WO2008138868A2

WO2008138868A2 - In vitro diagnosis and/or prognosis of cancer by the analysis of gtpase protein expression

Info

Publication number: WO2008138868A2
Application number: PCT/EP2008/055714
Authority: WO
Inventors: Mª Teresa GÓMEZ DEL PULGAR; Ana RAMÍREZ DE MOLINA; Juan Carlos LACAL SANJUÁN
Original assignee: Consejo Superior de Investigaciones Cientificas CSIC; Traslational Cancer Drugs Pharma SL
Current assignee: Consejo Superior de Investigaciones Cientificas CSIC; Traslational Cancer Drugs Pharma SL
Priority date: 2007-05-09
Filing date: 2008-05-08
Publication date: 2008-11-20
Anticipated expiration: 2009-11-09
Also published as: WO2008138868A3

Abstract

The present invention relates to an in vitro method of diagnosis / prognosis of cancer, principally of breast cancer and / or lung cancer by the analysis in a single sample of either the expression profile of the Rnd2/Rho 7 protein, or simultaneously, the expression profile of the Rnd subfamily of proteins, comprised of the following proteins: Rndl / Rho 6, Rnd2 / Rho 7 and Rnd3 / RhoE / Rho8. Other GTPases of other subfamilies can be added to the expression profile of these proteins, such as: Rho, Rac, Cdc42, RhoBTB and Miro. The invention further comprises a test device which includes sequences of nucleic acids capable of hybridizing with sequences belonging to the genes of the different families of GTPases whose expression profile is to be measured. The use of these test devices permits, in addition to diagnosing the cancer, making a prognosis of development of the disease, as well as monitoring the treatments for its cure.

Description

IN VITRO DIAGNOSIS AND/OR PROGNOSIS OF CANCER BY THE ANALYSIS

OF GTPASE PROTEIN EXPRESSION

FIELD OF THE INVENTION

The present invention can be included within several fields such as Genetic Engineering, Biotechnology and Pathology. Specifically, the present invention relates to a method of in vitro diagnosis and/or prognosis of breast cancer and/or lung cancer by the analysis of Rho-family GTPase protein expression.

STATE OF THE ART

Rho-family GTPases belong to the large superfamily of Ras proteins, with which they share a high homology of sequence and structure. There are a total of 22 members which can be divided in turn into six groups in accordance with the sequence homology, the structural motifs and the biological function. These subfamilies are Rho (RhoA, RhoB, RhoC), Rac (Racl, Rac2, Rac3, RhoG), Cdc42 (Cdc42, TClO, TCL, Chp, Wrch-1), Rnd (Rndl, Rnd2, Rnd3/RhoE), RhoBTB (RhoBTBl and RhoBTB2) and Miro (Miro-1 and Miro-2). Three of the members of the family, RhoD, Rif and RhoH/TTF are not grouped in any of the following subgroups (Wennerberg and Der, 2004). The most atypical members are the RhoBTB and Miro subfamilies, since these proteins are considerably larger than the classic GTPases and contain structural domains not present in other Rho GTPases.

This family of proteins mediates a wide spectrum of cell activities among which we can highlight the regulation of proliferation and apoptosis processes, organization of the actin cytoskeleton, control of gene expression and membrane traffic. It is therefore not surprising that the Rho proteins are involved in the tumourigenesis process at several levels (reviewed in (Gomez del Pulgar et al, 2005). Their activity is directly related to cell proliferation, the regulation of the cell cycle, cell adhesion and motility, and the degradation of the basal membrane of the original tissue for the extravasation to the blood stream and its later intravasation in a distant tissue. Similar to the Ras proteins, most Rho GTPases are guanine nucleotide -binding proteins which cycle between an active state, bound to GTP, and an inactive state, bound to GDP. This cycle is regulated by three distant families of proteins, the guanine exchange factors (GEFs), the activating proteins of the intrinsic activity of hydrolysis of GTP to GDP (GTPase activating proteins, GAPs), and the guanine dissociation inhibitors (GDIs). After the loading of GTP, the Rho proteins suffer a formational change which allows them to interact with several effectors which are in charge of propagating the signal to inside the cell.

The deregulation of the expression and the activity of some of the best known Rho GTPases, and of some of their most studied effects, has a very important role in human carcinogenesis (Aronheim et al, 1998; Gomez del Pulgar et al, 2005). The relevance of the rest of the family of Rho proteins, as well as of other effectors, in tumourigenesis has not yet been defined. The present invention centred on the analysis of the expression of the three genes Rnd, Rndl, Rnd2 and RhoE/Rnd3 in tumour tissues. These proteins constitute a subgroup of the Rho family with unusual biochemical properties compared with those of other family members. They lack GTPase activity, due to the low affinity for GDP, and because there are normally 10 times more GTP than GDP in the cells, it seems probable that the Rnd proteins are always bound to GTP, and are considered as constitutively active. Thus, the regulation of this subfamily of proteins could exclusively be controlled by the balance between transcription-translation and degradation.

Unlike most of the Rho proteins that are modified by the addition of the geranylgeranyl group at its C-terminal, the Rnd subfamily is farnesylated and suggests that the interaction of the Rnd proteins with membranes is probably different from other members of the family. The members of the Rnd subfamily have a specific tissue expression pattern. The results of a Northern blot in human tissues (Nobes et al., 1998) indicated that the mRNA expression of these three genes is different according to the tissue analysed. Rndl is largely expressed in the brain, liver and in smaller quantities in other tissues; Rnd2 is largely found in the testicles and Rnd3/RhoE is expressed in practically all the tissues at low levels, although their expression levels significantly vary among different cell types (Chardin, 2006). Therefore, an alteration in the expression of the genes which code for members of the Rnd subfamily may be relevant for cell functionality. In the present invention, a significant difference is observed in the expression of these genes among normal and tumour tissue of the breast and lung. Specifically, the present invention relates to an in vitro method of diagnosis / prognosis of breast cancer and/or lung cancer by the analysis in a single sample of the expression of the Rnd2/Rho 7 protein. A 2nd preferred embodiment of the present invention relates to an in vitro method of diagnosis / prognosis of breast cancer and/or lung cancer by the analysis in a single sample and simultaneously, of the expression profile of the subfamily of Rnd proteins, comprised of the following proteins: Rndl / Rho 6, Rnd2 / Rho 7 and Rnd3 / RhoE / Rho8.

The most relevant documents found in the state of the art are:

1 Dl: Jiang, Wen G. et al, "Prognostic value of Rho GTPases and Rho guanine nucleotide dissociation inhibitors in human breast cancers", Clinical Cancer Res., 15.12.2003, vol. 9, 6432-6440.

2 D2: Kleer, Celina G. et al., "Characterization of Rho C expression in benign and malignant breast disease", The American J. of Pathology, 2002, vol. 160:579-584.

