WO2009084740A1 - Method and composition for predicting the postsurgical recurrence in patient suffering from pulmonary adenocarcinoma - Google Patents

Abstract

Disclosed is a method for predicting the recurrence of pulmonary adenocarcinoma after surgically removing lung cancer in a patient suffering from pulmonary adenocarcinoma by using a specific gene set. Also disclosed is a composition for use in the method.

Description

 Methods and compositions for predicting postoperative recurrence of lung adenocarcinoma patients

 The present invention relates to methods and compositions for predicting post-operative recurrence of lung adenocarcinoma patients. Background art

 Lung cancer is the leading cause of cancer-related mortality in developed countries. Lung cancer is roughly divided into small cell carcinoma and non-small cell carcinoma, and non-small cell carcinoma is mainly adenocarcinoma, squamous cell carcinoma, large cell carcinoma.

Cell carcinoma. Among non-small cell lung cancer, lung adenocarcinoma accounts for about half of non-small cell cancer and is known to occur most frequently in Japan. In addition, the 10-year survival rate of patients with non-small cell carcinoma is very low at about 10%, and about 25-30% of the patients have only undergone surgical resection in stage I patients, Approximately 35 to 50% of these patients have relapsed within 5 years.

 In recent years, with advances in technologies such as microarrays and mass spectrometry, new technology areas such as bioinformatics and proteomics using these have been developed. Also in cancer research, comprehensive analysis of gene expression using such technology is actively conducted for the purpose of diagnosis and treatment of cancer.

 With regard to lung cancer, research has also been conducted to pursue correlation between classification of lung cancer, prediction of metastasis and prediction of recurrence in lung cancer patients and gene expression patterns and clinical parameters (Non-patent Document 1).

 The present inventors have also comprehensively analyzed gene expression patterns, focusing on lung cancer, and analyzed expression patterns associated with metastasis and prognosis prediction (Non-patent Document 2). In addition, the present inventors recently classified lung adenocarcinoma into TRU-a, TRU-b, and non-TRU types, and based on this classification, a method for predicting the postoperative prognosis of lung adenocarcinoma patients And discloses a composition (Patent Document 1).

With regard to lung adenocarcinoma, yet another group has a set of genes to distinguish lung adenocarcinoma, a set of genes to distinguish pleural invasion of lung adenocarcinoma, and a prediction of recurrence of lung adenocarcinoma. Discloses a gene set for the gene expression (patent documents 2 to 4). Among them, Patent Document 4 describes a method for predicting postoperative recurrence of lung adenocarcinoma and a set of 45 genes, and the discrimination rate of lung adenocarcinoma by this gene set is 50% or more. It is stated that In addition, Non-Patent Document 3 discloses the prediction of recurrence of lung adenocarcinoma by a set of 54 genes, which also accurately indicates the recurrence by the conventional method for determining the prognosis of patients with lung adenocarcinoma. The difficulty of prediction is stated.

 Patent Document 1 International Publication WO 200 7/04341 8

 Patent Document 2 Japanese Patent Application Laid-Open No. 2007-679 1

 Patent Document 3 Japanese Patent Application Laid-Open No. 200 7- 6 792

 Patent Document 4 Japanese Patent Application Laid-Open No. 200 7-1 3 5466

 Non-patent literature, 1 D. G. B e er et al., Na t u r e Med ic i n e 2 002, 8 (8): 8 1 6-8 24

 Non-patent literature 2 Shuta Kuwata and Takashi Takahata, "Diagnosis of treatment of lung cancer based on expression profile analysis' New development of treatment" Experimental medicine 2004, 22 (14): 45-50, Yodosha (Japan) Non-patent literature 3 J E. L Arsen et al., C Clinical Cancer R esearch 200 7, 1 3: 2946-29 54 Disclosure of the Invention

 Some cancer patients have a significantly worse prognosis due to metastasis, recurrence, etc. after surgical removal of the tumor and some patients die within 5 years; conversely, there are patients without metastasis or recurrence, that is, the prognosis is good. The present inventors have focused on the fact that there is no method that enables the prediction of recurrence particularly with high accuracy in predicting the prognosis of lung cancer in lung adenocarcinoma patients relatively many in lung cancer patients in Japan. The prediction of recurrence after surgery for lung cancer is very difficult because it is related to various factors. If the recurrence of lung adenocarcinoma patients can be predicted with a very high degree of certainty, it is necessary to identify a group that needs treatment among those currently being followed up without further treatment after surgery. It is possible that it will be possible and improve the prognosis.

The conventional recurrence prediction method for lung adenocarcinoma is a set of genes that can predict the risk of recurrence using leave—one—outcross-va 1 identification. Have been identified. At this time, when using the determination criteria that determination is impossible, the apparent determination rate and the true determination rate may be extremely different depending on the proportion of undecidable cases included. In addition, in leave-one-outcross-va 1 identification, even when there are very few data to be analyzed, the number of repetitive calculations is small, and this is an effective method for evaluating the accuracy of the prediction model. It is preferable to perform more accurate model evaluation by increasing the data and performing iterative calculations incorporating the 10 0 fo 1 d cross-parison method and random division.

 Furthermore, when linking clinical gene to comprehensive gene expression profiling of cancer cells using microarrays, there are doubts about reproducibility and reliability due to differences in sampling methods, platform differences, statistical analysis, etc. It is pointed out that there is a problem (for example, Shigeyuki Matsui, Experimental Medicine 2007, 25 (17): 72- 78 (extra edition) Yodosha (Japan)).

 The object of the present invention is to provide a method and composition for predicting the recurrence of lung adenocarcinoma with extremely high accuracy in predicting the recurrence of lung adenocarcinoma after surgical removal of lung cancer in patients with lung adenocarcinoma. It is.

 The present inventors are different from the judgment method of undecidable, that it is a method that presents judgment of either high recurrence risk or low recurrence risk to all cases, and is different from the conventional recurrence prediction gene set, 82 The use of a novel set of genes consisting of different types of genes for the determination made it possible to predict the presence or absence of recurrence with a probability of more than 70% in the evaluation model and completed the present invention.

 The invention, in summary, includes the following features.

 The present invention relates, in a first aspect, to a method for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients, comprising a gene comprising each nucleotide sequence of SEQ ID NO: 1-82 Measuring the expression of a gene set of 2 to 82 genes selected from the group: in a biological sample derived from the patient, wherein SEQ ID NOs: 1 to 5, 8 to 9, 12, 14 ~: 1 5, 17, 19-21, 23-28, 3

When the expression of a gene containing a nucleotide sequence of any of 0 to 33, 35 to 59, 61 to 68, 70 to 82 is relatively high compared to a postoperative recurrence-free case, or The expression of the gene containing the nucleotide sequence of any of the numbers 6 to 7, 10 to 11: 11, 13, 16, 18, 22, 29, 34, 60, 69 compared with the postoperative recurrence-free case The above method is provided, which is characterized in that the recurrence of lung adenocarcinoma is predicted to be high when it is relatively low.

 According to an embodiment of the present invention, the number of genes of the gene set is 5 or more and 82 or less.

 According to another embodiment of the present invention, the measurement of the expression of the set of genes is performed by measuring the presence or the amount of the nucleic acid or the protein corresponding to the gene.

 According to another embodiment of the present invention, the measurement of the expression of the gene set is performed by a hybridization method or a quantitative PCR method.

 According to an embodiment of the present invention, the hyperprime method is a microarray method or a plotting method.

 According to an embodiment of the present invention, the hybridization method is carried out using nucleic acid which hybridizes to the RNA transcript derived from the gene, cDNA, aRNA or cDNA.

 According to another embodiment of the present invention, the measurement of the expression of the gene set is performed by an immunological method.

 According to an embodiment of the present invention, said immunological method is performed using an antibody against the protein encoded by said gene or a fragment thereof.

 In the second aspect, the present invention relates to a composition for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients, which composition is one of SEQ ID NOS: 1-82. A plurality of nucleic acids which make it possible to measure the expression of two or more different genes comprising different nucleotide sequences, said nucleic acids comprising

 (1) Nucleotide sequences as shown in SEQ ID NOs: 1 to 2, nucleotide sequences complementary thereto, or nucleic acids having partial sequences of 15 bases to less than the total number of bases in their nucleotide sequences, and

(2) A nucleic acid hybridizing with a nucleic acid comprising the nucleotide sequence shown in SEQ ID NO: 1 to 82 or a nucleotide sequence complementary thereto, or the hybridizing agent The above composition is provided, which is selected from the group consisting of: a nucleic acid having a partial sequence consisting of 15 contiguous bases to less than the total number of bases of the nucleic acid of W.

