JP2004248502A - Method for evaluating malignancy of cancer - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for evaluating the malignancy of cancer, especially gastric cancer by which the malignancy of the cancer can simply and rapidly be evaluated with high reliability and to provide an array and a kit used therefor. <P>SOLUTION: The method for evaluating the malignancy of the gastric cancer comprises carrying out the following steps. (1) a step of measuring ≥8 types of genes varying the expression level according to the malignancy of the gastric cancer in a test specimen and (2) a step of evaluating the malignancy of the gastric cancer based on the expression level of each gene obtained in (1). The array for evaluating the malignancy of the gastric cancer is obtained by immobilizing a nucleic acid corresponding to the ≥8 types of genes or its fragment in a predetermined region on each support. The kit for evaluating the malignancy of the gastric cancer comprises a primer and/or a probe for detecting an mRNA expressed from the ≥8 types of genes or its fragment. The kit for evaluating the malignancy of the gastric cancer comprises an antibody to a polypeptide encoded with the ≥8 types of genes or its fragment or the fragment. <P>COPYRIGHT: (C)2004,JPO&NCIPI

Description

【００３４】
【表２】

【００３５】
に列挙される１０５種の遺伝子並びに後述の表６及び７にさらに列挙される５１種の遺伝子の発現量を測定し、癌の悪性度の評価を行なうことができる。
【００３６】
本発明の癌、特に胃癌の悪性度の評価方法においては、正確な評価を得る観点から、上記遺伝子のうち、本発明において発現量を測定される遺伝子は、少なくとも８種であり、好ましくは少なくとも２０種であり、より好ましくは３０種以上であり、さらに好ましくは４０種以上であり、特に好ましくは５０種以上であることが望ましく、評価方法における操作の容易性、特に後述のＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により遺伝子発現のパターン解析を行なう場合の操作の容易性の観点から、１００種以下であることが望ましい。
【００３７】
「胃癌の悪性度に応じて発現量が変化する遺伝子」の具体例としては、好ましくは、アポトーシスインヒビター サーバイビン〔ａｐｏｐｔｏｓｉｓ ｉｎｈｉｂｉｔｏｒ ｓｕｒｖｉｖｉｎ（ジーンバンク登録番号：Ｕ７５２８５）〕遺伝子、
Ｉ型コラーゲンα−２〔ｃｏｌｌａｇｅｎ ｔｙｐｅ Ｉ α−２（ジーンバンク登録番号：Ｊ０３４６４）〕遺伝子、
ＩＩＩ 型コラーゲン プロ−α−１鎖〔ｃｏｌｌａｇｅｎ ｔｙｐｅ ＩＩＩ ｐｒｏ−ａｌｐｈａ−１ （ジーンバンク登録番号：Ｘ１４４２０）〕遺伝子、
フィブロネクチン前駆体（ＦＮ）〔ｆｉｂｒｏｎｅｃｔｉｎ ｐｒｅｃｕｒｓｏｒ （ＦＮ）（ジーンバンク登録番号：Ｘ０２７６１）〕遺伝子、
ＭＭＰ−１；消化管コラゲナーゼ〔ｍａｔｒｉｘ ｍｅｔａｌｌｏｐｒｏｔｅｉｎａｓｅ １； ｉｎｔｅｓｔｉｎａｌ ｃｏｌｌａｇｅｎａｓｅ（ジーンバンク登録番号：Ｍ１３５０９）〕遺伝子、
ＭＭＰ−７；子宮マトリリシン〔ｍａｔｒｉｘ ｍｅｔａｌｌｏｐｒｏｔｅｉｎａｓｅ ７； ｍａｔｒｉｌｙｓｉｎ， ｕｔｅｒｉｎｅ （ジーンバンク登録番号：Ｚ１１８８７）〕遺伝子、
Ｕ−プラスミノーゲン活性化因子〔ｕｒｏｋｉｎａｓｅ−ｔｙｐｅ ｐｌａｓｍｉｎｏｇｅｎ ａｃｔｉｖａｔｏｒ ｐｒｅｃｕｒｓｏｒ （ＥＣ ３．４．２１．７３）； Ｕ−ｐｌａｓｍｉｎｏｇｅｎ ａｃｔｉｖａｔｏｒ （ＵＰＡ）； ｐｌａｓｍｉｎｏｇｅｎ ａｃｔｉｖａｔｏｒ， ｕｒｏｋｉｎａｓｅ（ジーンバンク登録番号：Ｘ０２４１９） 〕遺伝子、
ウロキナーゼ型プラスミノーゲン活性化因子受容体〔ｕｒｏｋｉｎａｓｅ−ｔｙｐｅ ｐｌａｓｍｉｎｏｇｅｎ ａｃｔｉｖａｔｏｒ ｒｅｃｅｐｔｏｒ； ＧＰＩ−ａｎｃｈｏｒｅｄ ｆｒｏｍ ｐｒｅｃｕｒｓｏｒ （Ｕ−ＰＡＲ）；ｍｏｎｏｃｙｔｅ ａｃｔｉｖａｔｉｏｎ ａｎｔｉｇｅｎ ＭＯ３； ＣＤ８７ ａｎｔｉｇｅｎ （ジーンバンク登録番号：Ｕ０９９３７）〕遺伝子、
アルファ−２−マクログロブリン〔ａｌｐｈａ−２−ｍａｃｒｏｇｌｏｂｕｌｉｎ （ａｌｐｈａ−２−Ｍ） （ジーンバンク登録番号：Ｍ１１３１３）〕遺伝子、
β型血小板由来増殖因子受容体前駆体〔ｂｅｔａ ｐｌａｔｅｌｅｔ−ｄｅｒｉｖｅｄ ｇｒｏｗｔｈ ｆａｃｔｏｒ ｒｅｃｅｐｔｏｒ ｐｒｅｃｕｒｓｏｒ （ＰＤＧＦＲ−ｂｅｔａ）； ＣＤ１４０Ｂ ａｎｔｉｇｅｎ （ジーンバンク登録番号：Ｊ０３２７８）〕遺伝子、
ＢＩＧＨ３ （ジーンバンク登録番号：Ｍ７７３４９）遺伝子、
ＸＶＩＩＩ 型コラーゲンα−１鎖〔ｃｏｌｌａｇｅｎ ｔｙｐｅ ＸＶＩＩＩ ａｌｐｈａ−１ （ジーンバンク登録番号：ＡＦ０１８０８１）〕遺伝子、
成長ホルモン依存性インスリン様増殖因子結合タンパク質〔ｇｒｏｗｔｈ ｈｏｒｍｏｎｅ−ｄｅｐｅｎｄｅｎｔ ｉｎｓｕｌｉｎ−ｌｉｋｅ ｇｒｏｗｔｈ ｆａｃｔｏｒ−ｂｉｎｄｉｎｇ ｐｒｏｔｅｉｎ （ジーンバンク登録番号：Ｍ３５８７８）〕遺伝子、
肝癌由来増殖因子〔ｈｅｐａｔｏｍａ−ｄｅｒｉｖｅｄ ｇｒｏｗｔｈ ｆａｃｔｏｒ （ＨＤＧＦ） （ジーンバンク登録番号：Ｄ１６４３１）〕遺伝子、
インスリン様増殖因子結合タンパク質２〔ｉｎｓｕｌｉｎ−ｌｉｋｅ ｇｒｏｗｔｈ ｆａｃｔｏｒ ｂｉｎｄｉｎｇ ｐｒｏｔｅｉｎ ２ （ＩＧＦＢｐ２） （ジーンバンク登録番号：Ｘ１６３０２）〕遺伝子、
ＩＶ型コラゲナーゼ〔ｔｙｐｅ ＩＶ ｃｏｌｌａｇｅｎａｓｅ （ジーンバンク登録番号：Ｍ５５５９３）〕遺伝子、