It is considered that document Dl is the closest to the object of the invention as it analyses the expression levels and distribution of Rho family GTPases (Rho-A, B, C and G, Rho-6, 7 and 8) and of guanine dissociation inhibitors (GDIs), (Rho-GDI-beta and Rho- GDI-gamma), in breast cancer and their evaluation with regard to the prognostic value. The distribution and localization of the described substances were determined by immunohistochemical staining of frozen sections. The increased concentrations of Rho- C, G and 6 and reduced expression of Rho-GDI-beta and gamma in breast tumour tissues are correlated with the formation of nodules and metastasis. Particularly, in document Dl, a study is made of mRNA and immunohistochemical levels. The GTPase proteins appear cited with alternative names Rho6, Rho7, Rho8 (Rndl, 2 and 3 respectively). In said document Dl, despite the Rho 6, Rho7, Rho8 proteins being related to breast cancer, it is concluded that Rho 6 is overexpressed in breast cancer. But, with regard to Rho-7 and Rho- 8 document Dl concludes that the level of said proteins is similar in tumour and normal tissues. In comparison, as cited in the description, the present invention reflects a change in the expression level of said Rho-7 and Rho-8, proteins in the tumour tissues, with respect to the normal tissues, the degree of expression of these two markers being what determines the cancer diagnosis or prognosis. Document D2 does not relate to GTPases of the invention object of the present invention but describes in it the identification of Rho C as breast cancer marker. It demonstrates that the overexpression of Rho C induces the malign transformation of immortalized human breast epithelial cells, producing an aggressive phenotype, which is very mobile and invasive. Rho C expression in different IBC breast cancers is produced at late stages, whilst the cancerous cells acquire invasive capacity. In IBC, the most lethal type, Rho C appears at early stages. Therefore, Rho C seems to be a marker of the metastastic potential in breast cancer. Finally, the detection of IBC by the Rho C protein may be a useful tool for identifying small invasive ductal carcinomas, with a great probability for producing metastasis. Furthermore, none of the documents located in the state of the art relate the

GTPases analysed in the present invention with the prognosis / diagnosis of lung cancer.

The objective technical problem resolved by the present invention constitutes the diagnosis of breast cancer and / or lung cancer in a greater percentage of patients than the existing methods, or alternatively to them, as well as a better prognosis of the evolution of said disease, on having new markers of the disease such as the Rnd2/Rho 7 protein, alone or in a combined analysis of the expression profile of the whole subfamily of Rnd GTPase proteins, constituted by Rndl/Rho 6, Rnd2/Rho 7 and Rnd3/RhoE/Rho8.

The document which could possibly be considered as closest to the invention is Dl because it analyses the expression and distribution levels of Rho-family GTPases (Rho-A, B, C and G, Rho-6, 7 and 8) and of guanine dissociation inhibitors (GDIs), (Rho-GDI-beta and Rho- GDI-gamma), in breast cancer and their evaluation in terms of prognostic value. Therefore, said document Dl discloses the fact that Rho 6 is overexpressed in breast cancer. But, with regard to Rho-7 and Rho-8, document Dl says that the level of said proteins is similar in the tumour and normal tissues. In comparison, as previously stated and exemplified in the description, the present invention reflects a change in the expression level of said Rho-7 and Rho-8 proteins in tumour tissues, with respect to normal tissues, the degree of expression of these two markers being what determines the prognosis.

The common part between the invention and Dl is that both diagnose breast cancer as they detect Rndl overexpression. However, Dl does not go beyond this, it being the work of the present invention to make it possible to discover the prognosis or evolution of the disease in accordance with the degree of Rnd2 and Rnd3 expression, which makes it possible to diagnose or make a prognosis of the development of the cancer in a greater number of patients that with the marker described in Dl would not be detected. Likewise, it makes it possible to evaluate the development of the disease in those patients suffering from them and which are subject to treatment, in a more complete and reliable way, as there is a greater number of markers available for said evaluation.

DESCRIPTION OF THE INVENTION

Brief description of the invention

The present invention has the main objective of the analysis of the gene expression levels of Rho-family GTPases (Table 6), specifically of the members of the Rnd subfamily: Rndl, Rnd2 and Rnd3, either specifically for one of its members, the Rnd2 protein, or simultaneously in a same sample of tumour tissue, of all them and its use for the diagnosis and / or prognosis of cancer. Said analysis was particularly developed in two of the currently most relevant human tumours: breast cancer and lung cancer, in order to identify new effective markers in the early diagnosis of cancer, the staging and the prognosis.

Therefore, the present invention relates to an in vitro method of diagnosis / prognosis of breast cancer and/or lung cancer by the analysis in a single sample of the expression of the Rnd2/Rho 7 protein. A 2nd preferred embodiment of the present invention relates to an in vitro method of diagnosis / prognosis of breast cancer and/or lung cancer by the analysis in a single sample and simultaneously, of the expression profile of the subfamily of Rnd proteins, comprised of the following proteins: Rndl / Rho 6, Rnd2 / Rho 7 and Rnd3 / RhoE / Rho8. Furthermore, the present invention also relates to a kit or test device which comprises at least sequences of nucleic acids capable of hybridizing with sequences belonging to the Rnd 2/ Rho 7 gene, specifically, combined with sequences of nucleic acids capable of hybridizing also simultaneously in a single sample with gene sequences of the genes Rndl/Rho6 and/or Rnd 3/Rho 8, or combinations thereof, for the in vitro diagnosis / prognosis of cancer which comprises the analysis in a single sample either of the expression level of the gene Rnd2/Rho 7, or, simultaneously, of the expression profile of said genes or of combinations thereof.

Although an embodiment of the present invention relates to a method of in vitro diagnosis / prognosis of cancer, principally of breast cancer and/or lung cancer by the analysis in a single sample of either the expression profile of the Rnd2/Rho 7 protein, or, simultaneously, of the expression profile of the subfamily of Rnd proteins, comprised of the following proteins: Rndl / Rho 6, Rnd2 / Rho 7 and Rnd3 / RhoE / Rho8, to the expression profile of these proteins it is possible to add the measurement of the expression profile corresponding to other GTPases of other subfamilies, such as: Rho, Rac, Cdc42, RhoBTB and Miro. The invention further comprises a test device which comprises sequences of nucleic acids capable of hybridizing with sequences belonging to the genes of the different families of GTPases whose expression profile one wants to measure. The use of these test devices further permits diagnosing cancer, making a prognosis of the development of the disease, as well as a monitoring of the treatments for its cure, or of the phases of development wherein the disease is found at any time.

Description of the figures

Figure 1. Analysis of Rnd2 expression in breast cancer samples. The y-axis (Y) represents the expression level of Rnd2 and the x-axis (X) represents the different tumour samples with N being the normal samples M each one of the tumour samples (A), tumour size (B), the level of lymph gland invasion (C), the histological grade (D) and the relapse

(E).

A. Value of the Rnd2 expression for each tumour sample compared with the normal tissue. B. Average value of Rnd2 expression in accordance with tumour size.

C. Average value of Rnd2 expression in accordance with lymph gland invasion.

D. Average value of Rnd2 expression in accordance with the histological grade.

E. Average value of Rnd2 expression in accordance with the relapse. For any of these parameters, a tendency to reduction is observed in Rnd2 expression, the loss of the expression with a greater histological grade being statistically significant (p<0.05).

Figure 2. A. The y-axis represents cumulative survival and the x-axis represents overall survival in months, in patients with breast cancer in accordance with Rnd2 expression. B. The y-axis represents cumulative survival and the x-axis represents disease-free survival in months, in patients with breast cancer in accordance with Rnd2 expression. (+) cases wherein Rnd2 expression is equal to or greater than the expression in normal tissue (upper curve).

(-) cases wherein a loss occurs in Rnd2 expression compared to normal tissue (lower curve).

The cases with reduced Rnd2 expression showed less disease-free survival (p<0.046). On analysing the overall survival practically, the association between the loss of Rnd2 expression and less overall survival (p = 0.059) is significant.

