WO2007055553A1

WO2007055553A1 - Diagnostic methods of lung cancer and its subtypes by cgh

Info

Publication number: WO2007055553A1
Application number: PCT/KR2006/004786
Authority: WO
Inventors: Yeun-Jun Chung
Original assignee: Catholic University Industry Academy Cooperation Foundation
Priority date: 2005-11-14
Filing date: 2006-11-14
Publication date: 2007-05-18
Also published as: KR100759288B1; KR20070051101A

Abstract

The present invention relates to methods to diagnosis of lung cancer and its subtype by using the method of screening the aberration in whole genome. Specifically, by applying genome wide array CGH to the extracted DNA from lung cancer and identifying specific aberration to lung cancer and its subtype, the present invent gives correct diagnosis of lung cancer and its subtype. Also, the present invention provides the methods of using prognostic marker of specific genetic aberration in lung cancer progress.

Description

DIAGNOSTIC METHODS OF LUNG CANCER AND ITS SUBTYPES BY CGH

Technical Field

The present invention relates to methods of diagnosis of lung cancer and its subtypes by using screening methods of genetic variation in whole genomic genes, and methods of using specific genetic aberrations in lung cancer as a prognostic marker for lung cancer progress.

Background Art

Lung cancer is the most common incident form of malignancy and also the leading cause of cancer death worldwide (1, 2). The primary lung cancer is classified into four major histological subtypes; squamous cell carcinomas (SqCs) , adenocarcinomas

(AdCs), large cell and small cell lung cancers. The former three classes, grouped as non-small cell lung cancer (NSCLC) , comprise up to almost 80% of the total incidence of lung cancer. Among NSCLC, SqCs and AdCs are two major subtypes. Histologically different subtypes are known to behave differently in clinical courses, and may require individual therapeutic approaches.

Some genomic aberrations in tumors have been suggested as prognostic markers or as identifiers of genes to target for therapy or prevention (3, 4) . Likewise in other solid tumors, chromosomal aberrations have been thought to be critical molecular events in the pathogenesis of lung cancer (5, 6) . However, clinically applicable screening tools or prognostic markers are still underdeveloped. Since the lack of efficient screening methods and therapy accounts for the poor outcome of lung cancer, genome-wide assessment of aberrations is expected to help in developing more accurate diagnostic and therapeutic strategies .

For this reason, previous cytogenetic studies using conventional comparative genomic hybridization (CGH) or fluorescence in situ hybridization have focused on the identification of chromosomal aberrations associated with NSCLC. Recurrent genomic alterations were observed in NSCLC, including the gains of partial or whole chromosomal arms on Iq, 3q, 5p, and 8q along with the losses on 3p, βq, 8p, 9p, 13q, and 17p (7- 11) - However, the resolution of conventional CGH being approximately 10 Mb is not high enough (12), for the precise identification of sub-microscopic changes. As accumulating evidence suggests that genomic dosage changes contribute to tumorigenesis by altering expression levels of cancer-related genes (13, 14), more detailed analyses with sufficient resolution are required.

Technical Problem

An object of the present invention is to provide methods for diagnosing lung cancer and its subtypes, screening marker and prognostic marker of lung cancer progress by identifying genetic aberrations by applying genome wide comparative genomic hybridization (CGH) to extracted DNAs from tissues of lung cancers.

Technical Solution

The present invention relates to identification and preparation of genomic map in a new part of nucleic acids associated with generation of cancer and tumor. The present invention provides a part of p arm in human chromosome 10 and a part of q arm in human chromosome 16 characterized by acquisition through CGH in a whole human genome; a part of q arm in human chromosome 13 characterized by deletion through CGH in a whole human chromosome for diagnosis of lung cancer; and the p21 part of human chromosome 6 and the q31 part of human chromosome 19 characterized by the aberration of MAR-G gene through CGH in a whole human genome for diagnosis of lung cancer.

In one aspect, the present invention provides a diagnosis kit for lung cancer containing the mentioned parts of chromosome. The said diagnosis kit can be DNA chip.

The present invention provides a part of p arm in chromosome 3 and a part of Y chromosome characterized by acquisition with CGH in a whole genome chromosome for diagnosis of squamous cell carcinomas; a part of chromosome 12 characterized by acquisition with CGH in a whole genome chromosome for squamous cell carcinomas which is a subtype of lung cancer; and a part of q arm in chromosome 6 characterized by deletion with CGH in a whole genome chromosome for diagnosis of adenocarcinomas which is a subtype of lung cancer.

In yet another aspect, the present invention provides a diagnosis kit for lung cancer containing the mentioned parts of chromosome. The said diagnosis kit can be DNA chip.

In yet another aspect, the present invention provides a pair of primers for polymerase chain reaction amplifying a part or whole nucleic acids of a chromosome part selected from the groups comprising a part of p arm of chromosome 10, a part of q arm of chromosome 16, a part of q arm of chromosome 13, p21 part of chromosome 6 and q31 part of chromosome 19 of human. The detection system includes detecting whether the polymerase chain reaction occurs or not. The PCR primer pair can be selected from the group comprising a part of p arm of chromosome 10, a part of q arm of chromosome 16, a part of q arm of chromosome 13, p21 part of chromosome 6 and q31 part of chromosome 19 of human

In yet another aspect, the present invention provides methods for screening a presence of an amplicon in a nucleic acids sample of human. The first step of the methods is providing a nucleic acid sample of human cell and probe. The second step is contacting the probe with the nucleic acids of human wherein the probe is contact with the nucleic acids in stringent condition that it selectively contacts with the nucleic acids of human to be hybridization complex. The third step is detecting the hybridization complex. As a one example, the nucleic acids can be genomic DNA of human nucleic acids and separated from a cell of lung cancer. The detection step includes determination of copy number of the amplicon.