ｓｐａｒｃ前駆体（オステオネクチン）〔ｓｐａｒｃ ｐｒｅｃｕｒｓｏｒ （ｓｅｃｒｅｔｅｄ ｐｒｏｔｅｉｎ ａｃｉｄｉｃ ａｎｄ ｒｉｃｈ ｉｎ ｃｙｓｔｅｉｎｅ； ｏｓｔｅｏｎｅｃｔｉｎ） （ＯＮ）； ｂａｓｅｍｅｎｔ ｍｅｍｂｒａｎｅ ｐｒｏｔｅｉｎ ＢＭ−４０ （ジーンバンク登録番号：Ｊ０３０４０）〕遺伝子、
トロンボスポンジン２前駆体〔ｔｈｒｏｍｂｏｓｐｏｎｄｉｎ ２ ｐｒｅｃｕｒｓｏｒ（ジーンバンク登録番号：Ｌ１２３５０）〕遺伝子、
コンドロイチン硫酸プロテオグリカンコアタンパク質２〔ｖｅｒｓｉｃａｎ ｃｏｒｅ ｐｒｏｔｅｉｎ ｐｒｅｃｕｒｓｏｒ； ｌａｒｇｅ ｆｉｂｒｏｂｌａｓｔ ｐｒｏｔｅｏｇｌｙｃａｎ； ｃｈｏｎｄｒｏｉｔｉｎ ｓｕｌｆａｔｅ ｐｒｏｔｅｏｇｌｙｃａｎ ｃｏｒｅ ｐｒｏｔｅｉｎ ２； ｇｌｉａｌ ｈｙａｌｕｒｏｎａｔｅ−ｂｉｎｄｉｎｇ ｐｒｏｔｅｉｎ （ＧＨＡＰ） （ジーンバンク登録番号：Ｕ１６３０６）〕遺伝子、
エンボプラキン〔ｅｎｖｏｐｌａｋｉｎ〕（ジーンバンク登録番号：Ｕ５３７８６）遺伝子、
インテグリンβ−４鎖〔ｉｎｔｅｇｒｉｎ， ｂｅｔａ ４〕（ジーンバンク登録番号：Ｘ５３５８７）遺伝子、
ビメンチン〔ｖｉｍｅｎｔｉｎ〕（ジーンバンク登録番号：Ｚ１９５５４）遺伝子、
ｃ−ｆｍｓ プロトオンコジーン〔ｃ−ｆｍｓ ｐｒｏｔｏ−ｏｎｃｏｇｅｎｅ〕（ジーンバンク登録番号：Ｘ０３６６３）、
増殖性細胞核抗原〔ｐｒｏｌｉｆｅｒａｔｉｎｇ ｃｅｌｌ ｎｕｃｌｅａｒ ａｎｔｉｇｅｎ〕（ジーンバンク登録番号：Ｍ１５７９６）遺伝子、
白血球抗原関連タンパク質〔ｌｅｕｋｏｃｙｔｅ ａｎｔｉｇｅｎ ｒｅｌａｔｅｄ ｐｒｏｔｅｉｎ 〕（ジーンバンク登録番号：Ｙ００８１５）遺伝子、
インターロイキン８〔ｉｎｔｅｒｌｅｕｋｉｎ ８ 〕（ジーンバンク登録番号：Ｍ２６３８３）遺伝子、
ＧＤＰ 解離阻害因子（ＧＤＩ） 〔ＧＤＰ ｄｉｓｓｏｃｉａｔｉｏｎ ｉｎｈｉｂｉｔｏｒ （ＧＤＩ）〕（ジーンバンク登録番号：Ｌ２０６８８）遺伝子、
脳特異的チューブリンα鎖〔Ｔｕｂｕｌｉｎ， ａｌｐｈａ， ｂｒａｉｎ−ｓｐｅｃｉｆｉｃ〕（ジーンバンク登録番号：Ｘ０１７０３）遺伝子、
細胞周期制御遺伝子ｃｄｃ２（細胞分裂サイクル２，Ｇ１期からＳ 期＆ Ｇ２ 期から Ｍ期）〔ｃｅｌｌ ｄｉｖｉｓｉｏｎ ｃｙｃｌｅ ２， Ｇ１ ｔｏ Ｓ ａｎｄ Ｇ２ ｔｏ Ｍ〕（ジーンバンク登録番号：Ｘ０５３６０）、
組織型プラスミノーゲン活性化因子〔ｐｌａｓｍｉｎｏｇｅｎ ａｃｔｉｖａｔｏｒ， ｔｉｓｓｕｅ 〕ジーンバンク登録番号：Ｍ１５５１８〕遺伝子、
ストローマ細胞由来因子１〔ｓｔｒｏｍａｌ ｃｅｌｌ−ｄｅｒｉｖｅｄ ｆａｃｔｏｒ １ 〕（ジーンバンク登録番号：Ｕ１６７５２）遺伝子、
マトリックスメタロプロテナーゼ１０（ストロメリシン）〔ｍａｔｒｉｘ ｍｅｔａｌｌｏｐｒｏｔｅｉｎａｓｅ １０ （ｓｔｒｏｍｅｌｙｓｉｎ ２） 〕（ジーンバンク登録番号：Ｘ０７８２０）遺伝子、
マトリックスメタロプロテナーゼ１５〔ｍａｔｒｉｘ ｍｅｔａｌｌｏｐｒｏｔｅｉｎａｓｅ １５ （ｍｅｍｂｒａｎｅ−ｉｎｓｅｒｔｅｄ） 〕（ジーンバンク登録番号：Ｚ４８４８２）遺伝子、
抗酸化タンパク質〔ａｎｔｉｏｘｉｄａｎｔ ｐｒｏｔｅｉｎ １ 〕（ジーンバンク登録番号：Ｄ４９３９６）遺伝子、
ラミニンα−４鎖〔ｌａｍｉｎｉｎ， ａｌｐｈａ ４〕（ジーンバンク登録番号：Ｓ７８５６９）遺伝子、
ユビキチン結合酵素Ｅ２Ｍ （酵母ＵＢＣ１２ に相同的である）〔ｕｂｉｑｕｉｔｉｎ−ｃｏｎｊｕｇａｔｉｎｇ ｅｎｚｙｍｅ Ｅ２Ｍ （ｈｏｍｏｌｏｇｏｕｓ ｔｏ ｙｅａｓｔ ＵＢＣ１２）〕ジーンバンク登録番号：ＡＦ０７５５９９〕遺伝子、
レチノブラストーマ結合タンパク質６〔ｒｅｔｉｎｏｂｌａｓｔｏｍａ−ｂｉｎｄｉｎｇ ｐｒｏｔｅｉｎ ６〕（ジーンバンク登録番号：Ｘ８５１３３）遺伝子、
Ｋ−サム〔Ｋ−ｓａｍ 〕（ジーンバンク登録番号：Ｍ８７７７０）遺伝子、
カテプシンＢ〔ｃａｔｈｅｐｓｉｎ Ｂ 〕（ジーンバンク登録番号：Ｌ２２５６９）遺伝子、
シグナル伝達＆転写活性化因子１；９１ｋＤ〔ｓｉｇｎａｌ ｔｒａｎｓｄｕｃｅｒ ａｎｄ ａｃｔｉｖａｔｏｒ ｏｆ ｔｒａｎｓｃｒｉｐｔｉｏｎ １， ９１ｋＤ〕（ジーンバンク登録番号：Ｍ９７９３５）遺伝子、
癌胎児性抗原関連細胞接着分子６（非特異的交差反応性抗原）〔ｃａｒｃｉｎｏｅｍｂｒｙｏｎｉｃ ａｎｔｉｇｅｎ−ｒｅｌａｔｅｄ ｃｅｌｌ ａｄｈｅｓｉｏｎ ｍｏｌｅｃｕｌｅ ６ （ｎｏｎ−ｓｐｅｃｉｆｉｃ ｃｒｏｓｓ ｒｅａｃｔｉｎｇ
ａｎｔｉｇｅｎ） 〕（ジーンバンク登録番号：Ｍ１８７２８）遺伝子、
ヘパラン硫酸プロテオグリカン〔ｈｅｐａｒａｎ ｓｕｌｆａｔｅ ｐｒｏｔｅｏｇｌｙｃａｎ〕（ジーンバンク登録番号：Ｍ８５２８９）遺伝子、
ｖ−Ｋｉ−ｒａｓ２ Ｋｉｒｓｔｅｎ ラット肉腫ウイルスオンコジーンホモログ〔ｖ−Ｋｉ−ｒａｓ２ Ｋｉｒｓｔｅｎ ｒａｔ ｓａｒｃｏｍａ ２ ｖｉｒａｌ ｏｎｃｏｇｅｎｅ ｈｏｍｏｌｏｇ〕（ジーンバンク登録番号：Ｍ５４９６８）遺伝子、
ラスホモログ遺伝子ファミリー、Ｄ因子〔ｒａｓ ｈｏｍｏｌｏｇ ｇｅｎｅ ｆａｍｉｌｙ， ｍｅｍｂｅｒ Ｄ 〕（ジーンバンク登録番号：Ｄ８５８１５）遺伝子、
ヘキサブラキオン（テナシンＣ、サイトタクチン）〔ｈｅｘａｂｒａｃｈｉｏｎ （ｔｅｎａｓｃｉｎ Ｃ， ｃｙｔｏｔａｃｔｉｎ） 〕（ジーンバンク登録番号：Ｘ７８５６５）遺伝子、
サイクリンＤ２〔ｃｙｃｌｉｎ Ｄ２ 〕（ジーンバンク登録番号：Ｄ１３６３９）遺伝子、
メタロプロテナーゼ３組織阻害因子〔ｔｉｓｓｕｅ ｉｎｈｉｂｉｔｏｒ ｏｆ ｍｅｔａｌｌｏｐｒｏｔｅｉｎａｓｅ ３ （Ｓｏｒｓｂｙ ｆｕｎｄｕｓ ｄｙｓｔｒｏｐｈｙ， ｐｓｅｕｄｏｉｎｆｌａｍｍａｔｏｒｙ） 〕（ジーンバンク登録番号：ＡＬ０２３２８２）遺伝子、
ラミニン受容体１、リボゾーマルタンパク質ＳＡ〔ｌａｍｉｎｉｎ ｒｅｃｅｐｔｏｒ １ （６７ｋＤ， ｒｉｂｏｓｏｍａｌ ｐｒｏｔｅｉｎ ＳＡ） 〕（ジーンバンク登録番号：Ｕ４３９０１）遺伝子、
乳酸脱水素酵素〔ｌａｃｔａｔｅ ｄｅｈｙｄｒｏｇｅｎａｓｅ Ａ 〕（ジーンバンク登録番号：Ｘ０２１５２）遺伝子、
ケラチン１８〔ｋｅｒａｔｉｎ １８〕（ジーンバンク登録番号：Ｍ２６３２６）遺伝子、
Ｉａ関連インバリアントγ鎖〔Ｉａ−ａｓｓｏｃｉａｔｅｄ ｉｎｖａｒｉａｎｔ ｇａｍｍａ−ｃｈａｉｎ 〕（ジーンバンク登録番号：ＡＨ００１４８４）遺伝子、
ＣＤ７２抗原〔ＣＤ７２ ａｎｔｉｇｅｎ〕（ジーンバンク登録番号：Ｍ５４９９２）遺伝子、
結合組織増殖因子〔ｃｏｎｎｅｃｔｉｖｅ ｔｉｓｓｕｅ ｇｒｏｗｔｈ ｆａｃｔｏｒ 〕（ジーンバンク登録番号：Ｘ７８９４７）遺伝子、
マトリックスメタロプロテナーゼ３（ストロメリシン１、ゼラチナーゼ前駆体）〔ｍａｔｒｉｘ ｍｅｔａｌｌｏｐｒｏｔｅｉｎａｓｅ ３ （ｓｔｒｏｍｅｌｙｓｉｎ １， ｐｒｏｇｅｌａｔｉｎａｓｅ） 〕（ジーンバンク登録番号：Ｘ０５２３２）遺伝子、
サイクリン依存性キナーゼ４〔ｃｙｃｌｉｎ−ｄｅｐｅｎｄｅｎｔ ｋｉｎａｓｅ ４ 〕（ジーンバンク登録番号：Ｕ３７０２２）遺伝子、