Figure 3. Analysis of the expression levels of RhoE/Rnd3 in breast cancer samples.

The y-axis represents the expression level of RhoE/Rnd3 and the x-axis represents the different tumour samples with N being the normal samples and M each one of the tumour samples (A), the age of the patient (B), tumour size (C), the level of lymph gland invasion

(D), the histological grade (E) and the relapse (F).

A. Value of the RhoE/Rnd3 expression for each tumour sample (M) compared with the normal tissue (N). B. Average value of RhoE/Rnd3 expression in accordance with the age. Patients with ages over 65 have a significant decrease in RhoE levels (p = 0.015).

C. Average value of RhoE/Rnd3 expression in accordance with tumour size. The reduction in RhoE expression also correlates to the tumour progression, since it is much more dramatic in larger tumours (p = 0.029).

D. Average value of RhoE/Rnd3 expression in accordance with the level of lymph gland invasion. The loss of the RhoE expression seems to be related to lymph gland invasion (p = 0.093)

E. Average value of RhoE/Rnd3 expression in accordance with the histological grade. The loss of RhoE expression seems to be related to the increase in the histological grade (p = 0.186).

F. Average value of RhoE/Rnd3 expression in accordance with the relapse. Significant reduction in RhoE expression in samples of patients who have had a relapse (p = 0.023). Figure 4.

A. The y-axis represents cumulative survival and the x-axis represents overall survival in months, in patients with breast cancer in accordance with RhoE/Rnd3 expression.

B. The y-axis represents cumulative survival and the x-axis represents disease-free survival in months, in patients with breast cancer in accordance with RhoE/Rnd3 expression.

(+) cases wherein the RhoE/Rnd3 expression is equal to or greater than the expression in normal tissue (upper curve).

(-) cases wherein a loss occurs in RhoE/Rnd3 expression compared to normal tissue (lower curve). The analysis of the overall survival is not significant (p = 0.608), although it improves if the analysis is performed only considering the cases with more reduced levels of RhoE (percentile 25) (p = 0.086) (Fig. A). However, when disease-free survival is analysed in accordance with RhoE expression, statistical significance is not achieved (p =

0.180) (Fig. B, Kaplan-Meier curve). Figure 5. Analysis of the Rndl expression levels in lung cancer samples. The y-axis represents the expression level of Rndl in absolute quantity (AQ) and the x-axis represents the different tumour samples with N being the normal samples L each one of the tumour samples (A), tumour size (B), level of lymph gland invasion (C) and the relapse (D). A. Value of the Rndl expression for each tumour sample (L) compared with the normal tissue (N).

B. Average value of Rndl expression in accordance with tumour size.

C. Average value of Rndl expression in accordance with the level of lymph gland invasion. D. Average value of Rndl expression in accordance with the relapse due to appearance ofmetastasis.

The distribution of Rndl expression is analysed comparing the values of the averages in accordance with the clinical-pathological parameter. The relation between the mRNA levels of the Rndl gene and the clinical-pathological parameters were based on the non- parametric Mann Whitney U test. It is observed that Rndl expression is not significantly associated to any of the parameters under study.

Figure 6

A. The y-axis represents cumulative survival and the x-axis represents overall survival in months, in patients with lung cancer in accordance with Rndl expression.

B. The y-axis represents cumulative survival and the x-axis represents disease-free survival in months, in patients with lung cancer in accordance with Rndl expression.

(+) cases wherein the Rndl expression is greater than the value of the average (lower curve).

(-) cases wherein a loss occurs of Rndl expression compared with the value of the average

(upper curve).

Analysis of the capacity of the Rndl mRNA levels for discriminating between patients which relapse and those that do not relapse using the value of the median, 0.99, as cut-off point (with a sensitivity of 82% and specificity of 58% according to analysis of the ROC curve). This value represents a reduction in the expression of 1.6 times below expression in the normal tissue. The average of disease-free survival is of 30.8 months for the patients with expression above the median and of 51.8 months if the levels are lower. Although the difference does not come to be statistically significant (p = 0.069), this result indicates that those patients with expression levels 1.6 times below the value of expression in normal tissue relapse less. When the overall survival is analysed in accordance with Rndl expression, the difference is significant (p<0.05), the average of overall survival being 31.5 months in patients with expression above the cut-off point and 56.4 months when the expression levels are below that value.

Figure 7. Analysis of Rnd2 expression in lung cancer samples. The y-axis (y) represents the expression level of Rnd2 in absolute quantity (AQ) and the x-axis (X) represents the different tumour samples with N being the normal samples and L each one of the tumour samples (A), tumour size (B), the level of lymph gland invasion (C) and the relapse (D).

A. Value of Rnd2 expression for each tumour sample compared with the normal tissue.

B. Average value of Rnd2 expression in accordance with tumour size.

C. Average value of Rnd2 expression in accordance with lymph gland invasion.

D. Average value of Rnd2 expression in accordance with the relapse.

Figure 8.

A. The y-axis represents cumulative survival and the x-axis represents overall survival in months, in patients with lung cancer in accordance with Rnd2 expression.

B. The y-axis represents cumulative survival and the x-axis represents disease-free survival in months, in patients with lung cancer in accordance with Rnd2 expression.

To discriminate between patients that relapse and those that do not relapse in accordance with Rnd2 expression we use as cut-off point the value 10.5 (which is a point with the same sensitivity as the value of the median, of 73%, but with greater specificity, 66%, according to analysis of the ROC curve). (+) cases wherein Rnd2 expression is greater than the cut-off point (lower curve).

(-) cases wherein a loss occurs of Rnd2 expression below the cut-off point (upper curve).

Analysis of disease-free survival. Significant indication (p = 0.037), that patients with an increase in Rnd2 expression very closely relapse over time and their survival is less than that of the those which have lower levels. Furthermore, it is observed that the overall survival is less (average of 36.4 months compared with an average of 53 months) in those patients with Rnd2 overexpression with respect to normal tissue. The difference is significant (p=0.020). Thus, these results suggest that Rnd2 overexpression in the lung is a factor of worse prognosis, since a significant reduction is observed in disease-free survival and in the overall survival of patients.

Figure 9. Analysis of the expression levels of RhoE/Rnd3 in lung cancer samples. The y-axis represents the expression level of RhoE/Rnd3 and the x-axis represents the different tumour samples with N being the normal samples and L each one of the tumour samples (A), tumour size (B), the level of lymph gland invasion (C) and the relapse (D).

A. Value of RhoE/Rnd3 expression for each tumour sample (L) compared with normal tissue (N).

B. Average value of RhoE/Rnd3 expression in accordance with tumour size.

C. Average value of RhoE/Rnd3 expression in accordance with the level of lymph gland invasion.

D. Average value of RhoE/Rnd3 expression in accordance with the relapse.

It is observed that most of the tumour samples have an expression level greater than the expression in normal tissue (the expression average is 7.48, median 5.43, the normal tissue expresses 2.12, the pool of normal tissues 1.67). When the averages are compared in accordance with the clinical-pathological parameters, no significant differences are found.