In one method of the present invention, the said probe can be attached to a surface of a solid and the attached probe can be one member of nucleic acids array. The human nucleic acids can be labeled with a detectable material and the detectable material can be fluorescein or Texas red. In other method, the probe can be labeled with a detectable component. The method according to present invention provides methods that nucleic acids is prepared from an reference cell and the human genomic nucleic acid is laid to contact with the nucleic acids from reference cell before or simultaneously an probe contacts with the human nucleic acids. Furthermore, the present invention provides a method that Cot-1 DNA contacts with the human genomic DNA before a probe contacts with the human nucleic acids. The present invention provides nucleic acids probe for detecting the presence of amplicons in a sample of human genome nucleic acids, containing nucleic acids specifically binding to a sequence located in the space of chromosome parts selected from the group comprising a part of p arm of chromosome 10, a part of q arm of chromosome 16, a part of q arm of chromosome 13, p21 part of chromosome 6 and q31 part of chromosome 19 of human.

In yet another aspect, the present invention provides PCR primer amplifying a part of whole nucleic acids of a chromosome part selected from the group comprising a part of p arm of chromosome 10, a part of q arm of chromosome 16, a part of q arm of chromosome 13, p21 part of chromosome 6 and q31 part of chromosome 19 of human.

The present invention provides a kit for detecting the presence of amplicons in human nucleic acids. The kit contains a space containing a probe, the probe contains nucleic acids specifically hybridize to the nucleic acids of the chromosome parts selected from the group comprising a part of p arm of chromosome 10, a part of q arm of chromosome 16, a part of q arm of chromosome 13, p21 part of chromosome 6 and q31 part of chromosome 19 of human.

In yet another aspect, the present invention provides a kit for detecting the presence of amplicons in human nucleic acids. The kit contains a space containing a probe, the probe contains nucleic acids specifically hybridize to the complementary nucleic acids of the chromosome parts selected from the group comprising a part of p arm of chromosome 10, a part of q arm of chromosome 16, a part of q arm of chromosome 13, p21 part of chromosome 6 and q31 part of chromosome 19 of human. In another aspect, in a kit detecting the presence of amplicons in human nucleic acids sample, the present invention provides a device containing a pair of PCR primers amplifying a part or whole of the nucleic acid of chromosome parts selected from the group comprising a part of p arm of chromosome 10, a part of σ arm of chjrpmo≤ome 16, a oart._of α arm .of_ chXnmosome 13.

The said probe can be cloned human nucleic acids, which can be attached to a solid surface. The said device can include an indicative material notifying that more than 2 times copy number observation of amplicons is a prognosis or diagnosis of cancer or tumor.

The present invention provides a method of diagnosis of lung cancer and a subtype of lung cancer by using the method of screening genetic aberration in whole genomic. Specifically, the screening method of genetic aberration is array CGH. More specifically, the screening method of genetic aberration uses a 1 Mb resolution array CGH. In the course of screening, an DNA extracted from normal tissue and cancer tissue is directly used without genomic amplification, analyzed statistically a repeated genetic aberration, and used to diagnose lung cancer and its subtype according to the analyzed results. The array CGH method and statistical analysis method is described in examples in detail .

In a diagnosis of lung cancer subtype according to genetic aberration, the two subtype of non-small cell lung cancer, SqCs and AdCs are known to behave differently with respect to their location, growth patterns or prognosis. According to present invention, the comparison of chromosomal aberrations between two subtypes of NSCLC has identified a number of genomic imbalances specific to each subtype; gains of 3q and 12p as well as losses of 3p and Y are specific to SqCs, while gain of 6p is specific to AdCs. Thus, it is possible to correctly diagnose subtypes of a non-small cell lung cancer by using specific genomic aberration identified by the present invention.

The present invention provides a method to use specific aberration to lung cancer as a prognostic marker. By aligning the genomic alterations, 17 MARs with variable genomic sizes were defined. Nine MAR-Gs on Ip, 2p, βp, 8p, 19p, and 2Op along with the MAR-Ls on 5q and 2Oq are thought to be novel features in lung cancer and could be used for accurate diagnoses of lung cancer. We also identified the significant correlation between 3 pairs of MARs suggesting their possible collaborative roles in the tumorigenesis of NSCLC. Among these genomic alterations, MAR-G on Ip32.3 and MAR-L on 13q21.1 showed significant correlations with clinical features (Table 5) . For example, MAR- L on 13q21.1 was correlated with early onset (younger than 60 yrs) and advanced stage (stage III and IV) , suggesting this alteration was acquired in later stage of tumor progression. MAR-G on Ip32.3 showed association with lymph node negative cases, which implies its implication for metastasis of lung cancer. Therefore, using these factors, we identified to elucidate lung cancer pathogenesis or to develop new prognostic markers for lung cancer.

The term "amplicons" used in this specification and claim is a site of genome nucleic acids that is related cancer when it varies its copy number.

The term "non-small cell lung cancer (NSCLC) " used in this specification and claims means a lung cancer except small cell lung cancer from primary lung cancer. Specifically, the primary lung cancer is classified into four major histological subtypes; squamous cell carcinomas (SqCs) , adenocarcinomas (AdCs) , large cell and small cell lung cancers. The former three classes, grouped as non-small cell lung cancer (NSCLC) , comprise up to almost 80% of the total incidence of lung cancer. Among NSCLC, SqCs and AdCs are two major subtypes.

The term "array CGH" used in this specification and claims means array comparative genomic hybridization.

The term "MAR" used in this specification and claims means minimally altered region. Although high copy number changes are relatively rare, single copy number changes were widespread among the analyzed cases. To select highly recurrent ones among such changes, minimally altered region (MAR) was defined as commonly altered segment recurring for at least u cases.

The term "MAR-G and MAR-L" used in this specification means genomic gain and loss of MRS respectively.

