there is a large archive of clinical samples available together with your clinical information and disease monitoring. By therefore, in a preferred embodiment the one analyzed in is extracted from samples of tissue fixed and embedded in paraffin.

RNA obtained from tissue samples fixed and embedded in paraffin is often very degraded. While studies using microarrays are very sensitive to RNA degradation, the use of RT-PCR, and in particular, quantitative RT-PCR (qRT-PCR) has proven to be a technique that offers better results in the face of degradation of the RNA In addition, expression analysis using microarrays is a complex technique that requires sophisticated equipment that is not available in many laboratories. Therefore, in a preferred embodiment, the detection of the mRNA of the genes is performed by the RT-PCR technique; and in a more preferred embodiment, the detection is performed by the qRT-PCR technique.

In a preferred embodiment, the detection of the gene expression product is performed by determining the level of protein derived from its transcription and translation, where the analysis of the level of CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 can be performed, by way of illustration and without limiting the scope of the invention, by incubation with a specific antibody in an immunoassay. The term "immunoassay", as used herein, refers to any analytical technique that is based on the reaction of conjugation of an antibody with the sample obtained. Examples of immunoassays known in the state of the art are, for example, but not limited to: immunoblot , enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), immunohistochemistry or protein microarrays .

The term "antibody" as used in the present description, it refers to molecules of immunoglobulins and immunologically active portions of molecules of immunoglobulins, that is, molecules that contain a site of antigen binding that specifically binds (immunoreacts) with any of the proteins CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 or TIMP2. Examples of portions of immunoglobulin molecules immunologically active, include F (ab) and fragments F (ab ') 2 that can be generated by treating the antibody with an enzyme such as pepsin. Antibodies can be polyclonal (typically include distinct targeted antibodies against determinants or different epitopes) or monoclonal (directed against a single determinant in the antigen). Antibody monoclonal can be altered biochemically, by manipulation genetics, or it can be synthetic, possibly lacking the antibody in whole or in part, from portions that are not necessary for the recognition of any of the proteins CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 or TIMP2 and being substituted by others that communicate properties to the antibody additional advantages. The antibody can also be recombinant, chimeric, humanized, synthetic or a combination of Any of the above. An "antibody or polypeptide recombinant "(rAC) is an antibody that has been produced in a host cell that has been transformed or transfected with the nucleic acid encoding the polypeptide, or produces the polypeptide as a result of homologous recombination.

The antibodies used to carry out the methods of the present invention can be polyclonal or monoclonal antibodies capable of recognizing any of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 or TIMP2 proteins. Antibodies that recognize the proteins CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 or TIMP2 can be used to carry out the methods of the present invention by, for example, but not limited to, immunoblot , ELISA, RIA, immunhistochemistry or protein microarray .

In a preferred embodiment, the immunoassay is an immunoblot or Western blot . To carry out an immunoblot or Western blot , a protein extract is obtained from an isolated biological sample of a subject and the proteins are separated in a support medium capable of retaining them by electrophoresis. Once the proteins are separated, they are transferred to a different support where they can be detected through the use of specific antibodies that recognize the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 proteins .

In another preferred embodiment, the immunoassay It is an immunhistochemistry (IHQ). Immunohistochemistry techniques allow identification, on tissue or cytological samples of characteristic antigenic determinants. The analysis by immunohistochemistry is performed on tissue cuts, either frozen or included in paraffin, from a sample biological isolated from a subject. These cuts hybridize with a monoclonal or polyclonal antibody that recognizes proteins CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 or TIMP2.

Among the immunohistochemical techniques that can be used to obtain expression profiles is the tissue microarray . Tissue microarray (TMA) is considered today a powerful tool for the massive analysis of the molecular profile of cancer from multiple tissue samples simultaneously, providing the possibility of studying new markers, or existing ones in bulk. In turn, the TMA preserves the original tissue samples that with traditional methods decrease in amount and deteriorate by use
frequent.

Another aspect of the invention relates to a kit comprising the suitable means for carrying out the methods of the invention. In a preferred embodiment, it comprises the means for determining the expression products of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 , in an isolated biological sample.