Detailed description of the invention

Specifically, the present invention relates to an in vitro method of diagnosis / prognosis of breast cancer and/or lung cancer by the analysis in a single sample of the expression of the Rnd2/Rho 7 protein. A 2nd preferred embodiment of the present invention relates to an in vitro method of diagnosis / prognosis of breast cancer and/or lung cancer by the analysis in a single sample and, simultaneously, of the expression profile of the subfamily of Rnd proteins, comprised of the following proteins: Rndl / Rho 6, Rnd2 / Rho 7 and Rnd3 / RhoE / Rho8. The study was developed in two of the currently most relevant types of cancer: breast cancer and lung cancer, with the aim of identifying new effective markers in the early diagnosis of these tumours, the staging and the prognosis.

METHODOLOGY The analysis of the expression levels of Rho family genes, as well as of its main effectors, is carried out by a quantitative PCR study in real time. For this type of analysis, the extraction of total RNA samples from the tissue samples is necessary, as well as their processing for the PCR quantification in real time of the messenger RNA levels (mRNA) of each gene.

CLINICAL SAMPLES

The breast and lung tissue samples belong to patients who have undergone surgery in the Hospital de La Paz, Madrid. None of the patients received adjuvant therapy prior to surgery. All the samples were collected at the time of resection and were immediately frozen in liquid nitrogen. The ethical committee of the Hospital La Paz approved this study.

The breast cancer samples are of infiltrative ductal histology type and correspond to patients with ages between 27 and 80 years of age (average 57 years of age). The median of the monitoring time was 70 months (range: 25-126). At the time of analysis, 12 of the patients had suffered relapse due to the disease, whilst the rest had not relapsed. The pathological parameters of the piece such as the degree of differentiation, the size (pT) and the data on regional lymph gland invasion (pN) are set down in Table 1, as well as the clinical data regarding relapse of the disease.

In the lung cancer study, 53 samples of patients have been processed with ages between 43 and 82 years of age (average 64 years of age). The clinical-pathological parameters of the samples are set down in Table 2. Among the samples processed are the three most frequent types of non-microcytic carcinomas which occur in clinical practice: squamous cell carcinoma or epidermoid carcinoma, adenocarcinoma and large cell carcinoma. The monitoring period ranges from 2 months to 61 months, with a monitoring average of 18 months.

TOTAL RNA EXTRACTION

The total RNA of the tissue samples was extracted using Trizol Reagent (Invitrogen) and then passing the samples through the RNeasy Mini kit (Qiagen), according to the manufacturer's instructions. The RNeasy Mini kit from Qiagen enables the extraction from small quantities of material. It is based on the selective bonding properties of the RNA to a silica-gel membrane and the contaminants can be eliminated by washing. The integrity of each one of the RNA samples was evaluated by electrophoresis in agarose gel and the samples were quantified with the Nanodrop spectrophotometer for the later retrotranscription reaction.

REVERSE TRANSCRIPTION

The synthesis of the complementary DNA (cDNA) of each RNA sample was carried out from 1 g of total RNA with the High-Capacity cDNA Archive Kit (Applied Biosystems). This kit contains all the necessary components, including a buffer optimized for the reaction, dNTPs, random primers and the reverse transcriptase MultiScribe™ MuLV.

QUANTITATIVE REAL-TIME PCR

In the real-time PCR technique, the amplification products are observed as the PCR cycles elapse. It is based on the detection and quantification of a fluorescent Reporter, whose signal increases in direct proportion to the quantity of PCR product in the reaction.

The data are acquired when the amplification is still in exponential phase. The data collected corresponds to the cycle number wherein the fluorescence exceeds the established threshold, on increasing the emission intensity of the fiuorochrome with respect to the Background noise. This cycle number is called Ct: Threshold Cycle. The Ct is determined in the exponential phase of the reaction and is more reliable than the conventional end point measurements. The Ct is inversely proportionate to the number of copies of the initial target: the greater the concentration the less Ct measured. The threshold set must be the same for all measurements. Low intensity arrays designed by Applied Biosystems, called TaqMan® Low

Density Arrays have been used for the amplification of the cDNA. It is a microfluidic card which contains the specific probes of each gene deposited in each one of the 384 wells, with a configuration selected according to the number of replicas desired of each gene. In the present study, the configuration chosen contained the genes selected in duplicate. The amplification of an endogenous control is used to standardize the quantity of RNA. For the quantification of the expression of each gene, the gene of glyceraldyde-3 -phosphate dehydrogenase (GAPDH), or 18S ribosomal RNA (rRNA) were used as endogenous controls.

The probes deposited in each well of the card are chosen from those available in Applied Biosystems for our genes of interest. The probes are called TaqMan® Gene Expression Assays (Applied Biosystems) and are designed to recognize a region without ambiguities in its sequence, without polymorphisms or repeated sequences. Whenever possible, the tests are designed in the exon-exon connections, which avoid the co- amplification of contaminant genomic DNA. The probes chosen in the present study are specified in Tables 3, 4.

The microfluidic technology uses 8 loading doors, each connected to 48 reaction chambers. This characteristic reduces the number of pipetting steps. 50 nanograms of cDNA are combined with the Taqman Universal Master Mix (Applied Biosystems) per loading door, the card is briefly centrifuged and it is sealed. The Array is taken to 50 ⁰C for 2 min and 95 ⁰C 10 min, followed by 40 cycles at 95 ⁰C for 15 s, and 60 ⁰C 1 min. The amplification data are collected using the SDS 2.1 software from Applied Biosystems, and analysed with the comparative method of relative quantification. The reference sample is a commercial RNA of normal breast tissue for the breast tissue samples. In the lung study, a commercial RNA from normal lung tissue and a pool of normal tissues adjacent to some tumours under study have been used. The commercial RNAs are from Stratagene. To calculate the relative expression of each gene, the 2^{" Ct} method was used

(Applied Biosystems User Bulletin no. 2 (P/N 4303859). The results are expressed as base-

10 logarithms of RQ (Relative Quantity), which is the relative quantity of the gene of interest in a tumour sample with respect to normal tissue, once normalized against the endogenous control gene.

The main advantage of this methodology is that the quantity of material required is minimal, which is very interesting when working with clinical samples, of which there is normally very little material available. Furthermore, it is a simple, quick process.

STATISTICAL ANALYSIS

The relation between the mRNA expression levels of Rho-family GTPase genes and the clinical-pathological factors was analysed statistically. The continuous variables with abnormal distribution were compared using the Mann Whitney and Kruskal-Wallis U-test. To analyse the continuous variables with normal distribution, the t-test was used. The survival curves were made according to the Kaplan-Meier method and the significance was evaluated with the log rank test. The quantification of the linear relation between the two variables was studied by calculating Spearman's non-parametric coefficient of correlation. All calculations were made with the SPSS software, version 13. The differences were considered statistically significant when the value oϊp was lower than 0.05.

RESULTS

Rnd2 mRNA expression levels in breast cancer

The results of this quantitative PCR study were expressed as base- 10 logarithm of RQ (Relative Quantity), which is the relative quantity of the gene of interest in a tumour sample with respect to the normal tissue, once normalized against the endogenous control gene GAPDH. In this study we have use a commercial RNA of normal breast tissue (Stratagene, catalogue no. 540045), due to the difficulties in obtaining RNA from normal breast samples, because of the fat of this type of tissues. From the study, those samples whose error bar involved a difference over 50% of the sample value were eliminated, since it cannot be determined what the state of this sample on comparing it with normal tissue is. After reviewing the data, the results of 47 patients are processed. The expression profile of Rnd2 of these tumour samples is very variable compared with normal tissue (Fig. IA), which led to trying to group the samples in accordance with the clinical-pathological parameters and analysing the average expression value in accordance with tumour size (Fig. IB), the capacity of invading lymph glands (Fig. 1C), the histological grade (Fig. ID) and the relapse of the patient due to the disease (Fig. ID). For any of these parameters, a tendency is observed for a reduction in Rnd2 expression, the loss of expression with a greater histological grade being statistically significant (p<0.05).