Advantageous Effects

Because SqCs and AdCs are known to behave differently with respect to their location, growth patterns or prognosis, specific genomic alterations identified in this study might be useful to clarify the subtype-specific pathogenesis. In this invention, the comparison of chromosomal aberrations between two subtypes of NSCLC has identified a number of genomic imbalances specific to each subtype; gains of 3q and 12p as well as losses of 3p and Y are specific to SqCs, while gain of 6p is specific to AdCs. Brief Description of the Drawings

FIG. 1 shows genome-wide copy number alterations in 50 cases of non-small cell lung cancers. (A) Genomic profiles of 29 SqCs (above) and 21 AdCs (below) . Fifty NSCLC cases are represented in individual lanes with corresponding sample numbers in two subtypes. Intensity ratios are schematically plotted in different color scales reflecting the extent of genomic gains (red) and losses (green) as indicated in the reference color bar. Total 2987 BAC clones were ordered (x-axis) according to the map positions and the chromosomal order from lpter to Yqter. (B) The genome-wide frequencies of all significant gains (> 0.2 of intensity ratio, above) and losses (< -0.2 of intensity ratio, below) for each clone are shown for 29 cases of SqCs (black, above the center) and 21 cases of AdCs (grey, below the center) , respectively. The boundaries of individual chromosome and the location of centromere are indicated by vertical bars and dotted lines below the plots, respectively. FIG. 2 shows individual profiles of high copy number changes. (A) The intensity ratio profile for the 92 BAC clones in the long arm of chromosome is shown for AdCs-I. The x-axis represents the map position of corresponding clone according to the UCSC human genome (May. 2004 Freeze) and the intensity ratios were assigned to y-axis. The schematic presentation of cytogenetic bands as well as map position is shown below the plot. (B) Localized high level of amplification in the long arm of chromosome 17. (C) The intensity ratio profiles of the long arm of chromosome 3 for SqCs9 and SqCs22 are shown along the 3q21-q29. (D) A homozygous deletion on chromosome 10q23.31.

FIG. 3 shows examples of minimal regions of genomic gain or loss. Minimally altered region (MAR) was defined as commonly altered segment recurring for at least 7 cases. Each sample is represented as an individual lane. MARs are schematically shown as colored box below the cytogenetic bands, red, genomic gains; green, genomic loss; black, no changes (A) Two minimally altered regions with different genomic sizes in 16 samples. While the proximal one is as large as ~30 Mb, distal one is less than 1 Mb. (B) Minimal region of chromosomal losses on chromosome 5 common to 7 NSCLC samples in chromosome 5.

FIG. 4 shows Kaplan-Meier survival curves. The survival curves for the cases with or without specific genomic changes are plotted using Kaplan-Meier method. The chromosomal changes associated with relatively poor survival are presented with the significance level; gain of 1Op (A) and lβq (B), loss of 9p (C) and 13q (D), and MAR-Gs on 6p21 (E) and 19q31 (F).

Mode for Carrying Out the Invention

Hereinafter, the present invention will be described in further detail with reference to examples. It is to be understood, however, that these examples are for illustrative purposes only and are not to be construed to limit the scope of the present invention.

Example 1 (1) Materials and Methods Study Materials

Frozen tissues were obtained from 50 patients with NSCLC, who underwent surgical operation at Dankook University Hospital, Cheonan, Korea. Tissue collection and full procedure of genetic analysis were performed under the approval of Institutional Review Boards, Kangnam St. Mary's Hospital, The Catholic University of Korea. The 50 NSCLC cases were histologically classified into SqCs (29 cases) and AdCs (21 cases) and the tumor staging was performed according to the standard TNM classification in the American Joint Committee on Cancer guidelines. Two histological subtypes of SqCs and AdCs were compared for general characteristics such as sex, age, tumor size, nodal status, and recurrence of the disease according to the post operational computational tomography follow-up. Using Pearson's χ² test and independent t-test, no significant difference was observed between two subtypes of NSCLC except for sex (p=0.029) .

Tissue Preparation and Microdissection

Frozen sections were prepared of 10 μm thickness on a gelatin coated slide using 2800 FRIGOCUT (Reighert-Jung, Germany) . The sections were fixed at 100% ethanol and rehydrated before H&E staining. After H&E staining, paired tumor area (more than 60% of tumor cells) and normal tissue area were selected from the same patient microscopically and dissected manually.

Microdissected tissues were transferred into the cell lysis buffer (1% proteinase-K in TE buffer) and DNA was extracted.

Extracted DNA was purified using a DNA purification Kit (Solgent,

Korea) and used for dye labeling reactions.

Array Comparative Genomic Hybridization and Image Analysis The construction of large insert clone array and hybridization were performed as described by Fiegler et al (19) with some modification. We used human large insert clone arrays with 1 Mb resolution across the whole genome printed by the Sanger Institute Microarray Facility (Table 1). Each sequence of clone ID and related information could check URl : http : //www . ensemble . org/index . html . [Table 1 ] Clone ID of each chromosome for microarray

(continued from Table 1)

(continued from Table 1]

Six hundred nanogram of each tumor DNA was labeled with Cy3-dCTP and same amount of control DNA from normal lung tissue was labeled with Cy5-dCTP by random priming (BioPrime Array CGH Genomic Labeling System, Invitrogen, USA) . After 4 hour incubation at 37 "C labeled genomic DNA was purified using a DNA purification column supplied in the Kit. Flow through DNA was precipitated with 140 μg of human Cot-1 DNA (Roche, Germany) . The DNA pellet was dissolved in 50 μl of hybridization buffer (50% formamide, 10% dextran sulfate, 0.1% Tween 20, 2X SSC, and 10 mM Tris-HCl, pH 7.4) containing 600 μg of yeast t-RNA. After 10 minutes of denaturation at 72 °C probe DNA was incubated at 37^°Cfor 1 hour prior to being applied onto the prehybridized BAC array slide. BAC array slide was pre-hybridized for 2 hours at 37 ^°C with 90 μl of hybridization buffer containing 540 μg of herring sperm DNA and 90 μg of human Cot-1 DNA. Prehybridization and hybridization procedures were performed as described previously (19) . Briefly, a rubber cement ring was applied around the array to make a reaction chamber. After application of the probe solution onto the array, the slide was then incubated in a light-tight humid chamber on a shaker for 2 days at 37 ^°C Slides were washed serially in solution 1 (2X SSC, 0.1% SDS) for 10 minutes at 37 ^°C in solution 2 (0.1X SSC, 0.1% SDS) at room temperature for 10 minutes, in solution 3 (0.1X SSC) at room temperature for 1 minute for three times, and once in distilled water at room temperature for 10 seconds. Finally, slides were spin-dried for 3 minutes at 1000 rpm. Arrays were scanned using GenePix 4100A scanner (Axon Instruments, USA) and the image was processed using GenePix Pro 6.0.