 In an embodiment of the invention, the composition is in the form of a kit, a microarray or a set of primers.

 In another embodiment of the present invention, the above nucleic acid enables measurement of the expression of 5 or more to 82 or less genes.

 In another embodiment of the present invention, the composition is for use in the above method of the present invention.

 In the third aspect of the present invention, there is provided a composition for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in a lung adenocarcinoma patient, the composition comprising the composition of SEQ ID NO: 1 to 82. Comprising a plurality of antibodies which make it possible to measure the expression products of two or more different genes comprising different nucleotide sequences, said antibodies comprising any of the nucleotide sequences shown in SEQ ID NO: 1 to 82. Selected from the group consisting of: an antibody which immunologically reacts with a protein encoded by or a polypeptide consisting of at least 8 consecutive amino acid residues of the amino acid sequence of said protein, and a fragment of said antibody The above composition is provided.

 In an embodiment of the invention, the composition is in the form of a kit.

 In another embodiment of the present invention, the above-mentioned antibody probe enables measurement of the expression product of 5 or more to 82 or less genes.

 In another embodiment of the present invention, the composition is for use in the above method of the present invention.

 As used herein, "post-operative recurrence" means that the patient recurs lung adenocarcinoma after surgical removal of the lung adenocarcinoma. Patients who are predicted to relapse are determined to be at high risk for recurrence and have a low survival rate. In contrast, patients who are predicted to have no recurrence are judged to have a low risk of recurrence and have high survival rates.

 As used herein, "patient" refers to a mammal, including humans, with a preferred mammal being humans.

In the present specification, "nucleic acid" is either DNA or RNA, and includes, for example, transcripts (mRNA) of each gene, cDNA, aRNA, cDNA and the like. As used herein, a "gene" encodes a specific protein and is composed of: exon, intron, 5, 1 and 3 '-untranslated region, etc., which enables the formation of precursor mRNA by its transcription, and Through splicing events, they become mRNA, which is translated into protein.

 As used herein, "protein" includes a protein encoded by a specific gene, a variant thereof or a fragment thereof. Such variants include variants based on natural events such as splicing, polymorphism and the like. In addition, such a fragment is a protein fragment naturally generated in vivo by the action of an enzyme such as a protease or peptidase.

In the present specification, "hyperdilution" is a reaction for forming a double strand by base pairing of nucleic acid and nucleic acid. The hybridizations include DNA-DNA hybridization, DNA-RNA hybridisation, and RNA-RNA hybridization. Hybridization is usually performed using a nucleic acid probe immobilized on a solid support. The solid support may be made of plastic, polymer, glass, etc., and may be in the form of a test plate, test tube, test strip, chip or the like. In hybridization, temperature, ionic strength, surfactant concentration, presence or absence of formamide, GC content of nucleic acid, etc. affect the stability of the hybrid and the hybridization rate. In general, the higher the temperature or the higher the ionic strength, that is, the more stringent (stringent) conditions, the lower the stability of the hybrid and conversely, the specificity of the hybridization. For hybridism, Au sbe 1 FM et al. 'S hort P rotocolsin Molecular biology {反 反 し A omp endi um of Me thodsfr om C urrent P rotocolsin Mocular B iology, 19 9 5 J, J ohn Wi ley & Sons, Inc. (US). As used herein, "immunologically" means forming a complex of an antigen and an antibody. A reaction that forms such a complex is referred to as an immunological reaction, and a method of detecting or quantifying an antigen or antibody utilizing this reaction is referred to as an immunological method. Immunological methods can be performed under conditions such as solid phase or liquid phase, homogeneous or heterogeneous, etc. For detection of the antigen-antibody complex, a method using a label (eg, a dye, a fluorescent substance, a radioactive isotope, an enzyme, etc.), a method using aggregation (eg, latex method), a method using turbidity, It can be performed by a method using immunoprecipitation, a method using a secondary antibody (eg, sandwich method), and the like.

 The present specification includes the contents described in the specification and / or drawings of Japanese Patent Application No. 200 7-3 39 2 14 which is the basis of the priority of the present application. Brief description of the drawings

 FIG. 1 shows a schematic of a training-one efficacy strategy for identifying relapse-associated signatures using 100 iterations of the training set randomly divided into 10 1 0 − f o d c r o s s − v a i i d a t i o n.

 Figure 2 shows the results of a training procedure to identify relapse-associated signatures. Figure 2A shows the results of the search for the optimal number of probes to define recurrence related signatures. Survival rates in the training group were assessed using K ap 1 a n -Me i e r survival curves. The patient's recurrence related curves at all stages (Figure 2B) and at stage I (Figure 2C) are also shown.

 FIG. 3 shows K ap 1 a n -Me ier survival in the presence or absence of EGFR mutations (FIG. 3 A), K 1 ras mutations (FIG. 3 B) and p 5 3 mutations (FIG. 3 C) in the training data set Indicates the fountain.

 Figure 4 shows the effectiveness of the RR S-82 signature using evaluation set I and evaluation set I I. Ka p 1 a n -Me i er survival curves were presented according to the risk predicted based on RRS-82. FIG. 4A shows all-stage recurrence-free survival curves, FIG. 4B shows all-stage overall survival curves, FIG. 4C shows stage I recurrence-free survival curves, and 4D stage I overall survival curves. In addition, the recurrence-free survival curve and the overall survival curve are shown in FIG. 4E and FIG. 4F, respectively, using the evaluation set II consisting of the stage I only samples. The recurrence-free survival curves in the evaluation set I + I I are shown in FIG. 4G, and the overall survival curves are shown in FIG. 4H.

 Figure 5 shows the relationship between RRS- 82 based risk groups and the TNM stage. Figure

5A is a recurrence-free survival curve for patients with low-risk RR S-82, Figure 5B. Each shows a relapse-free survival curve for patients with high-risk RRS—82 signatures.

 Figure 6 shows the effectiveness of the RR S _ 82 signature by analysis of independent data sets. Figure 6A shows the results of an unsupervised hierarchical clustering analysis performed using the components of RR S _ 82 2 signatures in the data set of University of Duke consisting of 39 lung adenocarcinomas. B shows Kaplan-Meier survival curves of clusters I and II. Also, Fig. 6C shows a conceptual diagram of constructing a prediction model using the data set of the University of Michigan. Fig. 6D shows Ka a n-Me i survival curves obtained by analyzing the gene expression data of the cancer center's cancer center gene 104 using the model constructed by Fig. 6 C.

 FIG. 7 shows the assessment of K ap 1 an −Mei er survival in the training data set. Figure 7A shows the overall survival curve with RR S-82 signatures in multiple patients at all stages, and Figure 7B shows the overall survival with RR S- 82 signatures in multiple patients with stage I. The survival curve is shown.

 FIG. 8 shows the evaluation of K a l one i me i e r survival by R R S s-2 in an evaluation set I middle stage I I to I I I cases. Of the stage II-III patients, recurrence-free (Figure 8A) and overall survival (Figure 8B) are better for patients with low risk RR S-82 signatures compared to patients with high risk signatures. There was a trend but it was not statistically significant.

 FIG. 9 shows that in the evaluation set I, the association between RR S-82 and the stage was evaluated using the K ap 1 an -Me ier survival curve. In the low-risk group shown in Figure 9A, the overall survival curve of stage I cases tends to be better, although not statistically significant, in comparing stage I cases with stage I cases. In the 髙 at risk group shown in Fig. 9B, the overall survival curves are almost the same in the comparison of Stage I and Stage I I I I I I cases. BEST MODE FOR CARRYING OUT THE INVENTION

The present invention, as described above, is a method of rescuing lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients. Provide a method to predict the development with high accuracy. The reason why such high-accuracy prediction has been made possible is, in part, the rigorous selection of postoperative recurrence cases and recurrence-free cases among lung adenocarcinoma patients based on the expertise and experience of specialists. Under conditions such as using reliable statistical methods, based on finding previously unpublished 82 gene sets related to postoperative recurrence / no recurrence in lung adenocarcinoma patients. There is.

<Determination of gene set>

 In the present invention, a set of 82 different genes for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients was found, and the procedure for determining this gene set will be described below. .