Said kit may comprise primers, probes, poly or monoclonal antibodies, and all those reagents necessary to use the techniques described above in this document and / or known in the state of the art. The kit can also include, without any limitation, the use of buffers, enzymes, polymerase enzymes, cofactors to obtain optimal activity of these, agents to prevent contamination, etc. On the other hand, the kit can include all the supports and containers necessary for commissioning and optimization. The kit may also contain other molecules, genes, proteins or probes of interest, which serve as positive and negative controls. Preferably, the Kit further comprises the instructions for carrying out the first method of the
invention.

Throughout the description and the claims the word "comprises" and its variants not they intend to exclude other technical characteristics, additives, components or steps. For those skilled in the art, other objects, advantages and features of the invention will be partly detached of the description and in part of the practice of the invention. The following examples and drawings are provided by way of illustration, and are not intended to be limiting of the present invention.

       \ vskip1.000000 \ baselineskip
    

Description of the figures

Figure 1. Shows the ROC curves (acronym for " Receiver Operating Characteristic ", or Receiver Operating Characteristic ) of the model and LOOCV validation for clinical response.

Figure 2. Illustrative graphic (boxplot) in the that profile changes are compared between sensitive patients and no to the treatment, both of the model and after the process of LOOCV validation

Figure 3. Representation of the response model clinical versus probability.

       \ vskip1.000000 \ baselineskip
    

Examples

The following specific example to be provided in this patent document serves to illustrate the nature of the present invention. These examples are included For illustrative purposes only and should not be construed as limitations to the invention claimed herein. Therefore, the Examples described below illustrate the invention without limiting the field of application of it.

       \ vskip1.000000 \ baselineskip
    

Example one

Profile identification associated with response Sample Population Selection

The study was conducted on 63 patients diagnosed with ovarian carcinoma, previously untreated and operated at La Paz University Hospital between 1986 and 2003. All samples are from primary tumors and classified according to Federation criteria. International Gynecology and Obstetrics (FIGO). The samples, fixed in formalin and included in paraffin, were selected based on a series of clinical and pathological criteria:

- Diagnosis of stage ovarian carcinoma III or IV

- Homogeneous treatment consisting of surgery cytoreducer and adjuvant chemotherapy consisting of a regimen of combination of carboplatin and paclitaxel.

- Piece in good condition.

- More than 70% of tumor cellularity in the sample.

Of all the patients included in the study the necessary clinical data were collected, among which it included overall survival, disease-free survival or progression-free time, as well as response data radiological treatment.

       \ vskip1.000000 \ baselineskip
    

MRNA extraction, cDNA synthesis and amplification by RT-PCR

The mRNA was extracted from each of the samples, using 7 \ mum sections and following the protocol of the kit extraction Masterpure RNA Purification Kit (EPICENTRE Biotechnologies, Madison, WI, USA). The mRNA was extracted from each of the selected samples and the concentration and degradation status of said material before proceeding with the study. 1 µg of each sample was used for cDNA synthesis, performed according to the standard kit protocol High Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA).

This material was used for the reaction of amplification, carried out through the card platform Applied Biosystems TaqMan Low Density Arrays microfluidics (TLDA), on an ABI PRISM 7900 HT Sequence Detection sequencer System (Applied Biosystems), according to the recommendations of the distributor, with a configuration that allows the detection of expression of 96 genes (Table 1) in duplicate in two samples of Simultaneously on the same card.

       \ vskip1.000000 \ baselineskip
    

TABLE 1 Genes whose expression has been measured for response profile identification using Taqman probes included in the microfluidic card. Bold includes the reference genes that were used for the study

2

       \ newpage
    

       \ global \ parskip0.930000 \ baselineskip
    

3

Pre-analysis

The raw data of the gene expression levels, obtained by means of SDS 2.2 software (Applied Biosystems) were normalized using the GeNorm program. This program allowed the selection of 5 of them (from a total of 14 reference genes) as the ideal set for the normalization of cases, based on the stability throughout the sample. The normalization process was carried out following the instructions of the Software (Vandesompele et al , 2002. Genome Biol. 2002 Jun 18; 3 (7): RESEARCH0034), and calculating a normalization factor based on the geometric mean of the selected reference genes .