ヒトＳＢクラスＩＩ組織適合性抗原〔Ｈｕｍａｎ ｍＲＮＡ ｆｏｒ ＳＢ ｃｌａｓｓＩＩ ｈｉｓｔｏｃｏｍｐａｔｉｂｉｌｉｔｙ ａｎｔｉｇｅｎ ａｌｐｈａ−ｃｈａｉｎ〕（ジーンバンク登録番号：Ｘ０３１００）遺伝子、
トランスホーミング増殖因子β−３鎖〔ｔｒａｎｓｆｏｒｍｉｎｇ ｇｒｏｗｔｈ ｆａｃｔｏｒ， ｂｅｔａ ３〕（ジーンバンク登録番号：Ｘ１４８８５）遺伝子、
甲状腺ホルモン受容体相互作用因子６〔ｔｈｙｒｏｉｄ ｈｏｒｍｏｎｅ ｒｅｃｅｐｔｏｒ ｉｎｔｅｒａｃｔｉｎｇ
ｐｒｏｔｅｉｎ ６〕（ジーンバンク登録番号：ＡＪ００１９０２）遺伝子、
血小板／内皮細胞接着分子（ＣＤ３１抗原）〔ｐｌａｔｅｌｅｔ／ｅｎｄｏｔｈｅｌｉａｌ ｃｅｌｌ ａｄｈｅｓｉｏｎ ｍｏｌｅｃｕｌｅ （ＣＤ３１ ａｎｔｉｇｅｎ）〕（ジーンバンク登録番号：Ｌ３４６５７）遺伝子、
プロフィリン１〔ｐｒｏｆｉｌｉｎ １〕（ジーンバンク登録番号：Ｊ０３１９１）遺伝子、
ＧＴＰ 結合因子〔ＧＴＰ−ｂｉｎｄｉｎｇ ｐｒｏｔｅｉｎ 〕（ジーンバンク登録番号：ＡＦ１２０３３４）遺伝子、
ホモサピエンスクローン２４７０３ 番 βチューブリンｍＲＮＡ〔Ｈｏｍｏ ｓａｐｉｅｎｓ ｃｌｏｎｅ ２４７０３ ｂｅｔａ−ｔｕｂｕｌｉｎ ｍＲＮＡ， ｃｏｍｐｌｅｔｅ ｃｄｓ〕（ジーンバンク登録番号：ＡＦ０７０６００）遺伝子、
Ｒｈｏ 関連コイルド−コイル含有プロテインキナーゼ１〔Ｒｈｏ−ａｓｓｏｃｉａｔｅｄ， ｃｏｉｌｅｄ−ｃｏｉｌ ｃｏｎｔａｉｎｉｎｇ ｐｒｏｔｅｉｎ ｋｉｎａｓｅ １ 〕（ジーンバンク登録番号：Ｕ４３１９５）遺伝子、
ＩＶ型コラーゲン、α−１鎖〔ｃｏｌｌａｇｅｎ， ｔｙｐｅ ＩＶ， ａｌｐｈａ １〕（ジーンバンク登録番号：Ｍ２６５７６）遺伝子、
トポイソメラーゼ（ＤＮＡ） Ｉ 〔ｔｏｐｏｉｓｏｍｅｒａｓｅ （ＤＮＡ） Ｉ 〕（ジーンバンク登録番号：Ｍ６０７０６）遺伝子、
Ｋｕｎｉｔｚ１型セリンプロテナーゼ阻害因子、〔ｓｅｒｉｎｅ ｐｒｏｔｅａｓｅ ｉｎｈｉｂｉｔｏｒ， Ｋｕｎｉｔｚ ｔｙｐｅ １〕（ジーンバンク登録番号：ＡＢ００００９５）遺伝子、
Ｒｈｏ ＧＴＰ エース活性化タンパク質１〔Ｒｈｏ ＧＴＰａｓｅ ａｃｔｉｖａｔｉｎｇ ｐｒｏｔｅｉｎ １ 〕（ジーンバンク登録番号：Ｕ０２５７０）遺伝子、
ニドジェン〔ｎｉｄｏｇｅｎ 〕（ジーンバンク登録番号：Ｍ３０２６９）遺伝子、
サイクリン依存性キナーゼ５、制御因子１〔ｃｙｃｌｉｎ−ｄｅｐｅｎｄｅｎｔ ｋｉｎａｓｅ ５， ｒｅｇｕｌａｔｏｒｙ ｓｕｂｕｎｉｔ １ （ｐ３５） 〕（ジーンバンク登録番号：Ｘ８０３４３）遺伝子、
ＣＤ９ 抗原〔ＣＤ９ ａｎｔｉｇｅｎ 〕（ジーンバンク登録番号：Ｍ３８６９０）遺伝子、
大動脈タイプ平滑筋α−アクチン遺伝子、エキソン９〔ａｏｒｔｉｃ−ｔｙｐｅ ｓｍｏｏｔｈ ｍｕｓｃｌｅ ａｌｐｈａ−ａｃｔｉｎ ｇｅｎｅ， ｅｘｏｎ ９（ジーンバンク登録番号：Ｍ３３２１６）〕、
骨形成タンパク質１〔ｂｏｎｅ ｍｏｒｐｈｏｇｅｎｅｔｉｃ ｐｒｏｔｅｉｎ １ （ＢＭｐ１） （ジーンバンク登録番号：Ｕ５０３３０）〕遺伝子、
ＶＩ型コラーゲンα−３鎖〔ｃｏｌｌａｇｅｎ ｔｙｐｅ ＶＩ ａｌｐｈａ−３（ジーンバンク登録番号：Ｘ５２０２２）〕遺伝子、
サイトカインｈｕｍｉｇ〔ｃｙｔｏｋｉｎｅ ｈｕｍｉｇ； ｉｎｔｅｒｆｅｒｏｎ−ｇａｍｍａ−ｉｎｄｕｃｅｄ ｍｏｎｏｋｉｎｅ （ＭＩＧ） （ジーンバンク登録番号：Ｘ７２７５５）〕遺伝子、
インテグリンα−３鎖〔ｉｎｔｅｇｒｉｎ ａｌｐｈａ−３ ｃｈａｉｎ（ジーンバンク登録番号：Ｍ５９９１１）〕遺伝子、
インターフェロン誘導性遺伝子ファミリー〔ｉｎｔｅｒｆｅｒｏｎ ｉｎｄｕｃｉｂｌｅ ｇｅｎｅ ｆａｍｉｌｙ（ジーンバンク登録番号：Ｘ５７３５１）〕遺伝子、
転移関連ＭＴＡ１〔ｍｅｔａｓｔａｓｉｓ−ａｓｓｏｃｉａｔｅｄ ＭＴＡ１（ジーンバンク登録番号：Ｕ３５１１３）〕遺伝子、
ノッチ２ ノッチ ホモログ３〔Ｎｏｔｃｈ２ Ｎｏｔｃｈ ｈｏｍｏｌｏｇ ３（ジーンバンク登録番号：Ｕ９７６６９）〕遺伝子、及び
プロト−オンコジーンｒｈｏＡ多剤耐性タンパク質〔ｐｒｏｔｏ−ｏｎｃｏｇｅｎｅ ｒｈｏＡ ｍｕｌｔｉｄｒｕｇ ｒｅｓｉｓｔａｎｃｅ ｐｒｏｔｅｉｎ； ＧＴＰ−ｂｉｎｄｉｎｇ ｐｒｏｔｅｉｎ （ｒｈｏＡ）（ジーンバンク登録番号：Ｌ２５０８０）〕遺伝子のそれぞれの遺伝子からなる群より選択された少なくとも８種、より好ましくは、少なくとも２０種の遺伝子が挙げられる。なお、少なくとも８種の遺伝子、少なくとも２０種の遺伝子の具体例としては、特に限定されないが、それぞれ後述の実施例３の表５に記載の遺伝子、実施例２の表４に記載の遺伝子が挙げられる。
【００３８】
前記遺伝子は、癌又は癌細胞を検出するための指標としても有用である。したがって、本発明の癌、特に胃癌の悪性度の評価方法により、癌、特に胃癌における癌細胞の検出をも可能にする。かかる癌又は癌細胞の検出方法も本発明の範囲に包含される。
【００３９】
癌又は癌細胞の検出方法においては、選択された遺伝子の多くは癌化によりその発現が変動し、さらに悪性化により変動パターンが異なるため、同一の手法により癌又は癌細胞の検出とその癌の悪性化判定とを同時に行なうことができる。
【００４０】
本発明の癌、特に胃癌の悪性度の評価方法においては、遺伝子の発現量は、該遺伝子より転写されるｍＲＮＡ量又は該遺伝子にコードされたポリペプチド量から測定することができる。
【００４１】
前記ｍＲＮＡ量の測定には、公知の種々の方法、例えば、ノーザン・ハイブリダイゼーション法、ｉｎ ｓｉｔｕハイブリダイゼーション法、ドット・ブロット・ハイブリダイゼーション法、ＤＮＡアレイを使用する方法等に代表されるハイブリダイゼーション法、ＲＴ−ＰＣＲ法に代表される核酸増幅法等が使用できる。十分なｍＲＮＡの検出感度を得る観点から、ハイブリダイゼーション法又は核酸増幅法が好ましく、より具体的には、発現量を測定しようとする遺伝子に対応する核酸又はその断片が支持体上のそれぞれ定められた領域に固定化されたアレイ（例えば、ＤＮＡアレイ）を使用するハイブリダイゼーション法、ＲＴ−ＰＣＲ法が好適である。多数の遺伝子の発現を同時に測定する観点から、ＤＮＡアレイを使用するハイブリダイゼーション法が特に好適である。
【００４２】
本明細書において「ＤＮＡアレイ」とは、遺伝子又は遺伝子由来のＤＮＡ断片が支持体上の定められた領域に固定されているものを指し、例えばＤＮＡチップと呼称されているものを包含する。また、遺伝子又は遺伝子由来のＤＮＡ断片が支持体上に高密度に固定されているものはＤＮＡマイクロアレイとも呼ばれる。
【００４３】
ＤＮＡアレイの支持体は、ハイブリダイゼーションに使用可能なものであれば特に限定はなく、通常スライドグラス、シリコンチップ、ニトロセルロースやナイロンの膜等が使用される。また、支持体上に固定される遺伝子又はその断片としては、特に限定するものではないが、例えばゲノムＤＮＡライブラリーやｃＤＮＡライブラリー、これらを鋳型としたＰＣＲ等によって増幅されたＤＮＡが挙げられる。
【００４４】
このようなＤＮＡアレイを使用することにより、核酸試料中に含まれる多種類の核酸分子の量を同時に測定することができる。また、少量の核酸試料でも測定できるという利点がある。例えば、試料中のｍＲＮＡを標識するか、もしくはｍＲＮＡを鋳型として標識されたｃＤＮＡを調製し、これとＤＮＡアレイの間でハイブリダイゼーションを実施することにより、試料中で発現されているｍＲＮＡを同時に検出し、さらにその発現量を測定することができる。