 As described in the background art, although a set of genes for predicting postoperative recurrence of lung adenocarcinoma has been reported by some groups, a set of genes described as being predictable from tens of thousands of gene groups Each have no overlap with one another or very low overlap. Also with regard to the 82 gene set found by the present inventors, genes overlapping with the 45 gene set described in, for example, Japanese Patent Application Laid-Open No. 2007-135466 are evaluated, and J. E. Larsen et al., Only 1 gene overlaps with the 54 gene set described in C clinical Cancer Research 2001, 1 3: 2 946-2 954. However, as shown in Fig. 2A, the correct answer rate when performing recurrence prediction using a single gene (ie, [number of correct answers / (number of correct answers + number of incorrect answers)] X 1 00 (%) drops significantly. In order to increase the correct answer rate, it is possible to use a set of multiple, preferably 5 or more, more preferably 10 or more, more preferably 30 or more, most preferably 82 genes. desirable. Figure 2A shows that even if more than 82 genes are used, the correct answer rate tends to get better or worse.

As described in the examples below, the present inventors prepared 87 lung adenocarcinoma samples from lung adenocarcinoma patients who underwent excision surgery, of which 60 were randomly assigned to the training set. The remaining 27 were used as the evaluation set. A supervised classification technique (weighted voting algorithm; Y anagisa wa et al., J. Na tl t) using a probe signal that exceeds background expression levels for the training set samples and uses probe signals that meet this filtering criteria. C cancelr I nst. 9 9: 8 5 8 – 6 7, 2 0 0 7), which allows 82 genes that can predict high risk or low risk of relapse with a high degree of certainty on average of more than 75%. That is, they were identified as having a high risk of recurrence or recurrence, or no risk of recurrence or low recurrence). At this time, a 10 — fo 1 d cross validation method was used to select a probe signal to identify a recurrence free signal. The outline of this method is shown in FIG.

 For example, the 10 0-fo 1 d cross validation method divides 10 0 samples into 10 samples at 1 0, and the remaining 9 divisions excluding 1 division from this 1 分割 division have characteristics (relapse related signatures). Make a selection, construct a classifier consisting of the selected top 1 to a maximum of 2 0 0 predicted probes, and perform 1-division classification without exclusion. Next, from the top of the selected 1st place up to a maximum of 0. Construct a classifier consisting of predicted probes of, and perform another one-division classification without. Do the same operation a total of 10 times. Here, for example, in consideration of the fact that there are innumerable allocation methods for dividing 100 samples into 10 0, in order to reduce bias due to allocation methods, random (random) methods are used in 100 methods. 10 ◦ divided the sample into 10 In this way, a maximum of 200 predicted probes can be generated for a total of 100 0 independent sets that have undergone 1 0-fo 1 d cross-validation methods for every 1 0 0 0 random partitions. Was built. As a result, we find that the number of predicted probes is 82, which leads to the least learning error (Fig. 2A), and that the most frequently shared 8 2 probes among independent 1, 00 0 sets are associated with recurrence Identify it as a signet, and build the final classifier in this way.

 Classify test samples using the classifier constructed above. As a test sample, lung adenocarcinoma resected from randomly selected lung adenocarcinoma patients

2 7 evaluation samples were used. Recurrence-associated signatures obtained from evaluation samples using Kaplan-Meier survival curves and Cox proportional hazards model analysis, and overall survival

Analyze the relationship between (overal l survival) and relapse-free survival (relapse-free survival). As a result, the 82 gene set of the present invention (referred to as "RRS-82") is preferably 75% or more of high risk patients with relapse-death and low risk patients with no recurrence and high survival. It is possible to determine with a high degree of accuracy of 80 ° / ₀ or more, and more preferably 90% or more.

 RR S-82 is a set of 82 kinds of genes including each nucleotide sequence of SEQ ID NOs: 1 to 82. This gene set is shown in Table 1.

table 1

No. 2 to No. 8 2 in the above table respectively correspond to SEQ ID NO: 2 to SEQ ID NO: 2 It corresponds to 8 Two types of genes are classified according to relative expression difference with postoperative recurrence-free cases, ie, for recurrence-associated siginthia, ie recurrence (death) signet or high relapse risk signal sign and relapse-free sign or low recurrence risk signet, Then I was able to divide it into the following two groups.

 (1) SEQ ID NOs: 1 to 5, 8 to 9, 12, 14, 15 to 17, 17 to 19, 21 to 2, 2 to 28, 30 to 33, 35 to 59, 6 to 68, A group in which the recurrence of lung adenocarcinoma can be predicted to be high (or the relative expression is low) if the expression of a gene containing 70 to 82 nucleotide sequences is relatively high compared to a postoperative recurrence-free case. A group where recurrence of lung adenocarcinoma can be predicted to be low), and

 (2) SEQ ID NOs: 6 to 7, 1 to 0: 1, 1, 13, 16, 18, 22, 29, 34, 60, 69 The expression of a gene containing each nucleotide sequence does not cause any recurrence after surgery A group whose lung cancer recurrence can be predicted to be high if it is relatively low compared to cases (or, if the relative expression is high, lung cancer recurrence is predicted to be low).

 That is, in the case of genes of group (1) above, lung adenocarcinoma patients are predicted as relapse (death) patients or high risk patients, or non-relapse patients or low recurrence risk patients depending on whether the relative expression is high or low. · Can be classified. Also, in the case of the gene of the above-mentioned (2) group, the lung adenocarcinoma patient is conversely a non-relapsed patient or a low risk patient depending on whether the relative expression is low or high, or a relapse (death) patient or Predictable as high risk patients-can be classified.

 If it is determined that the method of the present invention is a recurrence or high risk, it will be possible to formulate an early treatment plan for suppressing postoperative recurrence of lung adenocarcinoma patients, while, on the other hand, no recurrence or low risk. If determined, it can also serve as a basis for avoiding useless chemotherapy and radiation therapy.

 The control in comparing the relative expression levels of the above genes is a postoperative non-recurring case, and if the above prediction can be made, a predetermined period after surgery (eg 3 years, preferably 5 years, more preferably 10) It is a sample (lung adenocarcinoma tissue or cells) derived from a patient who is judged to have no recurrence.

No statistically significant association was found between RRS-82 of the present invention and EGFR, K-ras or p53 gene mutations (FIG. 3). When the discriminatory ability of RRS-82 in the present invention was verified using an evaluation set (27 lung adenocarcinoma samples), all patients showing a high risk sign of RR S _ 82 had lung glands within 5 years. About 70% of patients who relapsed cancer, but showed low risk signs did not show signs of recurrence after 5 years (Figure 4). In this sense, it is concluded that the method of the present invention is extremely effective in predicting a high risk group of postoperative recurrence in lung adenocarcinoma patients.

 The recurrence / no recurrence prediction according to the method of the present invention can be performed regardless of the tumor stage, and interestingly, even in the early stage.

<Prediction of postoperative recurrence of lung adenocarcinoma>

 The present invention is, in one aspect thereof, a method for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients, comprising a gene comprising each nucleotide sequence of SEQ ID NO: 1-82 Measuring the expression of a gene set of 2 to 2 genes selected from the group in a biological sample derived from said patient, wherein: SEQ ID NOs: 1 to 5, 8 to 9, 12, 14 Expression of a gene containing a nucleotide sequence of any one of 1 to 15, 1 7, 19 to 21, 1, 23 to 28, 30 to 33, 35 to 59, 6 to 68, 70 to 82 Is relatively high compared to the postoperative recurrence-free cases, or array numbers 6 to 7, 1 0 to: 1 1, 13, 16, 18, 22, 29, 34, 60, If the expression of a gene containing any of the 69 nucleotide sequences is relatively low compared to post-surgery recurrence-free cases, it is characterized in that the recurrence of lung adenocarcinoma is predicted to be high. How to provide.

 The above 82 genes are human genes comprising the nucleotide sequences shown by SEQ ID NOs: 1-8, as described above. Since it is a gene derived from an organism, it is considered that mutations such as polymorphic mutation, splice mutation, mutation may occur. Therefore, the target gene of the present invention may contain not only nucleotide sequences represented by SEQ ID NOs: 1 to 2 but also deletions, substitutions or additions of one or several nucleotides in these sequences, Or at least for these sequences

It may have an identity of 95%, preferably at least 97%, more preferably at least 98%, most preferably at least 99%.

As used herein, the term "identity (%)" introduces a gap or It can be determined using a known BL AS T or FA S TA algorithm without introducing a gap (K ar 1 in and AI t ch ul, 199 3, P roc. N atl. A ca d. S ci. USA, 90: 58 73- 5 8 77; Karlin and A 1 tschu 1, 1 990, Pr o C. N at 1. A cd S. USA, 8 7: 2264-22 68; A ltschul et al., 1 9 90, J. Mo 1. Biol., 215: 403; Altschul et al., 1 9 9 7, Nucleic Acids Res 25: 3389- 3402). In general, identity (%) can be calculated as a percentage of the number of matched bases relative to the total number of bases.