The genes used as a basis for normalization were not considered as candidate genes to be included in the risk score .

Selection of the model to define the risk score and its overall significance

To find the risk index for response using the expression levels of 82 genes as variables, we used a multivariate logistic regression model that is given by the equation:

4

Where in this case \ hat {p} is the estimated treatment response model probability and \ beta_ {i} are estimated for each covariate X {i} that includes the expression of 82 genes, parameters i = 1 , .. p. In what follows we will call Z the linear part of this equation, that is, Z = \ sum ^ {p} _ {i = 0} \ beta_ {i} X {i}.

The selection of candidate variables for become part of the risk index, were selected through a multi-stage algorithm:

In the first stage, the number was selected approximate variables that should enter, using a model of stepwise logistic regression with broad criteria of input (P to enter = 0.995) and output (P = 0.999). In each step of model Akaike information criterion value was observed (AIC). The function that represents the number of steps against the AIC function, has a minimum before reaching the saturated model. As the first model selected, the one constituted by the variables included up to that minimum value of the AIC. In the case of clinical response, the minimum was reached at step 8 so the Model consisted of 8 variables.

The second stage consisted of selecting others additional candidate models, using "the best subset" as statistical criteria for selecting variables among those who had the same number of variables as the number selected in the First stage or one more variable or one less variable. In all them, the AIC statistic was revalued, corrected according to the number of cases and variables.

Finally, among those who had a value similar, that model was taken that in the opinion of the researchers it had a greater biological plausibility, which turned out to be that of minimum AIC value. Table 2 shows the data of global significance of the model.

TABLE 2 Response Prediction Model clinic

5

The model is made up of genes indicated in Table 3.

TABLE 3 Genes that integrate the response model clinic

6

       \ newpage
    

       \ global \ parskip1.000000 \ baselineskip
    

So that the probability associated with the Clinical response is calculated according to the specific weight of the Expression of the selected genes as follows:

First, the term Z is calculated previously defined, based on the coefficients estimated in the regression model and the expression of the model genes according to the equation (1), this is:

Z = Term \ independent \ + \ \ sum (PE_ {gen} \ x \ Value \ of \ expression \ normalized_ {gen})

Based on this term, the probability of response to treatment, according to the formula:

       \ vskip1.000000 \ baselineskip
    

7

       \ vskip1.000000 \ baselineskip
    

Thus, the quantitative probability value is proportional to the value of the variables included in the predictor .

Discriminant capacity of the model and Validation of the risk score

The discriminant capacity of the model was assessed chosen by the index "area under the ROC curve" (Figure 1 and table 4). This index on the probability of response was estimated given by the model for each patient.

Since the selection criterion used was the AIC statistic, it was not considered necessary to perform a validation of the model construction process since there is a bibliography that indicates it, compared to cross-validation or bootstrapping methods (You et al , 2008 Clin Biochem. 41 (10-11): 785-95).

The discriminant capacity of each model was validated using the LOOCV cross-validation method ( Leave One Out Cross Validation ) for ROC curves (Simón et al , 2006. Pharmacogenomics J. 6 (3): 166-73; Varma et al , 2006 BMC Bioinformatics 23; 7: 91) as shown in Figure 1 and Table 4.

       \ vskip1.000000 \ baselineskip
    

TABLE 4 LOOCV cross validation for response clinic

8

       \ vskip1.000000 \ baselineskip
    

In figure 2 the difference in the score values in the responding and non-responding patients is represented by a box diagram, the two response groups being perfectly differentiated based on the predictor. Figure 3 shows the probability associated with the response to treatment, based on the values obtained for the risk score . As can be seen, high score values are associated with a higher probability of response. Although what is proposed in this study is the analysis of the score continuously, to produce less bias, there is the possibility of assigning cuts to classify in risk populations. These cuts can be given based on the discriminant capacity of the model, which as indicated above has been analyzed using ROC curves, and based on the sensitivity and specificity values generated from said analysis and which are included in Table 5.