【００４５】
標識に用いられる標識物質としては、放射性同位元素、蛍光物質、化学発光物質、発光団を有する物質等の物質を用いることができる。例えば、蛍光物質としては、Ｃｙ２、ＦｌｕｏｒＸ、Ｃｙ３、Ｃｙ３．５、Ｃｙ５、Ｃｙ５．５、Ｃｙ７、イソチオシアン酸フルオレセイン（ＦＩＴＣ）、テキサスレッド、ローダミン等が挙げられる。また、同時に検出できる点から、２種類以上の蛍光物質を用いて、被検試料、対照として用いる試料をそれぞれ異なった蛍光物質で標識することが望ましい。ここで試料の標識は、試料中のｍＲＮＡ、該ｍＲＮＡ由来のｃＤＮＡもしくは該ｃＤＮＡより転写あるいは増幅された核酸を標識することによって実施される。
【００４６】
標識の検出法は、用いられる標識物質の種類により、適宜選択することができる。例えば、前記Ｃｙ３及びＣｙ５を標識物質として用いる場合、Ｃｙ３は５３２ｎｍ、Ｃｙ５は６３５ｎｍの波長でスキャンすることにより検出することができる。なお、標識の強度を遺伝子の発現量の指標とする。
【００４７】
核酸増幅法、特にＲＴ−ＰＣＲ法によりｍＲＮＡ量を測定する場合、例えば、競合ＰＣＲ法、ＴａｑＭａｎ法〔例えば、リンダ・ジー・リー等（Ｌｉｎｄａ Ｇ．Ｌｅｅ）、ヌクレイック アシッズ リサーチ、第２１巻、第３７６１−３７６６頁（１９９３）等を参照のこと〕等により、ｍＲＮＡ量を測定できる。
【００４８】
前記遺伝子にコードされたポリペプチド量は、該ポリペプチド又はその断片に対する抗体を用いて測定することができる。具体的には、例えば、標識抗体を用い、慣用のＥＬＩＳＡ法等により行なうことができる。
【００４９】
本発明の評価方法においては、ステップ（２）において、被検試料における遺伝子発現量と対照試料における遺伝子発現量とを比較することにより遺伝子の発現量を解析し、それにより癌、特に胃癌の悪性度を評価することができる。ここで、対照試料とは、健常人由来試料又は非病変部位由来の試料をいう。
【００５０】
遺伝子の発現の解析は、パターン解析により行なうことができる。かかるパターン解析は、被検試料における各遺伝子の発現強度シグナル、又は対照試料に対する被検試料における各遺伝子の相対発現比率を用いたクラスタリング解析により行なうことができる。
【００５１】
すなわち、得られた各遺伝子の発現強度を用いコンピューターによるクラスタリング解析を行なう。解析に用いられるデータとしては、被検試料における各遺伝子の発現強度シグナル、又は対照試料に対する被検試料における各遺伝子の相対発現比率が挙げられる。
【００５２】
なお、ステップ（２）において、さらにステップ（１）で得られた各遺伝子の発現量の測定値の補正処理を行なうことができる。
【００５３】
前記相対発現比率とは、〔被検試料における遺伝子発現量／対照試料における遺伝子発現量〕により示される値をいう。クラスタリングの手法としては、特に限定されないが、Ｈｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ 、Ｎｅｕｒａｌ ｎｅｔｗｏｒｋ ｃｌｕｓｔｅｒｉｎｇ 等が挙げられる。
【００５４】
以下にバイオテクノロジー分野で比較的よく用いられるＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇについて説明する。
【００５５】
Ｈｉｅｒａｒｃｈｉｃａｌ ｃｌｕａｔｅｒｉｎｇ とは、個々のサンプルにつき測定されたデータを多次元座標に配置した場合、データ間の距離が最も近いサンプルから順に見つけ、最終的には似たサンプルが隣り合うように並べるもので、結果を樹形図で表現できるのが特色である。サンプル間の距離の測定法には種々あり、ユークリッド距離、マハラノビスの汎距離等がよく用いられる。また、樹形図を作成する過程で、個々のサンプルをまとめた組の相互の距離の定義の仕方も種々あり、特に限定されないが、主に最短距離法、最長距離法、群平均法、重心法、ウォード法が挙げられる。
【００５６】
被検試料、例えば、癌組織における各遺伝子の発現強度シグナルを用いる場合は、各遺伝子のスポットシグナルからバックグラウンドシグナルを差し引いた値（本値を発現強度シグナルとする）をコンピューターに入力するが、この際のシグナル強度の分布はアレイ毎に異なるので補正（ｎｏｒｍａｌｉｚａｔｉｏｎ） を行なう方が好ましい。このｎｏｒｍａｌｉｚａｔｉｏｎ の方法には、Ｚ−ｓｃｏｒｅ 、Ｍｅａｎ ｄｅｖｉａｔｉｏｎ、Ｍｅａｎ ａｂｓｏｌｕｔｅ ｄｅｖｉａｔｉｏｎ 、Ｍｅｄｉａｎ ａｂｓｏｌｕｔｅ ｄｅｖｉａｔｉｏｎ 等があり、適宜選択する。クラスタリングの対象となる遺伝子は、前記表１及び２に記載の遺伝子から選択され、数が多いほうがより精度の高い分類が可能である。具体的には、少なくとも８種の限定された遺伝子の組み合わせでも、限られた悪性度因子を判定可能である。
【００５７】
ついで、ｎｏｒｍａｌｉｚａｔｉｏｎ 後の発現強度シグナルについて、Ｈｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ 解析を行なうことにより、検体、例えば、胃癌検体のクラスタリング並びに遺伝子のクラスタリングが行なわれる。その結果、樹形図上での検体間分類、すなわち遺伝子発現の強弱パターンが似通った検体間相互には、病理診断における癌の浸潤程度、リンパ節転移、腹膜転移、肝転移等の悪性度の面での共通性があり、かつ悪性度の高い転移検体と浸潤程度の低い早期癌検体はもっとも離れた距離に配置された。この方法を用いることにより、手術前、バイオプシーにより少量の癌組織を採取するだけで、未知の検体の悪性度が判定可能となる。すなわち、癌組織よりｍＲＮＡを精製し、たとえば一色蛍光ラベリングしたプローブを用いハイブリダイゼーションを行ない、個々の遺伝子の発現強度シグナルを求める。この発現強度シグナルと既に悪性度の判明した標準となる胃癌検体の発現強度シグナルと共にクラスタリング解析を行なうことにより、未知の検体がどの悪性度レベルの標準検体に近いかが判定される。また従来の手術後の病理診断では、癌の浸潤が粘膜から筋層に広がっていた場合、予後の比較的良い癌なのか、将来、再発及び／又は転移する可能性がある悪性の癌なのかの分岐点に位置するため予後の判定不能であったが、本発明のクラスタリング解析を行なうことにより悪性度未知の検体が、悪性度の低い初期の癌か、あるいは転移した悪性度の高い癌検体のいずれに距離的に近いか判明し、予後が予想可能となる。
【００５８】
一方、被検試料（例えば、胃癌組織）の対照試料（例えば、対照正常組織）に対する各遺伝子の相対発現比率、すなわち、〔被検試料における遺伝子発現量／対照試料における遺伝子発現量〕（具体的には、〔癌組織における遺伝子発現量／対照正常組織における遺伝子発現量〕）を用いた場合においても、胃癌検体のクラスタリングは、癌組織の発現強度シグナルを用いた場合とほぼ同様の結果が得られる。相対発現比率を用いた場合は、遺伝子発現における個体差を排除した遺伝子発現プロファイルを見ることができる利点がある。遺伝子のクラスタリングは癌化及び悪性化による発現変動が似通った遺伝子ごとに分類されており、その発現変動と病理診断上の悪性度因子との相関性解析を行なうことにより悪性化に関与する遺伝子を特定することが可能である。これにより、表１及び２に示す１０５個の遺伝子、表６及び７に示す５１個の遺伝子と悪性度との関連性、すなわち癌の浸潤程度を示す深達度レベルとの相関性が見出された。