Data Processing, Normalization and Mapping of BAC Clones

Normalization and re-aligning raw array CGH data were performed using the web-based array CGH analysis interface, ArrayCyGHt (http://genomics.catholic.ac.kr/arrayCGH/) (20). In brief, raw signal intensities and relevant chromosome mapping information of every individual data point were uploaded to ArrayCyGHt and normalized by Print-tip Loess method. The normalized fluorescence intensity ratios in log₂ scale were automatically re-aligned according to linear chromosomal order. Mapping of large insert clones was done according to the genomic location in the UCSC genome browser (May, 2004 Freeze) . In total, 2987 BAC clones out of initial 3014 clones were successfully mapped both on the UCSC genome browser and on the Sanger Institute genome browser and subsequently processed. All the genomic coordinates such as cytogenetic bands or gene positions described in this invention are based on the same version of the human genome available in the UCSC genome browser.

Data Analysis for Chromosomal Alterations

To set the cutoff value for chromosomal alterations of individual large insert clones, we performed four independent series of normal hybridization (three sex-matched and one male versus female hybridizations) as a control. The average SD value of the control batch was 0.081, therefore the cutoff value for the aberrant copy number changes were set to be ±0.2 in Iog₂ scale, higher than twofold of control SD. Entire chromosome arm gain or loss was determined as previously described (21) . Regional copy number change was defined as DNA copy number alteration limited to part of a chromosome. High-level amplification of clones was defined when their intensity ratios were higher than 1.0 in Iog₂ scale, and vice versa for homozygous deletion. The boundary of copy number change was assigned to the halfway between the two neighboring clones. Definition of Minimally Altered Regions

To define minimally altered regions of chromosomal gain or loss, we used CGH-Miner (http://www-stat.stanford.edu/~wp57/CGH-

Miner/) to smooth the raw intensity ratio and to identify the breakpoints of chromosomal alterations (22) . Four series of normal hybridizations were combined as controls and the analysis was performed with recommended program parameter. The significant gains or losses reported by the program were directly used for subsequent aligning procedure. Minimal regions of chromosomal gains and losses were determined by altered segments recurring for at least 7 samples.

Statistical Analysis

Significance of the differences in chromosomal arm changes between SqCs and AdCs was tested by two-sided Fisher's exact test. The correlations between the recurrent genetic changes on minimally altered regions were assessed using univariate pair- wise Pearson's correlation. Because of multiple comparisons, stepdown Sidak method was used to adjust the overall level of significance. In this case, the pairs of genetic changes on the same chromosomal arm were excluded for the concordance analysis. The correlations between genetic alterations and clinicopathological parameters were analyzed using two-sided Fisher's exact test. For this purpose, the changes on minimally altered regions as well as those of chromosomal arms were included in the analysis. For the comparison, 4 kinds of clinicopathological parameters were treated as categorical variables such as early onset (diagnosed before the age of 60) versus late onset (≥ 60 yrs) , early stage (stage I and II) versus advanced stage (stage III and IV) , lymph node negative versus positive, and the presence versus absence of the recurrence of disease. Kaplan-Meier method was used for survival analysis and the difference between survival curves was compared using the log-rank test in univariate model. Significant factors among the variables were further assessed in multivariate analysis using Cox proportional hazard model. In all statistical analyses, p value less than 0.05 was considered significant.

(2) Results

Comprehensive Profiling of Genomic Alterations in NSCLC The overall genomic alterations observed in the 50 NSCLC cases (29 SqCs and 21 AdCs) are illustrated in Fig. IA. Array- CGH signal intensity ratio (Iog2 scale) data of the 50 NSCLC can be downloaded in our web site (http://lib.cuk.ac.kr/micro/CGH/lung.htm). Genomic alterations are either localized involving only small number of clones or extended even encompassing an entire chromosomal arm. Substantial amount of genomic alterations were observed in most NSCLC cases. The average numbers of gained and lost clones per sample were 415.8 (13.9%) and 364.8 (12.2%), respectively, which shows that genomic gains slightly outnumber genomic losses, at least in these NSCLC cases.

Although genome-wide chromosomal profiles show that the NSCLC cases are heterogeneous in terms of extent and location of genetic aberrations, the frequency plots of chromosomal gains and losses show that chromosomal changes are not randomly distributed. In fact, many of them clustered in different places throughout the 50 NSCLC cases (Fig. IB) . To identify recurrent genomic changes at the individual chromosome level, we first investigated the changes of whole chromosomal arm frequently observed in the 50 NSCLC samples. Eight chromosomal arms, 19q (40% of the 50 NSCLC cases), 2Oq (26%), 22q (24%), 3q (22%), 19p (22%), Iq (20%), 5p (20%), and 17q (20%), were frequently gained

(Table 2). Similarly, six chromosomal arms, Yp (52%), Yq (46%), 9p (42%), 3p (26%), 17p (24%), and 4q (20%), were frequently lost.

Some of these chromosomal alterations were distributed differently in SqCs and AdCs. The average numbers of clones with aberrations were 437.7 (14.7%; gain) and 405.0 (13.6%; loss) for SqCs, whereas AdCs showed less frequent copy number alterations, 385.6 (12.9%; gain) and 309.2 (10.4%; loss), but the difference was not significant. We then performed statistical analysis to identify the differential chromosomal changes between two subtypes. As a result, six chromosomal changes were found to be differentially distributed between SqCs and AdCs, i.e. gain of 3q, 6p, and 12p along with the loss of 3p, Yp, and Yq (Table 2) . Gains of 3q and 12p as well as losses of 3p, Yp and Yq are specific to SqCs, while gain of 6p is specific to AdCs.