TABLE 5 Sensitivity and specificity of ROC curves generated for the clinical response predictor, and after passing the LOOCV validation

9

10

TABLE 5 (continued)

eleven

Assessment of the risk score against clinical variables

The model is significantly associated with the response (p = 0.003, RH (95% CI) 2.72 (1.39-5.29). The effect of the same was not evaluated in the presence or adjusted by others clinical factors since none of them is associated with the clinical response using a logistic regression model multivariate Below is the graph in which observe how high values in the model are associated with a high probability of response.

Since the answer is important in the both disease-free survival (or time until failure of treatment) as overall survival, the variables were analyzed clinics and if the inclusion of the response model added something more to Survival prediction. This was studied by a multivariate Cox regression model. The data is attached in Table 6. The only factors with statistical significance are the own predictor and optimal surgery, which continue to maintain the meaning in multivariate analysis.

The following have been used for the analysis statistical programs:

SAS version 9.1.3,

SPSS 9

R version 2.2 with the software package version 2.0-12.

TABLE 6 Logistic Regression of the clinical response and the score and clinical variables

12

       \ global \ parskip0.000000 \ baselineskip
    

<110> FOUNDATION FOR RESEARCH BIOMEDICAL OF LA PAZ UNIVERSITY HOSPITAL.

\hskip1cm

Mendiola Sabio, Marta

 \ hskip1cm 

Mendiola Sabio, Marta

\hskip1cm

Redondo Sánchez, Andrés

 \ hskip1cm 

Redondo Sánchez, Andrés

\hskip1cm

Hardisson Hernáez, David

 \ hskip1cm 

Hardisson Hernáez, David

\hskip1cm

Barriuso Feijoo, Jorge

 \ hskip1cm 

Barriuso Feijoo, Jorge

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<120> Method of obtaining useful data for the prediction of clinical response to a treatment of a cancer patient