【００５９】
以上のように解析して得られた結果は、例えば、プリンター、ディスプレイまたはディスプレイ用グラフィックソフトウエアのような他のソフトウエアパッケージに対して出力が与えられる。かかる出力は、本発明の評価方法により得られた結果を診断などに用いる場合、特に有利である。
【００６０】
さらに、上記法より見出された悪性度に相関する遺伝子の分類を応用し、より簡易的な悪性度診断が可能なことを見出した。すなわち、コンピュータによる複雑なクラスタリング解析を行なうことなく、発現変動する遺伝子の数や発現変動の割合を指標として、遺伝子の発現変動レベルを評価することにより、癌、特に胃癌の悪性度の評価が可能であることを見出した。かかる態様も本発明に含まれる。
【００６１】
ここで、「遺伝子の発現変動レベル」は、▲１▼〔対照試料に比較して被検試料において発現量の上昇した遺伝子の相対発現比率若しくは該相対発現比率の対数の絶対値の総和により示された値〕、又は▲２▼〔対照試料に比較して被検試料において発現量の上昇した遺伝子を１ポイントとし、被検試料中の各遺伝子のポイントの総和により示された値〕により評価することができる。具体的には、例えば、対照試料に比較して、被検試料において発現量の上昇した遺伝子の相対発現比率が２以上のときを１ポイントとし、被検試料中の各遺伝子のポイントの総和により示された値により、遺伝子の発現変動レベルを評価することができる。
【００６２】
本明細書において「遺伝子の発現変動」とは、２倍以上の変動、すなわち、相対発現比率（〔被検試料における遺伝子発現量／対照試料における遺伝子発現量〕）が２以上若しくは１／２以下の場合をいう。ここで、前記相対発現比率が２以上の場合が発現上昇した遺伝子であり、１／２以下の場合が発現低下した遺伝子であることの指標である。前記▲２▼では、悪性度の各因子と大きく相関する遺伝子群から選択された８個以上からなるグループの発現変動遺伝子の個数を、すでに悪性度、予後がわかっている標準となる癌組織の発現変動遺伝子数と比較することにより評価可能である。
【００６３】
前記▲１▼の値においては、〔被検試料における遺伝子発現量／対照試料における遺伝子発現量〕の値を対数で示し、その絶対値を用いることで発現の増加率と減少率を対等に評価することができる。その際の対数値は自然対数ｌｏｇ_２、絶対対数ｌｏｇ_１０ のいずれで表されてもよい。
【００６４】
発現変動する遺伝子の数や発現変動の割合を指標に悪性度を評価する際には、同一チップ上に搭載した陰性対照遺伝子の強度シグナル（バックグラウンド値）を基準に判定すると信頼性の高い評価をすることが可能である。すなわち、被検試料における発現強度シグナル又は対照試料における発現強度シグナルの両者がバックグラウンド値の２倍より小さい場合は変動の信頼性は低いが、いずれかが２倍以上で４倍より小さい場合は信頼度が高く、またいずれかが４倍以上の場合はさらに信頼度が高い。これらの信頼度を悪性度判定に加味すると評価の信頼性があがる。すなわち癌の発現強度シグナル又は対照正常組織の発現強度シグナルのいずれかが陰性対照遺伝子の強度シグナルの２倍以上の場合の発現変動する遺伝子の数や発現変動の割合の和で評価することがより望ましい。
【００６５】
本発明においては、遺伝子発現量の解析は、胃癌の悪性度に応じて発現量が変化する少なくとも８種の遺伝子の発現量と、被検試料におけるハウスキーピング遺伝子の発現量との比較により行なうことができる。
【００６６】
本発明の癌、特に胃癌の悪性度の評価方法においては、ステップ（２）において、被検試料における遺伝子発現量と対照試料における遺伝子発現量とを比較することにより遺伝子の発現量を解析し、得られた遺伝子発現量の解析結果と各遺伝子に関する重み付け係数とにより癌、特に胃癌の悪性度を評価してもよい。すなわち、対照試料に対する被検試料における遺伝子の相対発現比率の対数により示された値に、各遺伝子の悪性化への寄与の大きさを考慮した重み付け係数を掛け合せた値の総和により評価を行なうことができる。本発明においては、それぞれの遺伝子の発現変動が悪性化に寄与する程度は均一ではないため、その程度により遺伝子の発現変動に重み付けをすることで悪性化をより反映した評価が可能となる。
【００６７】
上記の態様においては、悪性化への寄与程度の大きな遺伝子の重み付け係数を高く、寄与程度の小さい遺伝子の重み付け係数を小さく設定される。また、対照試料に比較して、被検試料において発現上昇した遺伝子が悪性化と逆相関する場合は、重み付け係数はマイナスの値となる。
【００６８】
重み付け係数の配点段階は、特に限定されないが、通常２から１０段階で配点される。
【００６９】
例えば、後述の実施例に記載の表６及び表７には７０種の遺伝子について、−３、−１、１、２、３、４、５の７段階からなる重み付け係数が例示されているが、これを５または３段階に変更してもよい。
【００７０】
重み付け係数を使用した具体的な評価の方法としては、例えば、被検試料において対照試料の２倍に発現量が上昇した重み付け係数５の遺伝子について、発現比率である２の自然対数値である０．６９３に重み付け係数の５を掛け合せた値３．４６５が当該遺伝子のポイントとなる。同様にすべての遺伝子のポイントを求め、ポイントの総和で悪性度が評価される。
【００７１】
上記の重み付け係数は、臨床病理データおよび予後のわかっている検体を用いて遺伝子の発現変動レベルを調べ、悪性度を適切に指示し得るような値が設定される。
【００７２】
２．本発明のアレイ
癌化、癌の悪性化を評価するために用いられる、癌、特に胃癌の悪性度を評価するための本発明のアレイは、癌、特に胃癌の悪性度に応じて発現量が変化する少なくとも８種の遺伝子に対応する核酸（例えば、ＤＮＡ等）又はその断片がそれぞれ支持体上の定められた領域に固定化されたアレイである。本発明のアレイにおいては、前記１．の項に挙げられた、癌、特に胃癌の悪性度に応じて発現量が変化する少なくとも８種の遺伝子に対応する核酸又はその断片が固定されているため、癌、特に胃癌の悪性度を簡便な操作により評価することができるという優れた性質を有する。
【００７３】
ここで、「それぞれ定められた位置に固定化される」とは、おのおのの遺伝子に対応する核酸又はその断片の固定化されている位置が支持体上においてあらかじめ決定されていることを意味する。すなわち、このようなアレイを使用した場合には、検出されたシグナルの位置からこのシグナルがどの遺伝子の核酸又はその断片に由来するのかを知ることができる。
【００７４】