High-Level Amplification and Homozygous Deletion In total, 98 large insert clones showed high-level amplifications at least in one case and they clustered in 36 different genomic segments. The identified amplicons are summarized in Table 3. The genomic size of amplicons ranged from 0.31 to 14.78 Mb with the maximum intensity ratio of 2.56 in Iog2 scale. Although most amplicons were small, amplicons larger than 5 Mb were observed in Iq, 3q, 7p, and 19q. One of them, as large as 6.8 Mb, is a good example of high copy number gain (Fig. 2A) . The putative oncogenes AFlQ and TPM3 as well as a potential tumor marker of lung cancer, CTSS, are located in this amplicon. Small amplicons were usually well-defined by a small number of clones with high-level amplification. For example, a high copy number change with intensity ratio of 2.49 (Iog2) was observed on 17q (Fig. 2B) . This 1 Mb-sized amplicon, composed of three adjacent clones, contains a putative oncogene MLLT6. Other highly amplified regions across various chromosomes contain putative cancer-related genes listed in Table 3.

Most of the amplicons (30/36 cases; 83.3%) were observed in one or two NSCLC cases. The exceptions were two large amplicons, one in 3q2β from 8 SqCs cases and the other in 3q27-q29 from 6 SqCs cases. Figure 2C illustrates the largest amplicons in 3q26 (SqCs22) and in 3q27-q29 (SqCs9) . The first one, as large as 14.78 Mb, harbors several putative oncogenes such as EVIl, SKIL, ECT2, and PIK3CA. The other one sized 10.31 Mb contains putative oncogenes of BCL6 and HES. In all 50 NSCLC cases, only three homozygous deletions were identified (Table 3) . Among them, homozygous deletion of RPIl- 765C10 (10q23.31) harbors tumor suppressor gene PTEN (Fig. 2D).

Minimal Regions of Recurrent Genomic Gains and Losses Although high copy number changes are relatively rare, single copy number changes were widespread among the analyzed cases. To select highly recurrent ones among such changes, minimally altered region (MAR) was defined as commonly altered segment recurring for at least 7 cases. MAR-G and MAR-L represent genomic gain and loss of MAR, respectively. In total, we have identified 13 MAR-Gs and 4 MAR-Ls (Table 4). The examples of MAR-G and MAR-L are illustrated in figure 3. The MAR-G on Ip3β-p34, observed in 12 cases, contains several putative cancer-related genes such as PKKl, FGR, LCK, and MYCLl. Also, in 9 cases, another MAR-G on Ip32.3 contains putative cancer-related gene, TTC4 (Fig. 3A) . The MAR-L on 5q, in 7 samples, includes several putative tumor suppressor genes such as IRFl, CDKLl, and RAD50 (Fig. 3B) .

Correlation between Minimally Altered Regions

Pairwise correlation analysis between MARs was performed to see whether such genomic changes appeared concordantly in a set of NSCLC cases. For the comparison, all of the possible combinations between 17 MARs were considered except for the pairs on the same chromosomal arms. Significantly positive relationships were observed for 3 pairs of MARs. The MAR-G on 19ql3.1 was found to be associated with MAR-Gs on 6p21.3-p21.1 (r=0.549, p=0.0482) and 19pl3.2-pl3.1 (r=0.672, p=0.0016), respectively. Another significant association was found between two MAR-Gs on 8pl2.2-pl2.1 and 8qll.2-12.1 (r=0.610, p=0.0370).

Association between Genomic Aberrations and

Clinocopathological Characteristics

Four kinds of clinicopathological variables (age, stage, lymph node, and recurrence) were analyzed for the association with all the genomic alterations identified (Table 4). Significant associations were observed for the MAR-L on 13q21 with early onset (younger than 60 yrs; p=0.0345) and advanced stage (stage III and IV; p=0.0451). The gain of chromosomal arm Xq, were also found to be significantly associated with advanced stage (p= 0.0461) and disease recurrence (p=0.0210). MAR-G on Ip32 and chromosomal gain of Yp were significantly associated with lymph node negative (p=0.00454 and p=0.03468, respectively).

To assess the prognostic impacts of the genetic aberrations identified, survival analysis was performed. Using Kaplan-Meier methods, we identified 6 genetic aberrations significantly associated with relatively poor survival (Fig. 4); chromosomal gain of 1Op (p=0.0091; 6%) and 16q (p=0.0262; 6%), loss of 9p (p=0.0082; 42%) and 13q (p=0.0019; 18%), and the MAR-Gs on 6p21 and 19ql3 (p=0.0265 and 0.0295, respectively). In multivariate analysis, only three factors remained significantly associated with poor survival; male (Hazard ratio (HR) =16.67, 95% confidence interval (CI) = 1.12-131.32, p=0.0075), MAR-G on 6p21

(HR = 4.376, 95% CI = 1.55-12.35, p=0.0053), and chromosomal loss of 9p (HR = 3.851, 95% CI = 1.17-8.65, p=0.001i;

[Table 2] The frequency of chromosomal changes : Ln NSCLC

NOTE: The frequency (%) of entire change of the corresponding chromosomal arms throughout 50 NSCLC cases is shown with those of 29 SqCs and 21 AdCs in parensthesis, respectively. The frequency difference between SqCs and AdCs was statistically analyzed by two-sided Fisher's exact test and boxed in grey when significant (p < 0.05). ^*, p-value is provided only in case of significance.

[Table 3] Genomic segments representing high copy number changes in NSCLC

NOTE: The boundary of each high copy number of change is defined by corresponding insert clone. Cytogenetic band and map position of clones are based on the public genome database (UCSC genome, May 2004 Freeze) . ^a Amp, amplification; HD, homozygous deletion. In case of more than two observed cases, the boundary of high copy number change was defined as the most extended set of clones, so they were not necessarily overlapping.