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<130> 1660.3

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<160> 16

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<170> PatentIn version 3.5

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<210> 1

         \ vskip0.400000 \ baselineskip
      

<211> 37595

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 1

         \ hskip0,8cm
      

14

         \ hskip0,8cm
      

fifteen

         \ hskip0,8cm
      

16

         \ hskip0,8cm
      

17

         \ hskip0,8cm
      

18

         \ hskip0,8cm
      

19

         \ hskip0,8cm
      

twenty

         \ hskip0,8cm
      

twenty-one

         \ hskip0,8cm
      

22

         \ hskip0,8cm
      

2. 3

         \ hskip0,8cm
      

24

         \ hskip0,8cm
      

25

         \ hskip0,8cm
      

26

         \ hskip0,8cm
      

27

         \ hskip0,8cm
      

28

         \ hskip0,8cm
      

29

         \ hskip0,8cm
      

30

         \ hskip0,8cm
      

31

         \ hskip0,8cm
      

32

         \ hskip0,8cm
      

33

         \ vskip0.400000 \ baselineskip
      

<210> 2

         \ vskip0.400000 \ baselineskip
      

<211> 2704

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 2

         \ hskip0,8cm
      

3. 4

         \ hskip0,8cm
      

35

         \ vskip0.400000 \ baselineskip
      

<210> 3

         \ vskip0.400000 \ baselineskip
      

<211> 15477

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 3

         \ hskip0,8cm
      

36

         \ hskip0,8cm
      

37

         \ hskip0,8cm
      

38

         \ hskip0,8cm
      

39

         \ hskip0,8cm
      

40

         \ hskip0,8cm
      

41

         \ hskip0,8cm
      

42

         \ hskip0,8cm
      

43

         \ hskip0,8cm
      

44

         \ vskip0.400000 \ baselineskip
      

<210> 4

         \ vskip0.400000 \ baselineskip
      

<211> 1103

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 4

         \ hskip0,8cm
      

Four. Five

         \ hskip0,8cm
      

46

         \ vskip0.400000 \ baselineskip
      

<210> 5

         \ vskip0.400000 \ baselineskip
      

<211> 26765

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 5

         \ hskip0,8cm
      

47

         \ hskip0,8cm
      

48

         \ hskip0,8cm
      

49

         \ hskip0,8cm
      

fifty

         \ hskip0,8cm
      

51

         \ hskip0,8cm
      

52

         \ hskip0,8cm
      

53

         \ hskip0,8cm
      

54

         \ hskip0,8cm
      

55

         \ hskip0,8cm
      

56

         \ hskip0,8cm
      

57

         \ hskip0,8cm
      

58

         \ hskip0,8cm
      

59

         \ hskip0,8cm
      

60

         \ hskip0,8cm
      

61

         \ vskip0.400000 \ baselineskip
      

<210> 6

         \ vskip0.400000 \ baselineskip
      

<211> 2438

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 6

         \ hskip0,8cm
      

62

         \ hskip0,8cm
      

63

         \ vskip0.400000 \ baselineskip
      

<210> 7

         \ vskip0.400000 \ baselineskip
      

<211> 100562

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 7

         \ hskip0,8cm
      

64

         \ hskip0,8cm
      

65

         \ hskip0,8cm
      

66

         \ hskip0,8cm
      

67

         \ hskip0,8cm
      

68

         \ hskip0,8cm
      

69

         \ hskip0,8cm
      

70

         \ hskip0,8cm
      

71

         \ hskip0,8cm
      

72

         \ hskip0,8cm
      

73

         \ hskip0,8cm
      

74

         \ hskip0,8cm
      

75

         \ hskip0,8cm
      

76

         \ hskip0,8cm
      

77

         \ hskip0,8cm
      

78

         \ hskip0,8cm
      

79

         \ hskip0,8cm
      

80

         \ hskip0,8cm
      

81

         \ hskip0,8cm
      

82

         \ hskip0,8cm
      

83

         \ hskip0,8cm
      

84

         \ hskip0,8cm
      

85

         \ hskip0,8cm
      

86

         \ hskip0,8cm
      

87

         \ hskip0,8cm
      

88

         \ hskip0,8cm
      

89

         \ hskip0,8cm
      

90

         \ hskip0,8cm
      

91

         \ hskip0,8cm
      

92

         \ hskip0,8cm
      

93

         \ hskip0,8cm
      

94

         \ hskip0,8cm
      

95

         \ hskip0,8cm
      

96

         \ hskip0,8cm
      

97

         \ hskip0,8cm
      

98

         \ hskip0,8cm
      

99

         \ hskip0,8cm
      

100

         \ hskip0,8cm
      

101

         \ hskip0,8cm
      

102

         \ hskip0,8cm
      

103

         \ hskip0,8cm
      

104

         \ hskip0,8cm
      

105

         \ hskip0,8cm
      

106

         \ hskip0,8cm
      

107

         \ hskip0,8cm
      

108

         \ hskip0,8cm
      

109