Given the tendency observed to the reduction in Rnd2 expression as the disease progresses, the samples were classified in accordance with their expression in cases with expression equal to or greater than normal tissue and cases with a reduced expression, less than normal tissue. Table 5 indicates the relation between the levels of Rnd2 expression with each clinical-pathological factor. The analysis of the data, as observed in Fig. 1 , indicates that the loss of Rnd2 expression in relation to normal tissue is significantly associated with the histological grade (p = 0.004).

Furthermore, the relation between the reduction of Rnd2 expression and the relapse of the patient due to the disease is statistically significant (p = 0.043). 5 out of the 10 samples of patients who have suffered relapse (appearance of metastasis in distant organs: skin, bone, lung or liver) have an expression level lower than normal tissue, suggesting that in these cases, the loss of the Rnd2 expression could favour the development of the disease in later stages. When the survival of the patients is analysed, the cases with reduced Rnd2 expression showed less disease-free survival (Fig. 2B) (p<0.046). On analysing the overall survival practically, the association between the loss of Rnd2 expression and a less overall survival is significant (Fig. 2A) (p = 0.059).

Thus, the results suggest that the loss of Rnd2 expression is associated to a worse prognosis. A relation has been observed with the histological grade, which is recognised as a classic prognosis factor, grade III being associated to a worse prognosis. Furthermore, the decrease in Rnd2 levels seems to favour the process of metastasis, leading to significantly less disease-free survival. These results suggest that Rnd2 could behave as a tumour suppressor gene in breast cancer, which would involve the first evidence of the relation of this GTPase with carcinogenesis. Rnd2 is located in chromosome 17, a region frequently altered in breast and ovarian tumours, since the translocations of said chromosome are the most frequently observed genetic alteration in this type of tumours. The Rnd2 gene is the closest to BRCAl (Smith et al, 1996), a gene whose mutations increase susceptibility to breast and ovarian cancer. Rnd2 is located in opposite orientation to BRCAl. However, there are no data on the effects of the deletion of the gene or translocations in breast cancer (Smith et al., 1996).

RhoE/Rnd3 mRNA expression levels in breast cancer

Those samples from the study with a deviation over 50% of its value have been eliminated, for which reason the study is reduced to 42 samples, which we analyse statistically. Most of the breast tumour tissues analysed show a reduction in the expression levels of RhoE compared with the levels of normal tissue (Fig. 3A). The distribution of the data is normal and the t-test is applied to the comparison of the averages in accordance with the clinical-pathological parameters (Fig. 3). It is observed that patients with ages over 65 years of age show a significant reduction in RhoE levels (p = 0.015) (Fig. 3B). The reduction in RhoE expression is also correlated with the progression of the tumour, since it is much more dramatic in tumours of greater size (p = 0.029) (Fig. 3C). Speaking of what could be called classic prognosis factors, the size is probably the second in importance after the axilliary lymph nodes. The relation between size and prognosis is practically linear. The loss of RhoE expression seems to also be related to lymph gland invasion (p =

0.093) (Fig. 3D), and histological grade (p = 0.186) (Fig. 3E), but it does not come to be significant from a statistical standpoint. What is significant is the reduction in gene expression in patients who have had a relapse (p = 0.023) (Fig. 3F). However, when the disease-free survival is analysed in accordance with RhoE expression the statistical significance was not reached (p = 0.180) (Fig. 4B, Kaplan-Meier Curve). Neither is the overall analysis of the overall survival significant (p = 0.608), although it is if the analysis is made only considering the cases with more reduced levels of RhoE (percentile 25) (p = 0.086) (Fig. 4A).

From the Rnd3 expression profile (Fig. 3A) three samples stand out (76, 109, 116) as they have expression levels 10 times greater than normal tissue, compared with the other samples which largely have lower levels. These patients have not suffered a relapse, nor did they show disease of the regional lymph glands at the time of the biopsy, therefore being cases of early stage.

The results suggest that the loss of RhoE expression is a factor of bad prognosis in breast cancer, since the reduced levels of this gene in comparison with normal tissue is associated to a greater tumour size and to the relapse of the patient.

This reduction or lack of RhoE expression indicates that it behaves as a tumour suppressor gene in breast cancer since RhoE expression may be critical in preventing the transformation of epithelial breast cells. As RhoE antagonizes the RhoA GTPase, which is overexpressed and activated in different human tumours, among those of the breast (reviewed in Gomez del Pulgar et al, 2005), RhoE could counteract the action of RhoA to maintain the cell growth processes and normal differentiation in breast tissue.

Therefore, our results suggest that RhoE/Rnd3 is a tumour suppressor gene and that its deregulation may be a marker of tumour progression, in accordance with its antagonist role of RhoA and as cell cycle inhibitor.

Correlation between the Rnd2 and RhoE/Rnd3 mRNA expression levels in breast cancer

The expression profile of the two genes Rnd, Rnd2 and Rnd3, in the breast tumour samples, suggest that they may have a potential role of tumour suppressors, since their expression significantly decreases with the development of the disease. The results suggest that the loss of both genes seems to be related to a worse prognosis, since a significant relation has been found with some of the factors which mark the prognosis of breast cancer. This led us to consider if there is a correlation in the reduction of its expression. To quantify the degree of linear relation existing between the two variables, we calculate Spearman's Rho coefficient of correlation. The coefficient takes values between -1 and 1 : a value of 1 indicating a positive perfect linear relation and a value of -1 a negative perfect linear relation. The coefficient of correlation between Rnd2 and Rnd3 expression is of 0.006 (p=0.972), i.e. a null linear relation. This result indicates to us that the loss of expression of these genes is independent.

Rndl mRNA expression levels in lung cancer mRNA expression levels of Rnd genes in lung cancer are expressed as the absolute quantity (AQ) of the Rnd gene in a tumour sample once normalized against the 18S ribosomal RNA endogenous control gene. As a consequence, it indicates the expression of mRNA of the gene in a sample of normal commercial tissue (Stratagene, catalogue no. 540019). It also represents mRNA expression in the pool of normal tissues included in the study.

Fig. 5 shows the results of the Rndl mRNA levels in lung cancer samples. The average of Rndl expression is 1.57, and the median 0.99, the quantity of the gene in the normal commercial tissue (Stratagene) being 1.63. (The normal pool of tissues has a quantity of 1.87). We take as reference the expression value of the gene in normal tissue and we observe that most of the tumour samples have lower expression levels (Fig. 5).

The distribution of Rndl expression is analysed comparing the values of the averages in accordance with the clinical-pathological parameter (Fig. 5). The relation between mRNA levels of the Rndl gene and the clinical-pathological parameters were based on the non-parametric Mann Whitney U test (Fig. 5). We observe that Rndl expression is not significantly associated to any of the parameters under study (Fig. 5A: size, 5B: invasion of lymph nodes or 5C: relapse).