[Table 4] Minimal regions of DNA copy number changes

NOTE: Gain and loss in the first column represent MAR-G and MAR-L, respectively. *The frequency represents the number of samples with the corresponding genomic changes in two kinds of NSCLC subtypes.

[Table 5] Correlations between genomic alterations and clinicopathological features

Age

MAR-L on 13q21 Early onset (< 60 yrs) Late onset (> 60 yrs) Total p-vahiQ

+ 6 1 7 0.0345 16 27 43 Total 22 28 50

Stage

Early Advanced

MAR-L on 13q21 (stage I and II) (stage III and IV) Total /7-value

+ 1 6 7 0.0451

25 18 43 Total 26 24 50

Gain of Xq

+ 0 4 4 0.0461 26 20 46 Total 26 24 50

Lymph node status

MAR-G on Ip32 Negative Positive Total p-value

+ 7 2 9 0.0044 10 31 41 Total 17 33 50

Gain of Yp

+ 3 0 3 0.0347 14 33 47 Total 17 33 50

Recurrence

Gain of Xq Negative Positive Total p-value

+ 0 4 4 0.0461 30 16 46 Total 30 20 50

(3) Result analysis

We have applied high-resolution array CGH to 29 SqCs and 21 AdCs for the systematic analysis of the complex genomic alterations in NSCLC. Adopting microdissection to remove non- tumor tissues and using paired normal tissue samples as a control, we uncovered highly recurrent or unique genomic aberrations specific to NSCLC. Microdissected DNA was hybridized to arrays without performing whole genome amplification, which enabled us to avoid the possible bias of random amplification. The frequently gained or lost chromosomal changes identified in this invention are in good agreement with previously published results (7-11), including the loss of the Y chromosome in male patients (5, 6) . It is notable that copy number alterations on small chromosomes such as 19, 20 and 22 were much more frequent in our results. This might be due to the differences in analytic methods used between the studies. However, it is more likely to reflect the potential of array CGH to improve low resolution of conventional CGH as described in a recent study (23) . In addition, the comparison of chromosomal aberrations between two subtypes of NSCLC has identified a number of genomic imbalances specific to each subtype; gains of 3q and 12p as well as losses of 3p and Y are specific to SqCs, while gain of βp is specific to AdCs. Because SqCs and AdCs are known to behave differently with respect to their location, growth patterns or prognosis, specific genomic alterations identified in this study might be useful to clarify the subtype-specific pathogenesis (9, 24).

Genomic profiles of individual NSCLC cases in this invention seem to be as heterogeneous as the stereotypic genomic changes in other solid tumors (4, 25). We focused on the genomic changes that seem to have plausible biological implications in lung cancer such as high-level amplifications or homozygous deletions as well as recurrent alterations of copy number gain or loss. Although it is still controversial whether genomic dosage changes are consistently correlated with the expression level of corresponding genes, good agreement between dosage change and expression level have been observed in recent studies (26, 27) including an animal trisomy model (28) . Therefore, it is reasonable to assume that such genomic alterations are likely to contribute to the tumor pathogenesis by altering expression profiles of critical cancer-related genes (3, 4).

The genomic size of high copy number changes ranged from 0.31 to 14.78 Mb, most of them were less than 5 Mb-sized. Genomic changes smaller than 5 Mb are thought to be novel undetectable by conventional CGH studies (12) . The amplification on 3q was the most frequent high copy change in this study. In previous studies, the role of this change has been implicated in the pathogenesis of invasive carcinoma (29) and activation of a gene in this region, PIK3CA which involves phosphoinositide-3-OH kinase (PI-3K) signaling pathway, might contribute to the tumorigenesis of SqCs (24) . We have identified amplicons around 3q2β harboring PIK3CA. Interestingly, a high-level amplification on 7q22 including PIK3CG that is also involved in PI-3K signaling, was observed in one SqCs case (SqCs25) . Although this gene was previously suggested as a putative tumor suppressor gene (30) , recent evidence indicates that activated form of PIK3CG may contribute to the development of malignancy (31) . Except for the amplifications of 3q, most high-level amplifications were usually observed in one case or two. They might reflect the individual nature of genomic evolution for the respective NSCLC cases. Proto-oncogenes such as REL, EGFR, MLL, DDX6, YESl, JUND, and HKRl as well as the genes such as PTPRF (32), NUDTl (33), and TYMS (34) showing correlations between a high-level of expression and clinical outcome of lung cancer are located in the amplicons identified in this study. Although the clinical implications of these genes as molecular targets for diagnosis or therapy have been suggested, they still need to be validated in future studies.

A homozygous deletion on 10q23.31 observed in one SqCs case contains well known tumor suppressor gene, PTEN. It is known to encode lipid phosphatase which negatively controls signaling proteins activated in PI-3K pathway such as protein kinase B/Akt. These data indicate a potential role of PI-3K signaling pathway in NSCLC pathogenesis. Recent studies suggest that inactivation of PTEN is partly caused by methylation change, which may explain the lack of the recurrent deletions of this locus in other samples.

Notably, recurrent gains and losses confined to minimal genomic regions were successfully identified using high- resolution array CGH in this invention. By aligning the genomic alterations, 17 MARs with variable genomic sizes were defined. Nine MAR-Gs on Ip, 2p, 6p, 8p, 19p, and 2Op along with the MAR- Ls on 5q and 2Oq are thought to be novel features in lung cancer, which indicates the advantage of high-resolution mapping of genomic alterations. In contrast to high copy number changes that were largely limited to a few samples, single copy changes were found in much more samples, which can be indicative of the shared mechanism common to the earlier stage of NSCLC. We also identified the significant correlation between 3 pairs of MARs suggesting their possible collaborative roles in the tumorigenesis of NSCLC. However, in terms of gene expression, the functional consequence of single copy number alterations is difficult to predict because of complex nature of such aberrations .