         \ hskip0,8cm
      

110

         \ hskip0,8cm
      

111

         \ hskip0,8cm
      

112

         \ hskip0,8cm
      

113

         \ hskip0,8cm
      

114

         \ hskip0,8cm
      

115

         \ vskip0.400000 \ baselineskip
      

<210> 8

         \ vskip0.400000 \ baselineskip
      

<211> 4913

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 8

         \ hskip0,8cm
      

116

         \ hskip0,8cm
      

117

         \ hskip0,8cm
      

118

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<210> 9

         \ vskip0.400000 \ baselineskip
      

<211> 7745

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 9

         \ vskip1.000000 \ baselineskip
      

         \ hskip0,8cm
      

119

         \ hskip0,8cm
      

120

         \ hskip0,8cm
      

121

         \ hskip0,8cm
      

122

         \ hskip1cm
      

123

         \ vskip0.400000 \ baselineskip
      

<210> 10

         \ vskip0.400000 \ baselineskip
      

<211> 1849

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 10

         \ hskip0,8cm
      

124

         \ hskip0,8cm
      

125

         \ vskip0.400000 \ baselineskip
      

<210> 11

         \ vskip0.400000 \ baselineskip
      

<211> 72728

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 11

         \ hskip0,8cm
      

126

         \ hskip0,8cm
      

127

         \ hskip0,8cm
      

128

         \ hskip0,8cm
      

129

         \ hskip0,8cm
      

130

         \ hskip0,8cm
      

131

         \ hskip0,8cm
      

132

         \ hskip0,8cm
      

133

         \ hskip0,8cm
      

134

         \ hskip0,8cm
      

135

         \ hskip0,8cm
      

136

         \ hskip0,8cm
      

137

         \ hskip0,8cm
      

138

         \ hskip0,8cm
      

139

         \ hskip0,8cm
      

140

         \ hskip0,8cm
      

141

         \ hskip0,8cm
      

142

         \ hskip0,8cm
      

143

         \ hskip0,8cm
      

144

         \ hskip0,8cm
      

145

         \ hskip0,8cm
      

146

         \ hskip0,8cm
      

147

         \ hskip0,8cm
      

148

         \ hskip0,8cm
      

149

         \ hskip0,8cm
      

150

         \ hskip0,8cm
      

151

         \ hskip0,8cm
      

152

         \ hskip0,8cm
      

153

         \ hskip0,8cm
      

154

         \ hskip0,8cm
      

155

         \ hskip0,8cm
      

156

         \ hskip0,8cm
      

157

         \ hskip0,8cm
      

158

         \ hskip0,8cm
      

159

         \ hskip0,8cm
      

160

         \ hskip0,8cm
      

161

         \ hskip0,8cm
      

162

         \ vskip0.400000 \ baselineskip
      

<210> 12

         \ vskip0.400000 \ baselineskip
      

<211> 6774

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 12

         \ vskip1.000000 \ baselineskip
      

         \ hskip0,8cm
      

163

         \ hskip0,8cm
      

164

         \ hskip0,8cm
      

165

         \ hskip0,8cm
      

166

         \ vskip0.400000 \ baselineskip
      

<210> 13

         \ vskip0.400000 \ baselineskip
      

<211> 159971

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 13

         \ hskip0,8cm
      

167

         \ hskip0,8cm
      

168

         \ hskip0,8cm
      

169

         \ hskip0,8cm
      

170

         \ hskip0,8cm
      

171

         \ hskip0,8cm
      

172

         \ hskip0,8cm
      

173

         \ hskip0,8cm
      

174

         \ hskip0,8cm
      

175

         \ hskip0,8cm
      

176

         \ hskip0,8cm
      

177

         \ hskip0,8cm
      

178

         \ hskip0,8cm
      

179

         \ hskip0,8cm
      

180

         \ hskip0,8cm
      

181

         \ hskip0,8cm
      

182

         \ hskip0,8cm
      

183

         \ hskip0,8cm
      

184

         \ hskip0,8cm
      

185

         \ hskip0,8cm
      

186

         \ hskip0,8cm
      

187

         \ hskip0,8cm
      

188

         \ hskip0,8cm
      

189

         \ hskip0,8cm
      

190

         \ hskip0,8cm
      

191

         \ hskip0,8cm
      

192

         \ hskip0,8cm
      

193

         \ hskip0,8cm
      

194

         \ hskip0,8cm
      

195

         \ hskip0,8cm
      

196

         \ hskip0,8cm
      

197

         \ hskip0,8cm
      

198

         \ hskip0,8cm
      

199

         \ hskip0,8cm
      

200

         \ hskip0,8cm
      

201

         \ hskip0,8cm
      

202

         \ hskip0,8cm
      

203

         \ hskip0,8cm
      

204

         \ hskip0,8cm
      

205

         \ hskip0,8cm
      

206

         \ hskip0,8cm
      

207

         \ hskip0,8cm
      

208

         \ hskip0,8cm
      

209

         \ hskip0,8cm
      

210

         \ hskip0,8cm
      

211

         \ hskip0,8cm
      

212