The disease-free survival was determined as the interval between the time of diagnosis and the detection of the first relapse. To analyse the capacity of Rndl mRNA levels to discriminate between patients that relapse and those that do not relapse, we use as cut-off point the value of the median, 0.99, (with a sensitivity of 82% and specificity of 58% according to analysis of the ROC curve). This value represents a reduction in expression of 1.6 times below the expression in normal tissue. The average of disease-free survival is 30.8 months for patients with expression levels greater than the median and 51.8 months if the levels are lower (Fig. 6). Although the difference does not come to be statistically significant (p = 0.069), this result indicates that those patients with expression levels 1.6 times below the value of the expression in normal tissue relapse less.

When the overall survival is analysed in accordance with Rndl expression (Fig. 6), the difference is significant (p<0.05), the average overall survival being 31.5 months in patients with expression above the cut-off point and 56.4 months when the expression levels are below that value.

The reduced expression levels of Rndl under (at least 1.6 times) the expression levels in normal tissue indicates to us that it is a factor independent of good progress in non-microcytic cancer, increasing the overall survival of these patients significantly. Our observation constitutes the first evidence of a first involvement of Rndl in the evolution of lung cancer.

Rnd2 mRNA expression levels in lung cancer Fig. 7A shows mRNA expression levels of the Rnd2 gene in lung cancer samples.

The expression average is 20.9 and the median 9.06, the quantity of mRNA in the normal tissue being 5.62. If we compare the expression averages in accordance with the clinical- pathological parameters, there are no significant differences (Fig. 7B, 7C and 7D).

Next, we analyse the survival curves in accordance with Rnd2 expression. To discriminate between patients that relapse and those that do not relapse in accordance with Rnd2 expression, we use as cut-off point the value 10.5 (which is a point with the same sensitivity as the value of the median, of 73%, but with greater specificity, 66%, according to analysis of the ROC curve). The average of disease-free survival is less for patients with expression levels over said cut-off point (36.4 months compared to 49.8 months) (Fig. 8). This difference is statistically significant (p = 0.037), indicating that the patients with an increase in Rnd2 expression relapse very rapidly in time and their survival is less than those with lower levels.

In the same way, we observe that the overall survival is less (average of 36.4 months compared with an average of 53 months) in those patients with Rnd2 overexpression with respect to the normal tissue (Fig. 8). The difference is significant (p=0.020).

Thus, these results suggest that Rnd2 overexpression in the lung is a factor of worse prognosis, since a significant reduction is observed in disease-free survival and in the overall survival of the patients.

Rnd3 mRNA expression levels in lung cancer

RhoE/Rnd3 expression in each sample is indicated in Fig. 9A, observing that most of the tumour samples have an expression level higher than expression in normal tissue (the average expression is 7.48, median 5.43, the normal tissue expresses 2.12, the pool of normal tissues 1.67). When we compare the averages in accordance with the clinical- pathological parameters (Fig. 9B, 9C and 9D), no significant differences are found.

The analysis of the disease-free survival or of the overall survival of the patient in accordance with the expression of Rnd3 does not give a significant relation. Despite observing that the average value of expression in this gene of the tumour samples is above the expression in normal tissue, we cannot conclude that Rnd3 is a prognosis factor in lung cancer from these results.

Below, the results are shown by way of synopsis: Breast cancer:

1 Rndl / Rho 6: is an oncogen which is overexpressed 100% of the times in the malignant cells of breast tissue and therefore can be used to diagnose breast cancer.

2 Rnd2 / Rho 7: behaves as a tumour suppressor since the results of the research suggest that the loss of Rnd2 expression is associated to the worst prognosis of the disease. Particularly well-known is the observation made with respect to the histological grade and to the disease-free survival on finding a statistically significant relation between the decrease in Rnd2 expression, the increase in histological grade (the lower levels of Rnd2 are located in histological grade III) and the decrease in disease-free survival. On analysing the overall survival practically, the association between the loss of Rnd2 expression and a less overall survival is significant. Therefore, Rnd2 can be considered as prognosis marker, their low levels being symptoms of bad prognosis. 3 Rnd3 / RhoE / Rho 8: the results of the research suggest that it behaves as a tumour suppressor gene in breast cancer since reduced levels of this gene, in comparison with normal tissue, are associated statistically to greater tumour size and to the relapse of the patient. With regard to survival analysis, it is considered that an amplification of the study could give significant results since there are only three cases of death in the analysis of the overall survival.

4 Rndl expression is homogeneous, it therefore being possible to behave as a marker useful for the diagnosis of breast cancer, the degree of expression of the other two markers (Rnd2 and Rnd3) being those which determine the prognosis or evolution of the disease.

Once Spearman's Rho coefficient of correlation is calculated: Coefficient of correlation

= 0.006, (p=0.972) it is deduced that the loss of expression of genes Rnd2 and Rnd3 do not have a linear relation, but it is independent.

Lung cancer: Rndl / Rho 6: Rndl underexpression is an independent factor of good prognosis, since these patients have a significantly greater overall survival.

Rnd2 / Rho 7: Rnd2 overexpression, two times above the expression levels of normal tissue, is a factor of worse prognosis, since the disease-free and overall survival are significantly less. Thus, the present invention relates in a first aspect to the method of in vitro diagnosis and/or prognosis of cancer which comprises the analysis of the expression profile of the Rnd 2/ Rho 7 or Rnd 3 /RhoE/ Rho 8 proteins, where the cancer analysed is breast cancer or lung cancer.

The present invention relates in a second aspect to a method of in vitro diagnosis and/or prognosis of cancer which comprises the analysis of the expression profile of the

Rnd 2/ Rho 7 or Rnd 3 /RhoE/ Rho 8 proteins, where the cancer analysed is breast cancer or lung cancer, which further comprises the analysis of the expression level of one of the

GTPases belonging to one of the following subfamilies of GTPases: Rho subfamily

(comprised of Rho A, Rho B and Rho C), Rac subfamily (comprised of Racl, Rac2, Rac3 and Rho G), Cdc42 subfamily (comprised of Cdc42, TClO, TCL, Chp and Wrch-1), Rho BTB subfamily (comprised of RhoBTB 1 and RhoBTB2), Miro subfamily (comprised of: Miro-1 and Miro-2) or the analysis of the expression level of one of the GTPases belonging to the group comprised of: Rho D, Rif and RhoH/TTF.

The present invention relates in a third aspect to a kit which comprises at least gene sequences capable of hybridizing with sequences belonging to the genes Rnd 2/ Rho 7 or Rnd 3 / Rho 8, for the in vitro diagnosis / prognosis of cancer where the cancer analysed is breast cancer or lung cancer.

The present invention relates in a fourth aspect to a kit which comprises at least gene sequences capable of hybridizing with sequences belonging to the genes Rnd 2/ Rho 7 or Rnd 3 / Rho 8, for the in vitro diagnosis / prognosis of cancer, where the cancer analysed is breast cancer or lung cancer, which further comprises at least gene sequences capable of hybridizing with one of the sequences belonging to the genes of the GTPases of one of the following subfamilies: Rho subfamily (comprised of Rho A, Rho B and Rho C), Rac subfamily (comprised of Racl, Rac2, Rac3 and Rho G), Cdc42 subfamily (comprised of Cdc42, TClO, TCL, Chp and Wrch-1), RhoBTB subfamily (comprised of RhoBTB 1 and RhoBTB2), Miro subfamily (comprised of: Miro-1 and Miro-2) or gene sequences capable of hybridizing with one of the sequences belonging to the genes of the GTPases comprised in the group: Rho D, Rif and RhoH/TTF.