Among these genomic alterations, MAR-G on Ip32.3 and MAR-L on 13q21.1 showed significant correlations with clinical features (Table 5). For example, MAR-L on 13q21.1 was correlated with early onset (younger than 60 yrs) and advanced stage (stage III and IV) , suggesting this alteration was acquired in later stage of tumor progression. MAR-G on Ip32.3 showed association with lymph node negative cases, which implies its implication for metastasis of lung cancer.

Survival analysis revealed six genetic alterations were significantly associated with poor survival in the univariate model. Among the 6 alterations, loss of 9p has been suggested to be associated with poor survival in previous studies about lung cancer (37) . However, the other 5 chromosomal alterations have not been studied for the association with lung cancer yet. Therefore, these chromosomal alterations seem to be a novel genetic indicator for the prognosis of NSCLC. Among 6 genomic alterations, two MAR-Gs on 6p21 and 19ql3 contain cancer-related genes such as PIMl, CCND3 (both in 6p21) and HKRl (19ql3) , all of which have been investigated in lung cancers as well as various kinds of tumors. For example, the high expression of HKRl after the administration of platinum drugs has been indicative of acquisition of resistance of lung cancer to chemotherapy (38). Intriguingly, these two MARs appeared concordantly (p=0.0482) as described above. It suggests their collaborative role for the prognosis of lung cancer. In subsequent multivariate analysis using Cox-regression model, only three factors, MAR-G on 6p21, chromosomal loss of 9p and male factor, remained as independent indicators of poor survival. However, it must be taken into consideration that treatment factors may affect the survival. We reviewed the treatment regimens for all 50 cases. Although we could not do subgroup analysis by postoperative treatment regimens due to the limited number of samples, all the operations were done by the same surgeon and preoperative chemotherapy has not been provided to any patient.

Using whole-genome array CGH strategy, we successfully identified chromosomal aberrations including novel MARs, high- level amplifications, or deletions specific to NSCLC as well as previously identified ones.

Industrial Applicability

Therefore the novel genomic alterations identified in this invention might be useful to clarify the subtype-specific pathogenesis and to diagnose its subtype correctly.

In this study, survival analysis revealed six genetic alterations were significantly associated with poor survival in the univariate model. Among the 6 alterations, loss of 9p has been suggested to be associated with poor survival in previous studies about lung cancer (37) . However, the other 5 chromosomal alterations have not been studied for the association with lung cancer yet. Therefore, these chromosomal alterations seem to be a novel genetic indicator for the prognosis of NSCLC. By aligning the genomic alterations, 17 MARs with variable genomic sizes were defined. Nine MAR-Gs on Ip, 2p, βp, 8p, 19p, and 2Op along with the MAR-Ls on 5q and 2Oq are thought to be novel features in lung cancer and could be used for accurate diagnoses of lung cancer. We also identified the significant correlation between 3 pairs of MARs suggesting their possible collaborative roles in the tumorigenesis of NSCLC. Among these genomic alterations, MAR-G on Ip32.3 and MAR-L on 13q21.1 showed significant correlations with clinical features (Table 5) . For example, MAR-L on 13q21.1 was correlated with early onset (younger than 60 yrs) and advanced stage (stage III and IV) , suggesting this alteration was acquired in later stage of tumor progression. MAR-G on Ip32.3 showed association with lymph node negative cases, which implies its implication for metastasis of lung cancer. Therefore, using these factors, we identified to elucidate lung cancer pathogenesis or to develop new prognostic markers for lung cancer.

Using whole-genome array CGH strategy, we successfully identified chromosomal aberrations including novel MARs, high- level amplifications, or deletions specific to NSCLC as well as previously identified ones. The novel genomic alterations identified in this invention will give a clue for further studies to elucidate lung cancer pathogenesis or to develop new prognostic markers for lung cancer and its subtype.

< References>

1. Boyle P, Ferlay J. Cancer incidence and mortality in Europe, 2004. Ann Oncol 2005; 16: 481-8. 2. Jemal A, Tiwari RC, Murray T, et al. Cancer statistics, 2004. CA Cancer J Clin 2004 ; 54 : 8-29.

3. Liotta L, Petricoin E. Molecular profiling of human cancer. Nat Rev Genet 2000; 1 : 48-56.

4. Albertson DG, Collins C, McCormick F, Gray JW. Chromosome aberrations in solid tumors. Nat Genet 2003; 34 : 369-76.

5. Balsara BR, Testa JR. Chromosomal imbalances in human lung cancer. Oncogene 2002; 21 : 6877-83.

6. Testa JR, Liu Z, Feder M, et al. Advances in the analysis of chromosome alterations in human lung carcinomas. Cancer Genet Cytogenet 1997 ; 95 : 20-32.

7. Berrieman HK, Ashman JN, Cowen ME, Greenman J, Lind MJ, Cawkwell L. Chromosomal analysis of non-small-cell lung cancer by multicolour fluorescent in situ hybridisation. Br J Cancer 2004; 90 : 900-5. 8. Luk C, Tsao MS, Bayani J, Shepherd F, Squire JA. Molecular cytogenetic analysis of non-small cell lung carcinoma by spectral karyotyping and comparative genomic hybridization. Cancer Genet Cytogenet 2001; 125 : 87-99.

9. Petersen I, Bujard M, Petersen S, et al. Patterns of chromosomal imbalances in adenocarcinoma and squamous cell carcinoma of the lung. Cancer Res 1997; 57 :2331-5.

10. Pei J, Balsara BR, Li W, et al. Genomic imbalances in human lung adenocarcinomas and squamous cell carcinomas. Genes Chromosomes Cancer 2001; 31 : 282-7. 11. Sy SM, Wong N, Lee TW, et al . Distinct patterns of genetic alterations in adenocarcinoma and squamous cell carcinoma of the lung. Eur J Cancer 2004;40: 1082-94. 12. Kallioniemi A, Kallioniemi OP, Sudar D, et al . Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 1992; 258 : 818-21.