         \ hskip0,8cm
      

213

         \ hskip0,8cm
      

214

         \ hskip0,8cm
      

215

         \ hskip0,8cm
      

216

         \ hskip0,8cm
      

217

         \ hskip0,8cm
      

218

         \ hskip0,8cm
      

219

         \ hskip0,8cm
      

220

         \ hskip0,8cm
      

221

         \ hskip0,8cm
      

222

         \ hskip0,8cm
      

223

         \ hskip0,8cm
      

224

         \ hskip0,8cm
      

225

         \ hskip0,8cm
      

226

         \ hskip0,8cm
      

227

         \ hskip0,8cm
      

228

         \ hskip0,8cm
      

229

         \ hskip0,8cm
      

230

         \ hskip0,8cm
      

231

         \ hskip0,8cm
      

232

         \ hskip0,8cm
      

233

         \ hskip0,8cm
      

2. 3. 4

         \ hskip0,8cm
      

235

         \ hskip0,8cm
      

236

         \ hskip0,8cm
      

237

         \ hskip0,8cm
      

238

         \ hskip0,8cm
      

239

         \ hskip0,8cm
      

240

         \ hskip0,8cm
      

241

         \ hskip0,8cm
      

242

         \ hskip0,8cm
      

243

         \ hskip0,8cm
      

244

         \ hskip0,8cm
      

245

         \ hskip0,8cm
      

246

         \ hskip0,8cm
      

247

         \ hskip0,8cm
      

248

         \ vskip0.400000 \ baselineskip
      

<210> 14

         \ vskip0.400000 \ baselineskip
      

<211> 5895

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 14

         \ hskip0,8cm
      

249

         \ hskip0,8cm
      

250

         \ hskip0,8cm
      

251

         \ hskip0,8cm
      

252

         \ vskip0.400000 \ baselineskip
      

<210> 15

         \ vskip0.400000 \ baselineskip
      

<211> 73610

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 15

         \ hskip0,8cm
      

253

         \ hskip0,8cm
      

254

         \ hskip0,8cm
      

255

         \ hskip0,8cm
      

256

         \ hskip0,8cm
      

257

         \ hskip0,8cm
      

258

         \ hskip0,8cm
      

259

         \ hskip0,8cm
      

260

         \ hskip0,8cm
      

261

         \ hskip0,8cm
      

262

         \ hskip0,8cm
      

263

         \ hskip0,8cm
      

264

         \ hskip0,8cm
      

265

         \ hskip0,8cm
      

266

         \ hskip0,8cm
      

267

         \ hskip0,8cm
      

268

         \ hskip0,8cm
      

269

         \ hskip0,8cm
      

270

         \ hskip0,8cm
      

271

         \ hskip0,8cm
      

272

         \ hskip0,8cm
      

273

         \ hskip0,8cm
      

274

         \ hskip0,8cm
      

275

         \ hskip0,8cm
      

276

         \ hskip0,8cm
      

277

         \ hskip0,8cm
      

278

         \ hskip0,8cm
      

279

         \ hskip0,8cm
      

280

         \ hskip0,8cm
      

281

         \ hskip0,8cm
      

282

         \ hskip0,8cm
      

283

         \ hskip0,8cm
      

284

         \ hskip0,8cm
      

285

         \ hskip0,8cm
      

286

         \ hskip0,8cm
      

287

         \ hskip0,8cm
      

288

         \ hskip0,8cm
      

289

         \ hskip0,8cm
      

290

         \ vskip0.400000 \ baselineskip
      

<210> 16

         \ vskip0.400000 \ baselineskip
      

<211> 3670

         \ vskip0.400000 \ baselineskip
      

<212> DNA

         \ vskip0.400000 \ baselineskip
      

<213> Homo sapiens

         \ vskip1.000000 \ baselineskip
      

         \ vskip0.400000 \ baselineskip
      

<400> 16

         \ hskip0,8cm
      

291

         \ hskip0,8cm
      

292

         \ hskip0,8cm
      

293

Claims (22)

1. Use of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes to predict the clinical response to a treatment comprising a platinum derivative and a taxane from a cancer patient. 2. Use of the genes according to claim 1, where the platinum derivative is selected from the list that comprises: carboplatin, oxaliplatin or cisplatin. 3. Use of the genes according to claim 2, where the derivative of platinum is carboplatin. 4. Use of genes according to any of the claims 1 to 3, wherein the taxane is selected from the list which comprises: paclitaxel or docetaxel. 5. Use of the genes according to claim 4, where the taxane is paclitaxel. 6. Use of genes according to any of the claims 1 to 5, wherein the cancer is from the list that comprises: ovary, tube, peritoneum, cervix, endometrium, lung or tumor of unknown origin. 7. Use of the genes according to claim 6, where the cancer is ovarian 8. Method to predict the clinical response to a treatment comprising a platinum derivative and a taxane or survival associated with a treatment comprising a derivative of platinum and a taxane from a cancer patient, who understands:

to.

obtain an isolated biological sample comprising a patient's cells, and

b.