本発明のアレイに用いる支持体は、ハイブリダイゼーションに使用可能なものであれば特に限定はなく、通常スライドグラス、シリコンチップ、ニトロセルロースやナイロンの膜等が使用される。更に好ましくは、非多孔性で、表面が滑らかな構造を有する材質であればよく、特に限定はないが、例えばスライドガラス等のガラスが好適に使用できる。支持体の表面は、共有結合又は非共有結合により一本鎖ＤＮＡを固定化できるものであればいずれでもよく、支持体の表面に親水性又は疎水性の官能基を有しているものが好適に使用でき、特に限定はないが、例えば、水酸基、アミノ基、チオール基、アルデヒド基、カルボキシル基、アシル基等を有しているものが好適に使用できる。これらの官能基は、支持体自体の表面特性として存在していてもよいが、表面処理によって導入してもよい。このような表面処理物としては、例えば、ガラスをアミノアルキルシラン等の市販のシランカップリング剤で処理したものや、ポリリジンやポリエチレンイミン等のポリ陽イオンで処理したもの等が挙げられる。また、これらの処理を施したスライドガラスの一部は市販されている。
【００７５】
本発明のアレイにおいては、核酸又はその断片は、一本鎖、二本鎖のどちらが固定されていてもよい。例えば、該核酸子又はその断片が、変性された二本の鎖として支持体に整列固定化されたＤＮＡアレイや、固定化されたＤＮＡの少なくとも一部が一本鎖ＤＮＡであるＤＮＡアレイでもよい。また、本発明のアレイは、二本鎖ＤＮＡを変性下において、同一支持体に整列させてスポットしたＤＮＡアレイでもよい。
【００７６】
さらに、本発明のアレイにおいては、固定化される遺伝子の核酸あるいはその断片の支持体上における密度について特に限定はないが、例えば、高密度のアレイでもよく、１００ドット／ｃｍ^２ 以上の密度で核酸、特にＤＮＡが固定化されたアレイが好適に使用できる。
【００７７】
支持体上に固定化される核酸又はその断片には特に限定はなく、ポリヌクレオチド、オリゴヌクレオチドのいずれであってもよい。また、その製法にも特に限定はなく、化学的に合成したもの、天然の核酸から単離、精製したもの、酵素的に合成したもの、さらにはこれらを組み合わせて使用することができる。
【００７８】
支持体上に固定される核酸又はその断片としては、特に限定するものではないが、例えば、鎖長が５０塩基長以上の二本鎖ポリヌクレオチド又はその誘導体であって、ＰＣＲ（ｐｏｌｙｍｅｒａｓｅ ｃｈａｉｎ ｒｅａｃｔｉｏｎ ）法等により、酵素的に増幅して調製されるものを、ＤＮＡの固定化支持体への固定化時に変性し、一本鎖ＤＮＡ又はその誘導体としたものが好適に使用できる。該誘導体としては、支持体表面への固定化を可能にするような修飾を付されたものであれば良く、特に限定するものではないが、例えばＤＮＡの５’末端にアミノ基やチオール基等の官能基が導入されたＤＮＡが挙げられる。
【００７９】
例えば、ゲノムＤＮＡライブラリーあるいはｃＤＮＡライブラリーを鋳型としたＰＣＲ等によって増幅されたＤＮＡを使用することができる。上記の遺伝子に対応する核酸又はその断片を公知の方法、例えばアミノ基を導入した支持体上に固定することにより該アレイを作製することができる。また、上記固定化の操作をＤＮＡアレイ作製装置、例えばＧＭＳ社製のＤＮＡチップ作製装置を使用して行なうことにより、遺伝子に対応する核酸が整列、固定化された本発明のアレイを作製することができる。
【００８０】
アレイに核酸の断片を固定する場合、該断片の鎖長は、特に限定されるものではなく、例えば、約１００塩基長〜約１キロ塩基長であることが望ましく、また、該鎖長より短いあるいは長いものであっても被検試料由来の核酸とハイブリダイゼーションにおいて特異的にハイブリダイズするものであればよい。
【００８１】
本発明のアレイに用いられる遺伝子は、癌、特に胃癌の悪性度に応じて発現量が変化する遺伝子であればよく、特に限定されるものではないが、前記１．の項で挙げられた遺伝子、例えば、癌遺伝子、癌抑制遺伝子、増殖因子、転移・浸潤因子、血管新生因子もしくはエネルギー代謝に関与するタンパクをコードする遺伝子等が挙げられる。さらに、例えば、機能は未知であるが種々の悪性度の異なる胃癌患者の癌組織及び対照正常組織より抽出された全ＲＮＡ中よりディファレンシアルディスプレイ（ＤＤ）法により胃癌の悪性化に伴い発現の減少、又は増加する遺伝子を見出してこれらを固定化してもよい。さらに、前記１．の項に記載の遺伝子選別方法で判明した癌、特に胃癌の悪性化の過程で発現変動する遺伝子は全て入手可能であり、かかる遺伝子に対応する核酸を全て固定化した癌、特に胃癌の悪性度評価用ＤＮＡアレイを作製することができる。
【００８２】
具体的には、前記１．の項に記載の「胃癌の悪性度に応じて発現量が変化する遺伝子」の具体例に挙げられた遺伝子からなる群より選択された少なくとも８種、好ましくは少なくとも２０種の遺伝子が挙げられる。
【００８３】
本発明のアレイは、癌、特に胃癌における癌細胞を検出するためにも使用することができる。
【００８４】
３．本発明の癌、特に胃癌の悪性度を評価するためのキット
本発明の癌、特に胃癌の悪性度を評価するためのキットとしては、検出対象物により、▲１▼胃癌の悪性度に応じて発現量が変化する少なくとも８種の遺伝子から発現されるｍＲＮＡ又はその断片を検出するためのプライマー及び／又はプローブを含有したキット、並びに▲２▼胃癌の悪性度に応じて発現量の変化する少なくとも８種の遺伝子にコードされたポリペプチド又はその断片に対する抗体又はその断片を含有したキットが挙げられる。
【００８５】
本発明のキットは、▲１▼胃癌の悪性度に応じて発現量が変化する少なくとも８種の遺伝子から発現されるｍＲＮＡ又はその断片を検出するためのプライマー及び／又はプローブ、あるいは▲２▼胃癌の悪性度に応じて発現量の変化する少なくとも８種の遺伝子にコードされたポリペプチド又はその断片に対する抗体又はその断片を含有しているため、前記癌、特に胃癌の悪性度の評価方法に用いることができ、より簡便、かつ迅速な操作を可能にする。また、本発明のキットは、癌、特に胃癌における癌細胞を検出するためにも使用することができる。
【００８６】
前記▲１▼のキットは、癌、特に胃癌の悪性度に応じて発現量の変化する少なくとも８種の遺伝子から発現されるｍＲＮＡ又はその断片を核酸増幅法、具体的には、ＲＴ−ＰＣＲ法によって検出するためのプライマー及び／又はプローブを含有する。
【００８７】
前記プライマー及び／又はプローブは、癌、特に胃癌の悪性度に応じて発現量の変化する少なくとも８種の遺伝子に、又は該遺伝子に相補的な塩基配列を有する核酸にストリンジェントな条件下にハイブリダイズするものであればよい。
【００８８】
上記「ストリンジェントな条件」とは、特に限定されないが、例えば、６×ＳＳＣ、０．５％ＳＤＳ、５×デンハルト、１００μｇ／ｍｌニシン精子ＤＮＡを含む溶液中、〔前記プライマー及び／又はプローブのＴｍ−２５℃〕の温度で一晩保温する条件等をいう。
【００８９】
また、上記のプライマーの塩基配列は、通常の核酸増幅法、特にＲＴ−ＰＣＲにおけるの反応条件下に、前記遺伝子に対応する核酸を特異的に増幅しうる配列であればよい。
【００９０】
一方、プローブの塩基配列も、前記遺伝子に対応する核酸に前記ストリンジェントな条件下にハイブリダイズしうる核酸の配列であればよい。
【００９１】
なお、プローブ又はプライマーのＴｍは、例えば、下記式：
Ｔｍ＝８１．５−１６．６ （ｌｏｇ_１０［Ｎａ^＋ ］）＋０．４１ （％Ｇ＋Ｃ）−（６００／Ｎ）
（式中、Ｎはプローブ又はプライマーの鎖長であり、％Ｇ＋Ｃはプローブ又はプライマー中のグアニン及びシトシン残基の含有量である）