TABLES

Table 1. Clinical-pathological parameters of the 54 samples of patients with infiltrative ductal breast carcinoma used in the study.

Breast samples n

Age 27-80 years

Histological grade

I 7 (13%)

II 20 (37%)

III 27 (50%)

PT

1 21 (38.9%)

2 33 (61.1%) pN

0 33 (61.1%)

1 12 (22.2%)

2 8 (14.8%)

X 1 (1.9%)

Relapse

No 41 (75.9%)

Yes 12 (22.2%) n.d. 1 (1.9%) Table 2. Clinical-pathological parameters of the 36 samples of patients with non- microcytic lung cancer used in the study.

Lung samples n

Age 43-82 years

Sex

Male 47 (88.7%)

Female 6 (11.3%)

Histological type

Bronchoalveolar

4 (7.5%) Adenoma

Adenocarcinoma 8 (15.1%)

Large-cell

5 (9.4%) carcinoma

Epidermoid 36 (67.9%)

PT

1 9 (17%)

2 40 (75.5%)

3 3 (5.7%)

4 1 (1.9%) pN

0 30 (56.6%)

1 13 (24.5%)

2 7 (13.2%)

3 1 (1.9%)

X 2 (3.8%)

Relapse

No 36 (67.9%)

Yes 13 (24.5%)

N.d. 4 (7.5%) Table 3. Taqman probes chosen for the amplification of GTPase family Mrna

son

Rho A Hs00357608 ml NM 001664 RHOA Family of genes homologous to ras, member A

Rho B HsOO26966θ^" ^~s1 NM 004040 RHOB Family of genes homologous to ras, member B

Rho C Hs00747110^~ ^~s1 NM 175744 RHOC Family of genes homologous to ras, member C

Rac 1 Hs00251654 ml NM 006908 RAC 1 Substrate 1 of Botulinum toxin C3 related to ras

Rac 2 Hs00427439 g1 NM 002872 RAC2 Substrate 2 of Botulinum toxin C3 related to ras

Rac 3 Hs00414037 g1 NM 005052 RAC3 Substrate 3 of Botulinum toxin C3 related to ras

Cdc42Hs Hs00741586 mH NM 044472 CDC42 Cell division cycle 42 (GTP-binding protein, 25 kDa)

Wrchi Hs00221873 ml NM 021205 RHOU Family of genes homologous to ras, member U N)

Chp/Wrch2 Hs00370444 g1 NM 133639 RHOV Family of genes homologous to ras, member V

TCL Hs00368156 ml NM 020663 RHOJ Family of genes homologous to ras, member J

TC10 Hs00817629 g1 NM 012249 RHOQ Family of genes homologous to ras, member Q

Rnd1 Hs00205507 ml NM 014470 RND1 GTPase 1 of the Rho family

Rnd2 HsOOI 83269 ml NM 005440 RND2 GTPase 2 of the Rho family

Rnd3/RhoE HsOOI 70603 ml NM 005168 RND3 GTPase 3 of the Rho family

RhoD Hs00205854 ml NM 014578 RHOD Family of genes homologous to ras, member D

RhoG Hs00750922^~ ^~s1 NM 001665 RHOG Family of genes homologous to ras, member G

Rif (RhoF) Hs00368034 ml NM 019034 RHOF Family of genes homologous to ras, member F

TTF(RhoH) HsOOI 80265 ml NM 004310 RHOH Family of genes homologous to ras, member H

RhoBTBI Hs00323872^~ ^~m1 NM 014836 RHOBTB1 Related to Rho containing domain BTB, 1

RhoBTB2 Hs00248529 ml NM 015178 RhoBTB2 Related to Rho containing domain BTB, 2

Miro-1 Hs00430266^~ ^"ml NM 018307 RHOT1 Family of genes homologous to ras, member T1

Miro-2 Hs00373355^~ ^"ml NM 138769 RHOT2 Family of genes homologous to ras, member T2

Table 4. Taqman Probes used in the study for the amplification of the mRNA of the main known effectors of the Rho GTPase family

Rocki HsOOI 78463_m1 NM 005406 ROCK1 protein kinase 1 associated to Rho with a "coiled-coil" motif

Rock2 HsOOI 53074_m1 NM 004850 ROCK2 protein kinase 2 associated to Rho with a "coiled-coil" motif

PKN HsOOI 77028_m1 NM 002741 PRKCL1 protein kinase similar to C, 1 mDiai HsOOI 93268_m1 NM 005219 DIAPH 1 homologue 1 of the "diaphanous" gene (Drosophila) mDia2 Hs00246501_m1 NM 006729 DIAPH2 homologue 2 of the "diaphanous" gene (Drosophila)

Citron Hs00392339_m1 NM 007174 CIT serine/threonine kinase 21 , which interacts with rho

Rhotekini Hs00369100_m1 NM 033046 RTKN rhotekin N)

Rhotekin2 Hs00377175_m1 NM 145307 PLEKHK1 member 1 of the K family with a domain homologous to pleckstrin

Rhophilini Hs00287628_m1 NM 052924 RHPN1 rhophilin 1 , Rho GTPase- binding protein

Rhophilin2 Hs00544666 ml NM 033103 RHPN2 rhophilin 2, Rho GTPase- binding protein

Table 4 continuation

Taqmaπ®

PAK 1 HS00176815 ml NM 002576 PAK1 kinase 1 activated by p21/Cdc42/Rac1

PAK2 HS00605586 ml NM 002577 PAK2 kinase 2 activated by p21 (CDKN1A)

PAK3 Hs00176828 ml NM 002578 PAK3 kinase 3 activated by p21 (CDKN1A)

PAK4 Hs00178686 ml NM 005884 PAK4 kinase 4 activated by p21 (CDKN1A)

MLK2 Hs00374249 ml NM 002446 MAP3K10 protein kinase kinase kinase 10 activated by mitogens

MLK3 Hs00176759 ml NM 002419 MAP3K1 1 protein kinase kinase kinase 1 1 activated by mitogens

MEKK1 Hs00394890 ml AF042838 MEKK1 MEK 1 kinase effectors MEKK4 Hs00245928 ml NM 005922 MAP3K4 protein kinase kinase kinase 4 activated by mitogens N⁾

IQGAP 1 HsOOI 82622 ml NM 003870 I QGAP 1 GTPase 1 activating protein containing IQ motif OD

IQGAP2 Hs00183606 ml NM 006633 IQGAP2 GTPase 2 activating protein containing IQ motif

ACK Hs00416477 ml NM 005781 ACK1 kinase activated by p21cdc42Hs

N-WASP Hs00187614 ml NM 003941 WASL gene similar to Wiskott-Aldrich's syndrome

WASP Hs00166001 ml NM 000377 WAS gene similar to Wiskott-Aldrich's syndrome

WAVE/Scar2 Hs00819075 gH NM 006990 WASF2 member 2 of the WAS gene family

POSH Hs00325806 ml NM 020870 SH3RF SH3 domain containing ring finger motif

PAR6 Hs00180947 ml NM 016948 PARD6A gene homologue of pair-6, alpha

MRCK Hs00207976 ml NM 014826 CDC42BPA CDC42 alpha-binding protein kinase, similar to DMPK

MRCK Hs00178787 ml NM 006035 CDC42BPB CDC42 beta-binding protein kinase, similar to DMPK

Table 5. Relation between the expression level of mRNA of Rnd2 and each clinical- pathological factor.