 The term "several" as used herein refers to an integer from 2 to 10, that is, any of 2, 3, 4, 5, 6, 7, 8, 9 or 10 , Preferably an integer of 2 to 5, more preferably 2 or 3.

 In the present invention, the number of genes in the gene set is 2 to 100 (82), and the number of genes in this range can predict post-operative recurrence. However, to further improve the prediction accuracy (or accuracy), it is preferable to set the number of genes to 5 to 82, more preferably 10 to 82, and still more preferably 30 to 82, for example 40 to 82, 50 to 82, 60 to 80, 70 to 82, and the most preferable number of genes is 82.

 Gene expression can be measured for the amount or presence of nucleic acids or proteins. In the measurement of nucleic acid, the transcripts of each gene, mRNA and its derivative nucleic acid, for example, cDNA, cRNA, RNA (an RNAi RNAi) can be used as a measurement target. On the other hand, in the measurement of protein, a protein encoded by each gene (also referred to as an expression product or gene product), a fragment thereof or a post-translational modification product (for example, modification by sugar etc.) can be the measurement target.

 The following describes a method for measuring the expression of the above-mentioned 82 genes at the nucleic acid and protein levels.

The test sample is lung adenocarcinoma tissue or cells of a lung adenocarcinoma patient, such as cancer tissue resected by surgery or tissue or cells obtained by biopsy. Measurement of gene expression by nucleic acid

 In the method of the present invention, the following nucleic acids can be used as probes or primers to detect the expression of 82 genes.

 (1) Nucleotide sequences as shown in SEQ ID NOs: 1 to 2, nucleotide sequences complementary thereto, or nucleic acids having partial sequences of 15 bases to less than the total number of bases in their nucleotide sequences, and

 (2) A nucleic acid that hybridizes to a nucleic acid comprising the nucleotide sequence shown in SEQ ID NO: 1 to 8 or a nucleotide sequence complementary thereto, or alternatively, from 15 bases to less than the total number of bases of the hybridizing nucleic acid. Nucleic acid having a partial sequence.

 The size of the nucleic acid fragment consisting of the above partial sequence is 15 bases to less than the total number of bases, for example, 17 bases or more, 20 bases or more, 30 bases or more, 30 bases or more, 50 bases or more, 70 bases or more, 100 bases The above is the range of 150 bases or more, 200 bases or more, 300 bases or more, 400 bases or more, or 500 bases or more to less than the total number of bases.

 The above-mentioned nucleic acid is a DNA molecule having a size of about 100 bases or less, the ability to be synthesized using a commercially available automatic DNA synthesizer using a phosphoramidite method or the like, or a DNA molecule having a size of about 100 bases or more. C DNA can be prepared by cloning and polymerase chain reaction (PCR). In addition, cDNA and a RNA can be prepared from cDNA using a commercial kit and the like.

 c In DNA cloning and PCR, total RNA is obtained from lung adenocarcinoma tissue, and poly A (+) RNA is obtained by oligo d T cellulose column treatment, followed by reverse transcription-polymerase chain reaction (RT-PCR) method. A cDNA library is prepared by the method described above, and a 15 base or longer probe or primer generated from this library based on the nucleotide sequences shown in SEQ ID NOS: 1-82 corresponds to the target gene. The cDNA can be selected / cloned or PCR amplified. The probe has a sequence complementary to the cDNA sequence, and its length is preferably

It is 20 bases or more, more preferably 30 bases or more, still more preferably 50 bases or more. Moreover, a primer consists of a sense primer and an antisense primer, and the length thereof is preferably 15 to 50 bases, more preferably 17 to 30 bases, and still more preferably 20 to 25 bases. The DNA obtained as described above may be inserted into a commercially available or literature vector, introduced into a prokaryotic or eukaryotic host cell and amplified, or the DNA of interest may be inserted. The PCR can be amplified by PCR. The vector is a vector such as plasmid, phage, plasmid or virus, for example, plasmid such as pBluescript system, pUC system, pBR system. The vector I includes, in addition to the target DNA, a promoter, an origin of replication, a ribosomal binding sequence (or Shine-Dalgarno sequence), a terminator I, a selective marker sequence (eg, a drug resistance gene), a multicloning site, etc. Can be included. Vectors suitable for the host cell are commercially available and it is desirable to use them. Prokaryotic host cells are bacterial cells, and generally include E. coli, B. subtilis, Pseudomonas bacteria, and the like. Eukaryotic host cells also include yeast, fungi, basidiomycetes, animal cells (such as insect cells and mammalian cells), plant cells and the like.

 Methods for introducing the vector into host cells are also known, and include, for example, electroporation, calcium phosphate method, Spheroplast fusion method, protoplast fusion method, microinjection, gene gun method, globulin method and the like.

 The details of the above-mentioned techniques can be found in, for example, J.Sambrook et al., Mol.Cla.I.Cloin.g.L.a.b.r. Ma.

(US); FM A usbel et al, S Hort Protocols, Molecular Biology (3 ed.) AC ompendi um of Memo ds fr om C o rrent Protocolsin, Molecular biology, 19 95, J ohn W iley & Sons , I nc. (US) etc.

Nucleic acids that can be used in the method of the present invention include, as described in (2) above, a nucleic acid that hybridizes with a nucleic acid comprising the nucleotide sequence shown in SEQ ID NO: 1-8 or a nucleotide sequence complementary thereto It is a nucleic acid having a partial sequence of 15 consecutive bases to less than the total number of bases of the nucleic acid. Hydridization is preferably carried out under stringent conditions. It is possible to cause hybridization with a nucleic acid having a sequence identity of at least 90%, preferably at least 95%, more preferably at least 98, most preferably at least 99% to the above-mentioned nucleotide sequence.

 The expression of 82 gene set is measured using the above nucleic acid. Preferred measurement methods include hybridization methods or quantitative PCR methods (such as real-time PCR methods).

 In hybridization, a nucleic acid probe attached or bound on the surface of a solid support is bound to a partially or completely complementary nucleic acid via base pairing to form a double strand. It means the method to form. The solid support includes, for example, a test plate, a test tube, a test piece, an array and the like, and the material is usually a polymer (or plastic, resin), glass or the like. The hybridization used in the method of the present invention may be any of DNA-DNA hybridization, DNA-RNA hybridization or RNA-RNA hybridization, and the method is not limited to this method. For example, microarray method, blotting method such as Northern blot, Southern blot, Northern hybridization, Southern hybridization, in situ hybridization etc. are included.

 In the microarray method, a 15 base or more nucleic acid probe that hybridizes with a gene containing the nucleotide sequence of SEQ ID NO: 1 to 8 or a corresponding nucleic acid (mRNA, cDNA, RNA or RNA) is bound to a support Make and use the microarray (or DNA chip). On the surface of the microarray, it is possible to introduce a spacer crosslinker containing a reactive group for covalently binding a nucleic acid probe. Binding of the array surface to the nucleic acid probe can be performed, for example, through a chemical reaction with light or heat.

The above hybridization enables more specific binding between the nucleic acid probe and the test nucleic acid when hybridization is performed under stringent conditions. Stringent conditions include the conditions defined above, but it is desirable to appropriately determine optimum conditions according to the type of hybridization method. In general, stringency refers to temperature, ion strength of solution, presence of formamide, relative GC content of nucleic acid, incubation time It greatly depends on etc. The temperature can be determined from the temperature range before and after the temperature (Tm) at which 50% of the oligonucleotide dissociates from its complementary strand. The temperature under stringent conditions is usually a temperature within the range of 42 to 68 ° C. High stringency hybridisation conditions include, for example, hybridization with 6 X SSC, 0.1 M EDTA, IX Denhardt's solution, 0.5% SDS at 68 ° C .; 1 MN a C 1, Hybridization at 60 ° C in 0.5% sarkosyl, 30% formaldehyde; washing at 55 ° C in 0.1 X SSC, 0.1% SDS, and so on. Here, 1 X SSC is 15 OmM sodium chloride and 15mM sodium taenoate aqueous solution (p H 7.2).

 The immobilization of the nucleic acid probe is not particularly limited, but a general method, for example, a method of spotting the DNA using a high-density dispenser called spotter or arrayer, or a method of jetting droplets from the nozzle It can be implemented using methods such as methods.