Detect the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes in the biological sample from step (a).

9. Method according to the preceding claim, which It also includes:

C.

assign the patient a probability of clinical response to treatment, where said clinical response depends on the amount of expression of the CAVI, CXCL1, EDN3 genes, EGF, EPOR, FGF2, NRP1 and TIMP2 in the sample obtained in (a), in relation to the amount of expression detected for said genes in a population of clinical response reference patients known.

10. Method according to claim 9, wherein the Probability of clinical response is assigned by the formula: 13 where Z = independent term + \ sumPE_ {gen} x Normalized expression value_ {gen}; Being the independent term a value between 0.299 and 7.0466 and the specific weight of each gene (PE_) a value between -3.7578 and 0.7914 for PE_ {CAV1}, between -7.3895 and -0.8323 for PE_ {CXCL1}, between -13.4446 and -0.9586 for PE_ {EDN3}, between -4.2346 and -0.3722 for PE_ {EGF}, between -3.1511 and -0.2347 for PE_ {EPOR}, between -0.7812 and 7.1596 for PE_ {FGF2}, between -0.5405 and 6.8703 for PE_ {NRP1} and between -0.2571 and 5.4977 for PE_ {TIMP2}. 11. Method according to claim 10, wherein the independent term is a value between 2.5228 and 4.8228 and the specific weight of each gene (PE_) a value between -2.6332 and -0.3332 for PE_ {CAV1}, enter
-5,2609 and -2,9609 for PE_ {CXCL1}, between -8,3516 and -6,0516 for PE_ {EDN3}, between -3,4534 and -1,1534 for PE_ {EGF}, between -2 , 8429 and -0.5429 for PE_ {EPOR}, between 2.0392 and 4.3392 for PE_ {FGF2}, between 2.0149 and 4.3149 for PE_ {NRP1} and between 1.4703 and 3.7703 for PE_ {TIMP2}. 12. Method according to any of the claims 8 to 11, wherein the platinum derivative is selected from the list comprising: carboplatin, oxaliplatin or cisplatin 13. Use of the genes according to claim 12, where the derivative of platinum is carboplatin. 14. Use of genes according to any of the claims 8 to 13, wherein the taxane is selected from the list which comprises: paclitaxel or docetaxel. 15. Use of the genes according to claim 14, where the taxane is paclitaxel.

         \ newpage
      

16. Method according to any of the claims 8 to 15, wherein the biological sample isolated from the passage (a) comprises tumor cells. 17. Method according to any of the claims 8 to 16, wherein the biological sample isolated from the passage (a) is fixed and embedded in paraffin. 18. Method to identify if a compound is useful for the treatment of cancer, which includes:

to.

get cells derived from a Cancer,

b.

detect the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes in the cells of step (a),

C.

administer the compound to the cells of (a),

d.

detecting the amount of expression product of the genes CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 , after administration of the compound according to (c), and

and.

compare the amount of expression product of the CAV1, CXCL1, EDN3, EGF, EPOR, FGF2, NRP1 and TIMP2 genes , before and after administration of the compound.

19. Method according to any of the claims 8 to 18 wherein the gene expression product It is determined by RT-PCR. 20. Method according to any of the claims 8 to 19, wherein the cancer is from the list that comprises: ovary, tube, peritoneum, cervix, endometrium, lung or tumor of unknown origin. 21. Method according to any of the claims 8 to 20, wherein the cancer is ovarian. 22. Kit for carrying out the method according to any of claims 8 to 21 comprising the assays indicated in Table 1 of the description to detect the amount of product of the expression expression of the CAV1, CXCL1, EDN3, EGF genes , EPOR, FGF2, NRP1 and TIMP2 .