* p < 0.05

Table 6. Rho-family GTPase proteins studied in the present invention

Bibliography 1. Aronheim, A., Broder, Y. C, Cohen, A., Fritsch, A., Belisle, B. and Abo, A. (1998). Chp, a homologue of the GTPase Cdc42Hs, activates the JNK pathway and is implicated in reorganizing the actin cytoskeleton. Curr Biol 8, 1125-8.

2. Bektic, J., Pfeil, K., Berger, A. P., Ramoner, R., Pelzer, A., Schafer, G., Kofler, K., Bartsch, G. and Klocker, H. (2005). Small G-protein RhoE is underexpressed in prostate cancer and induces cell cycle arrest and apoptosis. Prostate 64, 332-40.

3. Bektic, J., Wrulich, OR. A., Dobler, G., Kofler, K., Ueberall, F., Culig, Z., Bartsch, G. and Klocker, H. (2004). Identification of genes involved in estrogenic action in the human prostate using microarray analysis. Genomics 83, 34-44.

4. Chardin, P. (2006). Function and regulation of Rnd proteins. Nat Rev MoI Cell Biol 7, 54-62.

5. Gomez del Pulgar, T., Benitah, S. A., Valeron, P. F., Espina, C. and Lacal, J. C. (2005). Rho GTPase expression in tumourigenesis: evidence for a significant link. Bioessays 27, 602-13.

6. Nobes, CD., Lauritzen, L, Mattei, M.G., Paris, S., Hall, A. and Chardin, P. (1998) A new member of the Rho family, Rndl, promotes disassembly of actin filament structures and loss of cell adhesion. J Cell Biol, 141, 187-197.

7. Riento, K., Guasch, R. M., Garg, R., Jin, B. and Ridley, A. J. (2003). RhoE binds to ROCK I and inhibits downstream signaling. MoI Cell Biol 23, 4219-29.

8. Riento, K., Totty, N., Villalonga, P., Garg, R., Guasch, R. and Ridley, A. J. (2005). RhoE function is regulated by ROCK I-mediated phosphorylation. Embo J 24, 1170-80.

9. Smith, T. M., Lee, M. K., Szabo, C. L, Jerome, N., McEuen, M., Taylor, M., Hood, L. and King, M. C. (1996). Complete genomic sequence and analysis of 117 kb of human DNA containing the gene BRCAl. Genome Res 6, 1029-49. 10. Trojan, L., Schaaf, A., Steidler, A., Haak, M., Thalmann, G., Knoll, T., Gretz, N., Alken, P. and Michel, M. S. (2005). Identification of metastasis-associated genes in prostate cancer by genetic profiling of human prostate cancer cell lines. Anticancer Res 25, 183-91.

11. Villalonga, P., Guasch, R. M., Riento, K. and Ridley, A. J. (2004). RhoE inhibits cell cycle progression and Ras-induced transformation. MoI Cell Biol 24, 7829-40.

12. Wennerberg, K. and Der, C. J. (2004). Rho-family GTPases: it's not only Rac and Rho (and I like it). J Cell Sci 117, 1301-12.

Claims

1. Method of in vitro diagnosis and/or prognosis of cancer which comprises the analysis of the expression profile of the Rnd 2/ Rho 7 protein.

2. Method of in vitro diagnosis and/or prognosis of cancer according to claim 1, which further comprises the simultaneous analysis of the expression profile of one of the Rnd- subfamily GTPases, belonging to the group comprised of: Rndl/Rho 6, Rnd2/Rho 7 or Rnd 3 /RhoE/ Rho 8 or combinations thereof.

3. Method of in vitro diagnosis and/or prognosis of cancer according to either of claims 1 or 2, which further comprises the analysis of the expression profile of one of the Rho-sub family GTPases belonging to the group comprised of: Rho A, Rho B, Rnd 3 /RhoE/ Rho 8 or Rho C.

4. Method of in vitro diagnosis and/or prognosis of cancer according to any of claims 1-3, which further comprises the analysis of the expression profile of one of the Rac- subfamily GTPases belonging to the group comprised of: Racl , Rac2, Rac3 and Rho G.

5. Method of in vitro diagnosis and/or prognosis of cancer according to any of claims 1-4, which further comprises the analysis of the expression profile of one of the Cdc42- subfamily GTPases belonging to the group comprised of: Cdc42, TClO, TCL, Chp and Wrch-1.

6. Method of in vitro diagnosis and/or prognosis of cancer according to any of claims 1-5, which further comprises the analysis of the expression profile of one of the RhoBTB- subfamily GTPases belonging to the group comprised of: RhoBTBl and RhoBTB2.

7. Method of in vitro diagnosis and/or prognosis of cancer according to any of claims 1-6, which further comprises the analysis of the expression level of one of the Miro- subfamily GTPases belonging to the group comprised of: Miro-1 and Miro-2.

8. Method of in vitro diagnosis and/or prognosis of cancer according to any of claims 1-7, which further comprises the analysis of the expression profile of one of the GTPases belonging to the group comprised of: Rho D, Rif and RhoH/TTF.

9. Method of in vitro diagnosis and/or prognosis of cancer according to any of claims 1-8, where the cancer analysed is breast cancer.

10. Method of in vitro diagnosis and/or prognosis of cancer according to any of claims 1-8, where the cancer analysed is lung cancer.

11. Test device which comprises at least sequences of nucleic acids capable of hybridizing with sequences belonging to the Rnd2/Rho 7 gene for the diagnosis / prognosis /monitoring of the treatment of cancer.

12. Test device according to claim 11, which comprises at least sequences of nucleic acids capable of hybridizing with sequences belonging to the Rndl/Rho 6, Rnd2/Rho 7 or Rnd 3 / Rho 8 genes, for the in vitro diagnosis / prognosis of cancer.

13. Test device, according to claims 11 or 12, which further comprises sequences of nucleic acids capable of hybridizing with one of the sequences belonging to the genes of the Rho-subfamily GTPases belonging to the group comprised of: Rho A, Rho B and Rho C.

14. Test device, according to claims 11-13, which further comprises sequences of nucleic acids capable of hybridizing with one of the sequences belonging to the genes of the GTPases of the Rac subfamily belonging to the group comprised of: Racl, Rac2, Rac3 and Rho G.

15. Test device, according to claims 11-14, which further comprises sequences of nucleic acids capable of hybridizing with one of the sequences belonging to the genes of the Cdc42-subfamily GTPases belonging to the group comprised of: Cdc42, TClO, TCL, Chp and Wrch-1.

16. Test device, according to claims 11-15, which further comprises sequences of nucleic acids capable of hybridizing with one of the sequences belonging to the genes of the RhoBTB-subfamily GTPases belonging to the group comprised of: RhoBTBl and RhoBTB2.

17. Test device, according to claims 11-16, which further comprises sequences of nucleic acids capable of hybridizing with one of the sequences belonging to the genes of the Miro-subfamily GTPases belonging to the group comprised of: Miro-1 and Miro-2.

18. Test device, according to claims 11-17, which further comprises sequences of nucleic acids capable of hybridizing with one of the sequences belonging to the GTPase genes belonging to the group comprised of: Rho D, Rif and RhoH/TTF.

19. Test device according to any of claims 11-18, where the cancer analysed is breast cancer.

20. Test device according to any of claims 11-18, where the cancer analysed is lung cancer.