 A nucleic acid such as cDNA, cDNA, or a RNA derived from a lung adenocarcinoma tissue test sample surgically resected from a lung adenocarcinoma patient, such as a C y dye (for example, C r 3 or C y 5) Label with fluorescent substance and hybridize with the probe on the microarray. The fluorescence intensity is read using a laser scanning reader and data is analyzed by a computer.

In the plotting method, after the above-mentioned nucleic acid probe of the present invention is labeled with a radioactive isotope (for example, ³² P and ³⁵ S), a fluorescent substance (fluoresamine, rhodamine, derivatives thereof, C y dye etc.), etc. Hybridization is performed with the nucleic acid such as mRNA, cDNA, RNA, and RNA in the test sample that has been transcribed. The signal is detected using a radiation detector or fluorescence detector and its intensity is measured.

In the quantitative real-time PCR method, the cDNA is paired with cDNA so that cDNA regions prepared from mRNA in lung adenocarcinoma tissue test samples can be used as templates to amplify the regions of each target gene. PCR is performed to detect the amplification product double-stranded DNA in real time. The amplification product is exponentially amplified every cycle and eventually reaches a plateau. Real-time PCR using serially diluted standard samples Conduct the cycle number on the horizontal axis and the PCR width amount on the vertical axis to create an amplification curve for each cycle number. Set the threshold and calculate the Ct value (threshold cycle) at which the threshold and the amplification curve intersect. The Ct value is the number of cycles when a constant amount of PCR product is reached. Since there is a linear relationship between the C 1: value and the amount of initial template DNA, create a calibration curve. The detection of the amplification product can be carried out using a method such as an intercalator method or a fluorescent labeling method. The intercalator method is, for example, a method of monitoring the amount of amplification product utilizing the property that a fluorescent substance is bound between double strands of DNA formed by hybridization. In the fluorescent labeling method, the probe is labeled with a fluorescent substance, and the amount of amplification product is monitored by a fluorescent signal. Among these methods, there is also known a method using a probe having a fluorescent substance at the 5 'end and a quencher at the 3' end.

 Because real-time PCR systems are commercially available, they can be used to perform real-time PCR.

 As conditions for carrying out PCR, the amplification size should be in the range of 80 to 150 bp, the primer size should be in the range of 1 to 25 bases, and the GC content of the entire primer should be 40 to 60. %, The 3 'end of the primer is G or C, there is no complementary sequence inside or between the primers and the primers, and the Tm values of the upstream primer and the downstream primer are within 2 ° C. It is known that designing template-specific primers.

 PCR conditions are, for example, denaturation: 30 to 60 seconds at 92 to 94 ° C .; amino rings: 30 to 60 seconds at 50 to 55 ° C .; elongation: 30 to 60 seconds at 68 to 72 ° One cycle includes 30 to 40 cycles of reaction. The reverse transcriptase is a commercially available enzyme, for example, S.sup.Script.TM. III (Invitrogen), AMV R.E.Transcriptase ^ M-MLV R.E.T. transcriptase (P. rome ga), SYBR P remix E x T aq Carabio etc. can be used.

 In the present invention, the relative expression level difference between relapse and no relapse is 1.1 times or more, preferably 1.5 times or more, more preferably 2.0 times or more.

Measurement of gene expression by protein In the method of the present invention, gene expression can be measured by measuring the amount of a protein encoded by a gene comprising a nucleotide sequence represented by SEQ ID NO: 1-8 or a fragment thereof. An alternative measure of this expression level is an immunoepidemiological method.

 Proteins for producing antibodies can be produced using cDNA cloning and gene recombination technology.

Briefly, for example, a cDNA library is prepared from lung adenocarcinoma tissue, and a cDNA clone is prepared by PCR using a primer synthesized based on each nucleotide sequence of SEQ ID NO: 1 to 82. The obtained c DNA clone can be incorporated into an expression vector and obtained from the cell or culture supernatant by culturing a prokaryotic or eukaryotic host cell transformed or transfected with the vector. At this time, expression vectors can be commercially available or those described in the literature, and examples thereof include plasmids, cosmids, phages and the like. The vector can contain a DNA encoding a target protein, a promoter, a polyadenylation signal, a liposome binding sequence, an origin of replication, a terminator, a selectable marker, and the like. In order to facilitate separation and purification of the polypeptide, a labeled (poly) peptide (eg, mono-galactosidase gene etc.), (His) ₆ — ₁₀ tag, F LAG tag (amino acid sequence: DYKDDDDK (SEQ ID NO: 1) 50)) A protein or an oligopeptide such as GF P (green fluorescent protein) and a corresponding DNA sequence may be contained so as to form a fusion protein with the target protein. Host cells include prokaryotic cells such as bacteria (eg, E. coli, Bacillus subtilis, Pseudomonas bacteria, etc.), yeast (eg, Saccharomyces, Schizosaccharomyces, Pichia etc.), insect cells (eg, S f cells), mammalian cells (Eg CHO, COS, BHK, HEK2 93, NI H3 T3 etc) etc.

 The above-mentioned cDNA cloning and gene recombination techniques are described in, for example, Sammborok et al. (Above), Ausbe 1 et al. (Above) and the like.

The target protein thus obtained can be gel filtration, ion exchange chromatography, affinity chromatography, hydrophobic chromatography, isoelectric focusing, electrophoresis, ultrafiltration, salting out, dialysis etc. alone. Or in combination It can be made.

 The protein encoded by the gene of the present invention comprises the amino acid sequence shown in SEQ ID NO: 83 to 149 (this sequence corresponds to SEQ ID NO: 15 to 2 1, 23 respectively).

Encoded by a nucleotide sequence of -82. ). In addition, the sequences of proteins whose amino acid sequences are not described can be obtained by accessing a database such as G en B ank (US), EMB L (Europe) or the corresponding nucleotide sequences. The amino acid sequence of the target protein can be determined by cDNA cloning and sequencing.

 In the immunological method, mammals such as rabbits, mice, rats, and animals are immunized with the protein obtained by the above-described method or a fragment thereof as an antigen, and antibodies against the antigens are produced and purified. .

 The antibodies include polyclonal antibodies, monoclonal antibodies, anti-peptide antibodies, antibody fragments thereof and the like.

 The polyclonal antibody is used to immunize the animal subcutaneously with an appropriate amount (usually, ^ g order) of antigen, followed by boosting after about 2 to 3 weeks, and collecting blood about 3 weeks to 2 months after the primary immunization, and antiserum. Can be prepared by separating the IgG component containing the desired polyclonal antibody from ammonium sulfate by ion exchange chromatography. At this time, in order to increase the specificity of the antibody for the antigen, the obtained Ig G is bound to a carrier prepared by binding the target protein to a carrier such as cellulose or agarose, and then a high salt buffer is prepared. And then desalted by a method such as dialysis or ultrafiltration to obtain a specific polyclonal antibody. The antibody titer can be measured by a conventional immunoassay, such as enzyme-linked immunosorbent assay (ELISA), radioimmunoassay (RIA), fluorescent antibody assay and the like.

 A monoclonal antibody can be prepared, for example, by the following general method (for example, Hideaki Nagamune and Hiroshi Terada, preparation and characterization of monoclonal antibody, 1 990, Hirokawa Shoten; Yoshi Sado (200 1) “Saving antibody production” Biochemistry 7 3: 1 1 6 3 1 1 6 7 etc.).

The target protein or a fragment thereof is administered subcutaneously to mice or rats (eg, B a 1 b / c mice) as in the preparation of polyclonal antibodies, and at intervals of 1 to 4 weeks, approximately Perform 2-4 boosts. When the antibody titer reaches a plateau, the antigen is injected intravenously or intraperitoneally for final immunization. After 2 to 5 days, antibody-producing cells (eg, spleen cells or lymph node cells) are collected. Then, antibody-producing cells are fused to a myeloma cell line (preferably a hypoxanthine guanine phosphoribosyl transferase (HG PRT) deficient cell line) to generate hybridoma cells, and HAT (hypoxanthine, aminopterin, etc.) is generated. , Thymidine) selection. For cell fusion, antibody-producing cells and a myeloma cell line are mixed in an appropriate ratio in a medium for animal cell culture such as serum-free DMEM, R PM I- 1640 medium, polyethylene polyethylene glycol, etc. It is carried out in the presence of a cell fusion promoter. Confirmation of the target antibody can be performed by the above-mentioned immunoassay. Furthermore, for the growth of hybridomas, hybridomas are administered intraperitoneally to mice, and after hybridomas are allowed to grow, ascites fluid is collected 1 to 2 weeks later. Purification of the antibody can be carried out by appropriately combining methods such as ammonium sulfate fractionation, ion exchange chromatography, affinity chromatography, and generoch chromatography.