13. Tay ST, Leong SH, Yu K, et al. A combined comparative genomic hybridization and expression microarray analysis of gastric cancer reveals novel molecular subtypes. Cancer Res 2003;63:3309-lβ.

14. Mukasa A, ϋeki K, Matsumoto S, et al . Distinction in gene expression profiles of oligodendrogliomas with and without allelic loss of Ip. Oncogene 2002; 21 : 3961-8.

15. Pinkel D, Segraves R, Sudar D, et al. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays . Nat Genet 1998;20 : 207-11.

16. Albertson DG, Pinkel D. Genomic microarrays in human genetic disease and cancer. Hum MoI Genet 2003; 12 : R145-52.

17. Solinas-Toldo S, Lampel S, Stilgenbauer S, et al. Matrix-based comparative genomic hybridization: biochips to screen for genomic imbalances. Genes Chromosomes Cancer 1997;20:399-407. 18. Mantripragada KK, Buckley PG, de Stahl TD, Dumanski JP. Genomic microarrays in the spotlight. Trends Genet 2004;20:87-94.

19. Fiegler H, Carr P, Douglas EJ, et al. DNA microarrays for comparative genomic hybridization based on DOP-PCR amplification of BAC and PAC clones. Genes Chromosomes Cancer 2003;36:361-74.

20. Kim SY, Nam SW, Lee SH, et al . ArrayCyGHt : a web application for analysis and visualization of arrayCGH data. Bioinformatics 2005; 21 : 2554-5. 21. Hackett CS, Hodgson JG, Law ME, et al. Genome-wide array CGH analysis of murine neuroblastoma reveals distinct genomic aberrations which parallel those in human tumors. Cancer Res 2003;63:526β-73.

22. Wang P, Kim Y, Pollack J, Narasimhan B, Tibshirani R. A method for calling gains and losses in array CGH data. Biostatistics 2005; 6: 45-58. 23. Schraders M, Pfundt R, Straatman HM, et al. Novel chromosomal imbalances in mantle cell lymphoma detected by genome-wide array-based comparative genomic hybridization. Blood 2005;105:168β-93.

24. Massion PP, Kuo WL, Stokoe D, et al. Genomic copy number analysis of non-small cell lung cancer using array comparative genomic hybridization: implications of the phosphatidylinositol 3-kinase pathway. Cancer Res 2002; 62 : 3636- 40.

25. Stoler DL, Chen N, Basik M, et al. The onset and extent of genomic instability in sporadic colorectal tumor progression. Proc Natl Acad Sci U S A 1999; 96: 15121-6.

26. Virtaneva K, Wright FA, Tanner SM, et al. Expression profiling reveals fundamental biological differences in acute myeloid leukemia with isolated trisomy 8 and normal cytogenetics. Proc Natl Acad Sci U S A 2001; 98 : 1124-9.

27. Mukasa A, Ueki K, Matsumoto S, et al . Distinction in gene expression profiles of oligodendrogliomas with and without allelic loss of Ip. Oncogene 2002;21 : 3961-8.

28. Kahlem P, Sultan M, Herwig R, et al . Transcript level alterations reflect gene dosage effects across multiple tissues in a mouse model of down syndrome. Genome Res 2004 ; 14 : 1258-67.

29. Heselmeyer K, Schrock E, du Manoir S, et al. Gain of chromosome 3q defines the transition from severe dysplasia to invasive carcinoma of the uterine cervix. Proc Natl Acad Sci U S A 1996;93:479-84.

30. Sasaki T, Irie-Sasaki J, Horie Y, et al. Colorectal carcinomas in mice lacking the catalytic subunit of PI (3) Kgamma. Nature 2000; 406: 897-902.

31. Barbier M, Attoub S, Calvez R, et al. Tumour biology. Weakening link to colorectal cancer? Nature 2001; 413 : 796.

32. Harder KW, Saw J, Miki N, Jirik F. Coexisting amplifications of the chromosome Ip32 genes (PTPRF and MYCLl) encoding protein tyrosine phosphatase LAR and L-myc in a small cell lung cancer line. Genomics 1995; 27 : 552-3.

33. Kennedy CH, Pass HI, Mitchell JB. Expression of human MutT homologue (hMTHl) protein in primary non-small-cell lung carcinomas and histologically normal surrounding tissue. Free Radic Biol Med 2003; 34 : 1447-57.

34. Shintani Y, Ohta M, Hirabayashi H, et al. New prognostic indicator for non-small-cell lung cancer, quantitation of thymidylate synthase by real-time reverse transcription polymerase chain reaction. Int J Cancer 2003;104:790-5.

35. Cantley LC, Neel BG. New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3-kinase/AKT pathway. Proc Natl Acad Sci U S A 1999;96:4240-5.

36. Soria JC, Lee HY, Lee JI, et al . Lack of PTEN expression in non-small cell lung cancer could be related to promoter methylation. Clin Cancer Res 2002; 8 : 1178-84.

37. Tomizawa Y, Adachi J, Kohno T, et al. Prognostic significance of allelic imbalances on chromosome 9p in stage I non-small cell lung carcinoma. Clin Cancer Res 1999; 5:1139-46.

38. Oguri T, Katoh O, Takahashi T, et al. The Kruppel- type zinc finger family gene, HKRl, is induced in lung cancer by exposure to platinum drugs. Gene 1998 ; 222 : 61-7.

Claims

1. A kit for diagnosis of lung cancer containing CGH markers of q arm part of chromosome 13, p21 part of chromosome 6, q31 part of chromosome 19, and optionally p arm part of chromosome

10 or q arm part of chromosome 16; and having means for identification of the loss of q arm part of chromosome 13, gain of minimally altered region (MAR) of p21 part of chromosome 6, gain of MAR in q31 part of chromosome 19, and optionally gain of p arm part of chromosome 10 or gain of q arm part of chromosome

16.