により求められる。
【００９２】
また、プローブ又はプライマーの鎖長が１８塩基より短い場合、Ｔｍは、例えば、Ａ＋Ｔ（アデニン＋チミン）残基の含有量と２℃との積と、Ｇ＋Ｃ残基の含有量と４℃との積との和〔（Ａ＋Ｔ） ×２＋（Ｇ＋Ｃ） ×４ 〕により推定することができる。
【００９３】
前記プローブの鎖長は、特に限定はないが、非特異的なハイブリダイゼーションを防止する観点から、１５塩基以上であり、好ましくは１８塩基以上であることが望ましい。
【００９４】
また、プライマーの鎖長は、特に限定はないが、例えば、１５〜４０塩基長であり、好ましくは１７〜３０塩基長であることが望ましい。
【００９５】
前記「胃癌の悪性度に応じて発現量の変化する少なくとも８種の遺伝子」としては、好ましくは、前記１．の項に記載の「胃癌の悪性度に応じて発現量が変化する遺伝子」の具体例に挙げられた遺伝子からなる群より選択された少なくとも８種、より好ましくは少なくとも２０種の遺伝子が挙げられる。
【００９６】
前記▲２▼のキットにおいて、ポリペプチドは、前記胃癌の悪性度に応じて発現量の変化する少なくとも８種の遺伝子によりコードされたポリペプチドが挙げられる。具体的には、前記１．の項に記載の「胃癌の悪性度に応じて発現量が変化する遺伝子」の具体例に挙げられた遺伝子にコードされたポリペプチドからなる群より選択された少なくとも８種、より好ましくは少なくとも２０種のポリペプチドが挙げられる。
【００９７】
抗体は、前記ポリペプチドに特異的に結合する能力を有するものであれば、特に限定はなく、ポリクローナル抗体、モノクローナル抗体のどちらでもよい。さらに、公知技術により修飾された抗体や抗体の誘導体、例えばヒト化抗体、Ｆａｂフラグメント、単鎖抗体等を使用することもできる。前記抗体は、例えば、１９９２年、ジョン・ワイリー＆サンズ社（Ｊｏｈｎ Ｗｉｅｌｙ ＆ Ｓｏｎｓ， Ｉｎｃ）発行、ジョン・Ｅ・コリガン（Ｊｏｈｎ Ｅ． Ｃｏｌｉｇａｎ ）編集、カレント・プロトコルズ・イン・イムノロジー（Ｃｕｒｒｅｎｔ Ｐｒｏｔｏｃｏｌｓ ｉｎ Ｉｍｍｕｎｏｌｏｇｙ ）に記載の方法により、前記ポリペプチドの全部又は一部を用いてウサギやラット、マウス等を免疫することにより、容易に作製され得る。こうして得られた抗体を精製後、ペプチダーゼ等により処理することにより、抗体の断片が得られる。また、遺伝子工学的に抗体を作製することもできる。さらに、本発明の抗体又はその断片は、酵素免疫測定法、蛍光免疫測定法、発光免疫測定法等による検出を容易にするために、各種修飾をしてもよい。
【００９８】
上記の抗体又はその断片には、ポリペプチドのある部分断片に特異的に結合しうるものも含まれる。
【００９９】
本発明のキットは、検出用試薬等を適宜含有してもよい。
【０１００】
また、本発明は、前記評価方法等により、得られた結果を解析手段（例えば、コンピューターによる画像処理方法等）により解析し、その結果を、例えば、紙等の記録媒体；コンピューター読み取り可能な記録媒体等に記録又は表示して提供する診断方法をも提供することができる。かかる診断方法も本発明に含まれる。
【０１０１】
【実施例】
以下に実施例をもってさらに詳細に本発明を説明するが、本発明は実施例の範囲に限定されるものではない。
【０１０２】
実施例１
（１）ＤＮＡアレイの作成
ディファレンシャルディスプレイ（ＤＤ）法により見いだされた、国際公開公報第９８／３７１８７号公報に記載の癌関連遺伝子のｐＢｌｕｅｓｃｒｉｐｔファージミドクローンに挿入されているＤＮＡ断片を鋳型とし、当該ファージミドのマルチクローニングサイトの両端に設定されたプライマーを使用したＰＣＲ法により目的のｃＤＮＡ断片を増幅、回収した。さらに、ハウスキーピング遺伝子としてグリセルアルデヒド３−リン酸デヒドロゲナーゼ（ＧＡＰＤ）、シクロフィリン遺伝子を、陰性対照としてプラスミドｐＵＣ１８をそれぞれ同様に調製した。
【０１０３】
これらのＤＮＡをエタノール沈殿法により回収して１００ｍＭ 炭酸バッファー（ｐＨ９．５）で１μＭとなるように溶解し、ＤＮＡチップ作製装置（ＧＭＳ社製）を用いてアミノ基導入スライドガラス（シグマ社製）にスポットし、ＵＶ照射により固定した。スライドを０．２％ＳＤＳ、次いで蒸留水で洗浄乾燥してＤＮＡアレイとした。
【０１０４】
（２）蛍光標識ｃＤＮＡの調製
表３に示す、癌の浸潤程度、転移の有無、又は進行度の異なる１０人の胃癌患者より、癌摘出手術時に摘出された組織から胃癌組織と対照正常胃組織とを取りわけた。ついで、ＡＧＰＣ（Ａｃｉｄ Ｇｕａｎｉｄｉｕｍ Ｐｈｅｎｏｌ−Ｃｈｌｏｒｏｆｏｒｍ）法により、各組織から個々に全ＲＮＡを抽出した。これらの全ＲＮＡのそれぞれからＯｌｉｇｏｔｅｘ−ＭＡＧ ｍＲＮＡ Ｐｕｒｉｆｉｃａｔｉｏｎ ｋｉｔ（宝酒造社製）を用いて、ｍＲＮＡを精製した。
【０１０５】
【表３】

【０１０６】
なお、表３中の各項目は以下のような所見を示す：
進行度：
進行度Ｉは、癌の浸潤程度が浅く、転移の認められない早期の癌である。
進行度ＩＩは、癌が浸潤、近接リンパ節に転移した状態をいう。
進行度ＩＩＩは、遠隔リンパ節に広がった状態をいう。
進行度ＩＶは、さらに進行し、より遠隔リンパ節に広がるとともに腹膜転移、肝転移した状態をいう。
【０１０７】
分化度：
一般型として、乳頭腺癌（ｐａｐ） 、高分化腺癌（ｔｕｂ １） 、中分化腺癌（ｔｕｂ ２） 、低分化腺癌（ｐｏｒ） 、印環細胞癌（ｓｉｇ） 、粘液癌（ｍｕｃ） に分類される
【０１０８】
深達度：
癌浸潤の及んだ最も深い胃壁の層で表し、粘膜（ｍ） 、粘膜下層（ｓｍ）、筋層（ｍｐ）、しょう膜化組織（ｓｓ）、しょう膜を破り腹膜に露出（ｓｅ）で表現される。
【０１０９】
リンパ節転移（組織学的に検索されたリンパ節群の転移程度を表わす）：
全く転移を認めない（Ｎ_０）
第１群リンパ節のみに転移を認める（Ｎ_１）
第２群リンパ節に転移を認めるが、第３群、第４群には転移を認めない（Ｎ_２）
第３群リンパ節に転移を認めるが、第４群には転移を認めない（Ｎ_３）
第４群リンパ節に転移を認める（Ｎ_４）
【０１１０】
リンパ管侵襲：
侵襲が認められない（０）
侵襲が軽度に認められる（１）
侵襲が中程度に認められる（２）
侵襲が高度に認められる（３）
【０１１１】
静脈侵襲：
侵襲が認められない（０）
侵襲が軽度に認められる（１）
侵襲が中程度に認められる（２）
侵襲が高度に認められる（３）
【０１１２】
腹膜転移：
全く播種を認めない（０）
近接腹膜に播種を認める（１）
遠隔腹膜に少数の転移（２）
遠隔腹膜に多数の転移（３）
【０１１３】
肝転移：
全く肝転移を認めない（０）
一葉にのみ肝転移を認める（１）
両葉に少数散在性の転移（２）
両葉に多数散在性の転移（３）
【０１１４】
浸潤増殖様式：
癌層が膨張性の発育を示し周囲と一線を画する （α）
病巣が浸潤性の増殖を示し周囲組織との境が不明瞭 （γ）
その中間にあるもの （β）
【０１１５】
上記のｍＲＮＡを鋳型とし、対照正常胃癌組織群にはＣｙ３−ｄＵＴＰ（アマシャム−ファルマシア社製）、胃癌組織群にはＣｙ５−ｄＵＴＰ（アマシャム社製−ファルマシア社製）をそれぞれ使用した逆転写反応を行い、蛍光標識ｃＤＮＡを調製した。
【０１１６】