 An anti-peptide antibody is an antibody to a linear peptide on the surface of a protein, and can enhance immunological specificity. Such peptides may be used, for example, to estimate the hydrophilic-hydrophobic region of Kyte-D oolittle et al., Probability of being located on the surface of a specific peptide site on a protein molecule according to E mini et al., Polypeptide chain The degree of bending, for example, α-helitas such as Chou-Fasman et al., Β-sheet, secondary structure prediction of a protein indicating a turn, etc. can be estimated using alone or in combination. The deduced peptide can then be synthesized using a peptide synthesizer.

The antibodies described above can be used for detection of target proteins or fragments thereof in lung adenocarcinoma tissue test samples. By preparing antibody microarrays, in which a large number of antibodies are bound on a microarray support, or by spotting a large number of antibodies on a polymer membrane (filter), such as a VDF membrane, or a solid phase, such as a multiwell plate. By adsorbing the antibody to the surface of the protein, it becomes possible to detect or quantify a large number of target proteins at one time. In addition, conventional immunological assays, such as enzyme-linked immunosorbent assay (EL ISA, EI Α), fluorescent antibody, radioactive immunity The target protein or its protein in the test sample by epilepsy assay (RIA), luminescence immunoassay, immunoturbidimetric assay, latex agglutination, latex turbidimetric assay, hemagglutination, particle agglutination, Western blot, etc. Fragments can be detected or quantified.

 When the reaction is carried out on a solid phase, solid phase supports include membranes of polymers such as polystyrene, polycarbonate and polyethylene, multi-well plates, test tubes, test strips (test strips), latex, magnetic particles, etc. Be Immobilization can be physical adsorption or chemical. For chemical binding, for example, a solid phase can be treated with a reagent such as a maleylation reagent, cyanogen bromide or the like to introduce a functional group that reacts with an amino group or the like of a protein onto the solid phase.

Examples of labels include Western Wa rust pel O Kishida over peptidase, enzymes such as alkaline phosphatase, Furuoresin, rhodamine, a fluorescent substance, luciferase system, such as their derivatives, luminescent substances such as luminol, ^{3 2} P, ^{1 2 5} Radioactive isotopes such as I are included. Labeling includes, for example, the Daltal aldehyde method, the maleimide method, the pyridyl disulfide method, the chloramine T method, the Bolton hunter method and the like.

<Composition for predicting postoperative recurrence>

 The present invention is also a composition for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients, wherein the composition comprises the nucleotide sequence of any one of SEQ ID NO: 182. And a plurality of nucleic acids that make it possible to measure the expression of two or more, preferably two or more, of two different genes, including

 (1) Nucleotide sequences as shown in SEQ ID NOs: 1 to 2, nucleotide sequences complementary thereto, or nucleic acids having partial sequences of 15 bases to less than the total number of bases in their nucleotide sequences, and

 (2) A nucleic acid that hybridizes to a nucleic acid containing the nucleotide sequence shown in SEQ ID NO: 1 to 8 or a nucleotide sequence complementary thereto, or alternatively, from 15 bases to less than the total number of bases of the hybridizing nucleic acid. The above composition is characterized in that it is selected from the group consisting of: a nucleic acid having a partial sequence.

The plurality of nucleic acids constituting the composition of the present invention may be selected from 2 or more to 82, preferably 5 or more, of which the composition comprises any of the nucleotide sequences of SEQ ID NOS: 1-82. As long as it makes it possible to measure the expression of different genes, more preferably 10 or more to 82, more preferably 30 or more to 82, even more preferably 40 or more to 82 different genes. It may be any combination of the nucleic acids described in (1) and (2) above.

 The nucleic acid of (1) above is

 (a) The nucleotide sequences shown in SEQ ID NOs: 1-8,

 (b) A nucleotide sequence complementary to the nucleotide sequence of the above (a), and (c) 15 bases or less of the total number of bases continuous in the nucleotide sequence of the above (a) or the nucleotide sequence of the above (b) Preferably, 17 bases or more, 20 bases or more, 30 bases or more, 50 bases or more, 70 bases or more, 100 bases or more, 150 bases or more, 200 bases or more, 300 bases Partial sequence of the above, or more than 400 bases or more than 500 bases to less than the total number of bases,

A nucleic acid having

 The nucleic acid of (2) above is a nucleic acid having a nucleotide sequence that hybridizes with the nucleotide sequence of (a) or (b) above, or alternatively, from 15 consecutive bases in the nucleotide sequence of the nucleic acid Less than the total number of bases, preferably 17 bases or more, 20 bases or more, 30 bases or more, 50 bases or more, 70 bases or more, 100 bases or more, 150 bases or more, 200 bases or more, It is a nucleic acid having a partial sequence of 300 bases or more, 400 bases or more, or 500 bases or more to less than the total number of bases. Preferably, such a nucleic acid is 90% or more, preferably 95% or more, more preferably 98 or more, most preferably 9 9 or more, with the nucleotide sequence of (a) or (b) in (1) above. It is a nucleic acid of a sequence having% identity or more, or a fragment of 15 bases to less than the total number of bases.

Preferred nucleic acids are the nucleic acids of (1) above, in particular the nucleic acids of (1) (c) above. The compositions of the present invention may be in the form of kits, microarrays or primer sets, as well as mixtures in the form of a simple mixture of nucleic acid components. Rather, this latter form is preferred. Here, in the case of a kit, the above nucleic acids can be packaged individually or in the form of a mixture of multiple nucleic acids, and the kit further contains a buffer for hybridization, detection reagents (enzymes, etc.) , Instructions for use Can be included. The microarray and the primer set are as described above.

 The composition of the present invention can be used in a method for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients as described above. The method of the present invention is as described above.

 The present invention is further directed to a composition for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients, wherein the composition comprises the nucleotide sequence of any one of SEQ ID NOS: 1-22. 2 or more, preferably 5 or more to 82, more preferably 10 or more, still more preferably 30 or more, still more preferably 40 or more to 82 or less different genes including A plurality of antibodies that make it possible to measure the protein, wherein the antibody is a protein encoded by the nucleotide sequence shown in SEQ ID NOs: 1 to 2 or a contiguous at least 8 amino acid residue of the amino acid sequence of the protein The above composition is provided with an antibody immunologically reactive with the polypeptide consisting of: and a fragment of the antibody, characterized in that it is selected from the group consisting of:

 The preparation of antibodies is as described above.

The antibodies include, but are not limited to, polyclonal antibodies, monoclonal antibodies, anti-peptide antibodies, etc. as produced by the above method. The type of antibody may be any type, class, or subclass, including, for example, IgG, IgM, IgE, IgD, IgA, and the like. Furthermore, antibody fragments include Fab, (Fab ') ₂ , Fv, scFv and the like.

 The composition of the present invention may be in the form of a kit, an antibody binding support or an antibody microarray, as well as a mixture of the form in which the antibody components are simply mixed. Solid phase supports include membranes of polymers such as polystyrene, polycarbonate and polyethylene, multiwell plates, test tubes, test strips (test strips), latex, magnetic particles and the like. Materials of the array include polymers, glass and the like. As described above, the antibody-containing composition of the present invention can be used in a method for predicting the recurrence of lung adenocarcinoma after surgical resection of lung cancer in lung adenocarcinoma patients. 'The method of the present invention is an immunological method, as described above.

The invention is further illustrated by the following examples, which illustrate the invention. It shall not be restricted by the example. Example

Materials and methods

Patient

 1 9 9 5 1 1 February 1 999 8 7 cases of lung adenocarcinoma of the patient who underwent resection surgery with possibility of healing, and ethics committee of Aichi Cancer Center and Nagoya University The patient's consent was obtained from the Aichi Cancer Center (Nagoya Sakai). None of the cases received adjuvant chemotherapy. The median follow-up period for patients who survived at the final follow-up was 90 months (range: 64-1 08 months). All tumor specimens were embedded in OCT compound (Sakura Seiki Co., Ltd., Tokyo, Japan) and stored at -80 ° C. Two-thirds of the patients (n = 60) were randomly assigned to the training set, and the remaining one-third (n = 27) was considered evaluation set I. Similarly, among the lung adenocarcinoma samples of patients who underwent surgery between February 2002 and February 2004, 30 samples diagnosed as Stage I were used as evaluation set I I (Table 2).

Table 2

RNA isolation

 Frozen tissues of the tumor specimens were subjected to DAROS microdissection under the guidance of a pathologist (Y. Y.) using Giemsa-stained sections. The total RNA is extracted using the RNae ay kit (Qia g e n, Valencia, Calif.), And the DN as e

Treated with I The amount of RNA is Nano D rop (R) ND — 1 000 UV — V s Spectrophotometer (Nano D rop T cnnologies, Winnore De La Wa The properties of the RNA were observed using an Ag i ent 2 1 00 Pio analyzer and shown as complete RNA numbers (RIN) (Agilent Technologies, Palo Alto, Calif.). A large amount of common standard RNA was prepared using 20 lung cell lines showing all major lung cancer histotypes.

EGFR, from p 5 3 and K over _r the as expression profiles and mutation status of acquiring 5 00 ng of total RNA, Moroni one murine leukemia virus reverse transcriptase

Double-stranded cDNA was synthesized using poly dT primers containing (Agilent T e cho n o i g i s, Palo Anoret, Calif.) And a T 7 promoter. A pair of L o w R N A B o U o s e x n. JU ι n 6 a r Ά. M p 1 ι f i c i a t i n kit (Agilent T e c o n o l i o i s) was used to generate cRNA and labeled with C y 3 or C y 5. Cy 5-sample c RNA and C y 3-common standard cDNA with 4 thousand different probes Whole Human G ome oligo DNA microarray kit (G41 1 2 F, Agilent Technology) I was given a noize idea. After hybridisation and washing, the slides were scanned using an Agi l ent DNA microarray scanner (G250 5 B, Ag l l t t e n o g i o s). Expression data were acquired using Ag ile t F e t r E r E ソ フ ト ウ ェ ア E E E t ソ フ ト ウ ェ ア t ソ フ ト ウ ェ ア ソ フ ト ウ ェ ア ソ フ ト ウ ェ ア ソ フ ト ウ ェ ア 9. software 9.5. 1 (A g i l t t c t c n n n i i i i s s). The mutational status of EGFR, p53 and K-ra as has already been reported for the same set of patients (Takyu et al., JC〇, 2006). Duke data set is downloaded from Accession No. GS E 359 3 "d ene e x p s s s s o om om n i b u s h t t p: / / www. N c b i. N l m.

(Mon.) For clinical information such as the situation and cell type, D D u I I I I s s t t f-------S S S S S c h h h h h h h h h h h h h www www www www www www www www. Of the 45 adenocarcinoma samples, 3 9 adenocarcinoma samples annotated with clinical information were used for this analysis.

Statistical method First, at least 90% of the samples were selected for expression levels above background. Weighted voting algorithm, which is an established technique of supervised classification, to identify recurrence related signatures using signals that have passed this filtering criterion (each weighted value is calculated as the signal-to-noise ratio and is already detailed The method described (Yanagisa wa b J. N atl. C incer. I nst. 9 9: 85 8-6 7, 200 7) was used, in a nutshell, overtraining specific to the training data set. In order to obtain a classifier that can generally be used without doing, we perform 100 100 random interactions in selecting a probe signal to identify a recurrence free signal. — The fo 1 d cross-validation method was used (AM Molinaro et al., Bioinformatics 2 1: 3 30 1-7, 2005) Therefore, the whole process of classifier construction is a fixed number of predicted genes. Repeat 1 000 times for each classifier with K ap 1 an -Obtained relapse-associated signatures, overall survival and relapse-free survival in a set for evaluation of 27 patients, using Meier's survival curve and Cox proportional hazards model analysis (Stata purgeon 7.0) The hazard ratios of death and relapse are analyzed respectively The P values presented in the multivariate Co x model were obtained in a similar ratio test All statistical tests were two-tailed C Average linkage hierarchy between genes and cases using the luster program ('MB E isen et al., Pr. N. Ac. d. Sci. USA 95: 1 48 63-8, 1 9 98) At this time, the display used the program Trie V iew (MB E isen et al., Above).

Result

Identification of relapse-associated signatures

 The present inventors constructed and verified a gene expression signature for recurrence and death of lung adenocarcinoma patients surgically treated without information leakage between training set / evaluation set I (Figure

1). First, the expression profile data of 87 cases were classified into 60 training sets and 27 evaluation sets I. Second, out of the 60 training sets, 21 patients who died within 5 years of surgical resection have general high-risk signatures in their tumors due to recurrence, whereas all follow-up 5 years or more without showing signs of recurrence during the up period It was hypothesized that the 28 patients who survived did not have any of the aforementioned signs in the tumor. One remaining patient in the training set was excluded from the process of identifying recurrence-related signatures because of the unclear grade of the tumor. Of the excluded patients, 5 survive for more than 5 years with some signs of recurrence during follow-up, 5 survive for 5 or more years and die from cancer, and 1 has no evidence of recurrence. I died.

 Total genome microarray (Agi 1 ent; Who 1 e Humen G en ome (4 x 44 K) O ligo M icroarray Kit it; No. G 4 1 1 2 F) 4 thousand pieces Of the probes, 23, 828 probes passed the first filtering criteria to select informative probes and were classified based on SN metrics to distinguish between patients who died from recurrence and those who were cured by surgery. Were used to identify the best relapse-associated signature. The training error for each model to which increasing numbers of prediction probes were applied was calculated by 10_f o d cross validation, and the calculation was repeated 1000 times with the new randomly divided data set. A set of 1000 independent sets of up to 200 predicted probes was selected to construct a classifier using recurrence related signatures. As a result, it can be seen that the use of 82 prediction probes causes a minimum learning error (Fig. 2A), and the independent 1, 00 consisting of 82 prediction probes.

The 82 most frequently shared probes among the 0 sets were identified as recurrence related signatures (hereinafter referred to as "RRS-82". See Table 1 above). RR

The S-82 signature seems to be able to identify patients with a very poor prognosis when considering all stages or stage I (see figure for recurrence free survival).

2 B and 2 C, see Figure 7 for overall survival).

 Comparison and comparison of RR S-82 and various genetic changes

Lung cancer is known to conceal various genetic changes that may affect the clinical nature of patients with hereditary changes. Therefore, we analyzed the association of RRS-82 with the presence of EGFR, K-ras or p53 gene mutations. However, no statistically significant association was found between them. Furthermore, the prognostic significance of the genetic change was also analyzed using the same patient sample set, thereby constructing a prognostic classifier using RR S-82. With the presence of EGFR mutations No association between postoperative prognosis was observed (Figure 3A). There was no statistically significant difference in overall survival between patients with and without the K-ras mutation (Figure 3B) or the P53 mutation (Figure 3C).

Verification of RRS-82 using evaluation set

 In order to evaluate the buccal properties of RRS-82, the discrimination ability of RRS-82 was analyzed using an independent evaluation set I of lung adenocarcinoma 27 samples. Results obtained using Evaluation Set I indicated that RRS-82 can distinguish high-risk and low-risk patients with relapse and death. Recurrence-free survival is significantly different between the two groups, with a median recurrence-free survival of 29 patients in the high-risk group of 29 days compared with 108 months in the low-risk group of 18 patients. (P. 0. 0003, Fig. 4A). The proportion of recurrence-free patients in the high-risk and low-risk groups was 38% and 78% at two years, respectively. Of note, all eight patients with RRS-82 high risk signs have relapsed during the five-year follow-up period, while RRS-82 shows low risk signs. Sixty-seven percent of the patients showed no signs of recurrence 5 years after surgery. Among patients in the high-risk group, overall survival after surgical resection was significantly lower than those in patients in the low-risk group (P = 0.226, Figure 4B). In a multivariate Co X proportional hazards model, RRS—82 is an independent and significant prognostic factor that predicts both relapse-free survival (P = 0. 03) and overall survival (P = 0. 03) I understand. In addition, also in the multivariate C 0 X proportional hazards model when evaluation sets I and II are added, RRS-8 2 has no recurrence (P <0. 001) and overall survival (P = 0. 005). It was found to be an independent significant prognostic factor that predicts both (Table 3).

Table 3 Multivariate COX regression analysis

 Evaluation Set I (n = 27) Evaluation Set I + II (n = 57)

 Bad / Good Hazard Ratio 95% CI * P Hazard Ratio 95% Gf P

> 61 / <61 1.60 0.44-4.28 0.594 0.91 -2.02 0.817 Sex Sity ^ Annual Male / Female 1.30 0.42-4.03 0.651 1.19 0.54-2.60 0.668 ^ Tete Age Regained Age Renewal Growth--/ / 1.92 0.64-5.77 0.246 2.00 0.84-4.72 0.1 15 † ^, high risk / low risk 6.12 1.87-20.1 0.003 4.92 2.17-1 1.2 0.001 0.001

 > 61 / <61 1.84 0.54-6.23 0.329 1.21 0.50-2.91 0.668 male / female 1.31 0.40-4.31 0.653 1.33 0.55-3.19 0.526

Π-ΙΠ / Ι 1.72 0.52-5.65 0.370 2.15 0.84-5.47 0.106 Risk / low risk 3.79 1.14-12.6 0.030 1.60 1.48-8.77 0.005

, * 95% CI: 95% confidence interval

Usefulness of RR S-82 in stage I patients

We further evaluated the discrimination of RR S- 82 in the subgroup of adenocarcinoma cases with early or more advanced disease using 16 samples of Stage I in Evaluation Set I. Stage I patients predicted to be at high risk based on RR S-82 experienced relapses within 5 years, and died during follow-up (see Figure 4 C (P = 0 for recurrence free survival). 0008), Overall survival is shown in Figure 4D (P = 0. 043) Both cases by log rank test). In addition, using 30 samples of evaluation set II, which consisted of all stage I, was confirmed to be statistically significant (Fig. 4 E, 4 F). Furthermore, in the analysis of 46 samples in which 16 samples of evaluation set I and 16 samples of evaluation stage I were combined with 30 samples of evaluation set II, the discrimination power of RR S- 82 was confirmed to be statistically significant ( Figure 4G, 4H) ₀

 Of the stage II-III patients, recurrence-free and overall survival tended to be better for patients with low risk RR S-82 signature compared to those with high risk signaling. Was not significant (Figures 8A and 8B).

 From another aspect, the present inventors have also diagnosed RR S-82 and TNM (T: spread of primary tumor; N: presence or absence of regional lymph node metastasis; M: presence or absence of distant metastasis) (eg lung cancer · esophagus) We investigated the relationship between cancer, breast cancer, and handling regulations (excerpt) Kinbara Publishing. Among patients in the low-risk group, the K ap 1 an-Me ier curve for recurrence-free survival (Fig. 5 A) and the K ap 1 an-Me ier curve for overall survival (Fig. 9 A) are both pathological diseases. There was a tendency to show a moderate relationship with the stage of (P = 0.15 for recurrence-free survival, P = 0.18 for overall survival). On the other hand, in patients in the high-risk group, recurrence-free survival (Fig. 5B) and overall survival curve (Fig. 9B) were both unrelated to the stage of the pathological disease (P = for recurrence-free survival) 0. 7 1, P = 0. 8 7 for overall survival. As a result of the above-described test of the present invention, relapses were observed in 86% and 100% of the cases judged as risk R RS-2 in the training set and evaluation set I respectively. This is described by Duke et al. (DG B e er, N a t. Med. 8: 8 16-2

4, 2002) report high risk of recurrence in 69% and 79% of the ACOSOG and CALGB efficacy assessment data sets, respectively. It can be seen that the accuracy (or precision) of the judgment is significantly improved compared to the measured results.

Checking the General Applicability of RRS-2 Using Independent Data Sets

 The robustness of RRS-2 in predicting the viability of lung adenocarcinoma is further evaluated using other completely different data sets for 39 samples of lung adenocarcinoma and the gene expression dataset from Duke University. Verified. As only 46 genes out of RR S-82 on Agi 1 ent platform correspond to newer version of Affymetrix platform 1, form Duke University dataset probe (Table 4), RRS- 82 It was not possible to apply the classifier directly using. Therefore, we analyzed based on the expression profiles of 46 corresponding genes by hierarchical clustering without teacher. Twenty-nine samples of adenocarcinomas showed significantly different post-operative viability (P = 0. 028, Figure 6 B) Clearly clustered into two different subsets (Figure 6 A). In addition, 31 genes overlapping with RRS-8 2 were extracted using the gene expression data set of Michigan University, and a prediction model was constructed (Fig. 6C). The gene expression data of Memorial S loo n K ettering Cancer Center 1 04 samples were analyzed using the modelle, and the high risk group and low risk group identified by this model were statistically related to recurrence free survival. Significant differences were observed. (P = 0. 0003) (Figure 6 D). These results indicated that the genes contained in RR S-82 have generality (general applicability) related to risk assessment for recurrence.

Table 4

-LZ-

ZS6 L0 / 800 Zdr / 13d 01 ^ orchid 600Z OAV

Industrial applicability

According to the method of the present invention, the possibility of postoperative recurrence of lung adenocarcinoma patients can be predicted with extremely high accuracy. This method allows prediction with high accuracy even in the case of stage I patients. This will make it possible to develop a treatment plan early in the postoperative period to improve the survival rate for patients at high risk of recurrence. The present invention makes it possible to predict post-operative recurrence of adenocarcinoma in lung adenocarcinoma patients, so that it is possible to make a treatment plan suitable for patients after surgery. Thus, the present invention can contribute to the medical industry for the treatment of lung adenocarcinoma.

 All publications, patents and patent applications cited herein are incorporated herein by reference in their entirety.

Claims

The scope of the claims

1. A method for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients, which is selected from the group consisting of genes comprising the nucleotide sequences of SEQ ID NO: 1 to 82. 2-82 Measuring the expression of a gene set of individual genes in a biological sample derived from said patient, wherein: SEQ ID NOs: 1-5, 8-9, 12, 14, 15. 15, 17. The expression of a gene containing a nucleotide sequence of any of 2-21, 23-28, 30-33, 35-59, 61-68, 70-82 is relatively postoperative compared with the case without recurrence If it is high, alternatively, expression of a gene containing a nucleotide sequence of any one of SEQ ID NOs: 6 to 7, 10 to 11, 13, 16, 18, 22, 29, 34, 60, 69 The method is characterized in that the recurrence of lung adenocarcinoma is predicted to be high when it is relatively low compared to the case without recurrence.

 2. The method according to claim 1, wherein the number of genes of the gene set is 5-82.

3. The method according to claim 1, wherein the measurement of the expression of the gene set is performed by measuring the presence or amount of a nucleic acid or protein corresponding to the gene.

 4. The method according to claim 1, wherein the measurement of the expression of the gene set is performed by a hybridization method or a quantitative PCR method.

 5. The method according to claim 4, wherein the hybridization method is a microarray method or a blotting method.

 6. The method according to claim 4 or 5, wherein the hybridization method is performed using a nucleic acid that hybridizes with the RNA transcript, cRNA, aRNA or cDNA derived from the gene.

 7. The method according to claim 1, wherein the measurement of the expression of the gene set is performed by an immunological method.

 8. The method according to claim 7, wherein the immunological method is performed using an antibody against a protein encoded by the gene or a fragment thereof.

9. A composition for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in lung adenocarcinoma patients, wherein the composition comprises at least two nucleotide sequences of any one of SEQ ID NOs: 1-82 Contains multiple nucleic acids that make it possible to measure the expression of different genes The nucleic acid is

 (1) Nucleotide sequences as shown in SEQ ID NOs: 1 to 2, nucleotide sequences complementary thereto, or nucleic acids having partial sequences of 15 bases to less than the total number of bases in their nucleotide sequences, and

 (2) A nucleic acid hybridizing with a nucleic acid comprising the nucleotide sequence shown in SEQ ID NOs: 1 to 2 or a nucleotide sequence complementary thereto, or a portion from 15 contiguous bases to less than the total number of bases of the hybridized nucleic acid A nucleic acid having a sequence, characterized in that it is selected from the group consisting of

 10. The composition according to claim 9, wherein the composition is in the form of a kit, a microarray or a set of primers.

 The composition according to claim 9, wherein the nucleic acid enables measurement of the expression of 5 or more to 82 or less genes.

 A composition according to claim 9, which is for use in the method according to claim 1.

 A composition for predicting recurrence of lung adenocarcinoma after surgical removal of lung cancer in patients with lung adenocarcinoma, the composition comprising a nucleotide sequence of any one of SEQ ID NOS: 1 to 82. A protein encoded by any of the nucleotide sequences set forth in SEQ ID NOs: 1-8 or comprising a plurality of antibodies which make it possible to measure the expression products of two or more different genes. An antibody which immunologically reacts with a polypeptide consisting of at least 8 consecutive amino acid residues of the amino acid sequence, and a fragment of the antibody.

 14. The composition of claim 13, wherein the composition is in the form of a kit.

The composition according to claim 13, wherein the antibody enables measurement of an expression product of 5 or more to 82 or less genes.

 A composition according to claim 13, which is for use in the method according to claim 1.