反応液組成を以下に示す。
反応液Ａ：ｍＲＮＡ約１μｇ、３００ｐｍｏｌのオリゴｄＴプライマー（宝酒造社製）及び最終的に１１．９μｌになるようジエチルピロカーボネート（ＤＥＰＣ、ナカライテスク社製）処理水を添加；
反応液Ｂ：５×ＡＭＶ ＲＴａｓｅ用緩衝液（ライフサイエンス社製）４μｌ、０．１ｍＭのｄＡＴＰ、ｄＣＴＰ、ｄＧＴＰ及び０．０６５ｍＭのｄＴＴＰ、６０ＵのＲＮａｓｅインヒビター（宝酒造社製）、０．０３５ｍＭのＣｙ３又はＣｙ５標識ｄＵＴＰ（アマシャムファルマシア社製）を混合し、最終容量６．５μｌの溶液を得た。
【０１１７】
反応液Ａを７０℃で５分間保持した後、氷浴上で冷却した。その後、前記反応液Ａに反応液Ｂ（６．５μｌ）とＡＭＶ ＲＴａｓｅ（ライフサイエンス社製）３０Ｕとを加え、ＲＴ反応液を得た。このＲＴ反応液を５５℃で３０分間保持した後、温度を４２℃とした。前記ＲＴ反応液に、さらに３０ＵのＡＭＶ ＲＴａｓｅを追加して液量を２０μｌに調整した後、得られた混合液を再度４２℃で６０分間保持した。得られた反応液に５００ｍＭのＥＤＴＡ溶液 ２．５μｌと１Ｍの水酸化ナトリウム ５μｌとを加え、３７℃で１０分間保持し、鋳型ＲＮＡを分解させた。反応液を室温まで冷却した後、１Ｍのトリス−塩酸（ｐＨ７．５）を１２．５μｌ添加した。この溶液をＣｅｎｔｒｉ−Ｓｅｐ ｓｐｉｎ ｃｏｌｕｍｎ（アプライド・バイオシステムズ社製）を用いてゲルろ過した。これにより、Ｃｙ３標識ｃＤＮＡ溶液（約３５μｌ）及びＣｙ５標識ｃＤＮＡ（約３５μｌ）を得た。
【０１１８】
こうして得られたＣｙ３標識ｃＤＮＡ、Ｃｙ５標識ｃＤＮＡを同一患者ごとに対になるように混合してエタノール沈殿濃縮し、ハイブリダイゼーションバッファー（６×ＳＳＣ／０．２％ＳＤＳ／５×デンハート液／０．１ｍｇ／ｍｌサケ精子ＤＮＡ）８μｌに溶解して、蛍光標識ｃＤＮＡを調製した。
【０１１９】
（３）標識ｃＤＮＡとＤＮＡアレイとのハイブリダイゼーション
上記（１）で作成したＤＮＡアレイ、ならびに市販のＤＮＡアレイ（Ｉｎｔｅｌｌｉｇｅｎｅ Ｃａｎｃｅｒ Ｃｈｉｐ 、宝酒造社製）にプレハイブリダイゼーションバッファー（６×ＳＳＣ／０．２％ＳＤＳ／５×デンハート液／１ｍｇ／ｍｌサケ精子ＤＮＡ）を滴下し、カバーグラスをかけて周囲をフィルムで密閉した。これを室温で２時間保持した後、カバーグラスを除いて、２×ＳＳＣで洗浄、次いで０．２×ＳＳＣで洗浄し、風乾した。
【０１２０】
（２）で調製した蛍光標識ｃＤＮＡを熱変性（９４℃、３分）後、その８μｌをプレハイブリダイズされたＤＮＡアレイに滴下し、カバーグラスをかけて周囲をフィルムで密閉した。得られたアレイを６５℃で１６時間保持した後、カバーグラスを除いて、２×ＳＳＣ／０．２％ＳＤＳ中で５５℃、３０分洗浄を２回、次いで６５℃、５分洗浄を１回、さらに０．０５×ＳＳＣ中で室温、１０分洗浄し、風乾した。得られたアレイをマイクロアレイスキャナー（ＧＭＳ社製）にかけて各スポットの蛍光シグナルを測定した。測定されたシグナルを発現データ解析ソフトウェアＩｍａｇｅｎｅ（バイオディスカバリー社製）で解析し、癌組織、対照正常組織における各遺伝子の発現量を調べた。
【０１２１】
使用したＤＮＡアレイに固定化された個々の遺伝子の、癌組織と対照正常組織における相対発現比率（［癌組織の発現強度シグナル／対照正常組織の発現強度シグナル］）、すなわち〔被検試料における遺伝子発現量／対照試料における遺伝子発現量〕は、１０人の胃癌患者によりそれぞれ複雑に異なっていた。しかし一部の遺伝子は、癌の浸潤程度、転移の有無又は進行度に依存した、発現パターンを示した。これらの遺伝子の中から以下のステップを経て、胃癌の悪性度に関連する遺伝子を見い出した。
【０１２２】
まず使用したＤＮＡアレイ上の遺伝子全ての癌組織と対照正常組織とにおける〔相対発現比率の自然対数変換値〕について、クラスタリングソフトＧｅｎｅＳｉｇｈｔ （ＢｉｏＤｉｓｃｏｖｅｒｙ 社製） を使用したＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ を実施した。各サンプル間の距離の測定法にはユークリッド距離を、距離の定義には重心法を用いた。なお、この解析により得られる樹形図において、縦軸は、検体の分類を示し、大きく、〔初期胃癌検体〕、〔癌が固有筋層又はしょう膜下組織まで浸潤しているが転移はしていない胃癌検体〕及び〔転移検体〕の３つのグループに分類されることが示される。また、樹形図の横軸は遺伝子の分類を示しており、これらは、〔すべての胃癌患者癌組織で発現強度比の低い遺伝子群〕、〔すべての胃癌患者癌組織で発現強度比の高い遺伝子群〕及び〔胃癌患者ごとに発現パターンの異なる遺伝子群〕の３分類に分類されることがわかる。
【０１２３】
胃癌の悪性度に関連する遺伝子は、前記３分類のうち、〔胃癌患者ごとに発現パターンの異なる遺伝子群〕に属すると考えられたため、本グループの遺伝子をまず候補として選択した。これら候補遺伝子から、癌の発現強度シグナル及び対照正常組織の発現強度シグナルがバックグラウンド値の２倍より小さい遺伝子は、発現が低く変動の評価信頼性が低いため除去した。胃癌患者ごとに発現パターンの異なる候補遺伝子群は、およそ３から１０遺伝子を１グループとしてサブクラスタリングしているため、次に、選択遺伝子からサブクラスター遺伝子のグループごとに除いていき、残りの遺伝子を用い同様にクラスタリングしていった。その結果、検体間の分類と悪性度との相関性が失われるサブクラスタリンググループを見出した。このような遺伝子はその発現が胃癌の悪性度に密接に関連していると考えられる。
【０１２４】
上記ステップを経て選択された１０５の遺伝子名とそのアクセッション番号は、前出の表１及び２に示されたものと同様である。
【０１２５】
これら１０５遺伝子を用いたＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により得られた樹形図を図１及び２に示す。図１は、胃癌患者の分類を示す。また、図２は、各胃癌患者における各遺伝子の発現パターンの樹形図を示す。図１に示すように、癌の深達度が低くリンパ管侵襲の見られない初期胃癌患者（ＭＫ３８及びＭＫ１６８ ）と、侵襲の見られる中程度から高度の進行胃癌検体とに大きく分かれることがわかる。また、後者は、癌が深く浸潤しリンパ節に転移している検体群（ＭＫ１５３ 、ＭＫ４４及びＭＫ１４３ ）とＭＫ１４４ 、ＭＫ８０及びＭＫ９６のサブクラスターを含むことがわかる。これらのうちＭＫ９６は、癌の深達が比較的低く、転移が見られなかったが、手術切片を顕微鏡観察したところ、リンパ管に癌が軽度に侵襲していた。現在、手術後４年１０カ月経過するが生存中である。ＭＫ８０も手術時転移は見られなったが癌が深く浸潤し、摘出された手術切片の顕微鏡観察でリンパ管に癌が高度に侵襲していた。その後、癌が再発し、死亡している。ＭＫ１４４ は、リンパ節にも腹膜にも転移している悪性度の高い癌患者であった。
【０１２６】
ＤＮＡアレイ上の遺伝子の癌組織における発現強度を指標とした、Ｈｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ も実施した。癌組織における発現強度のｎｏｒｍａｌｉｚａｔｉｏｎ にはＺ−ｓｃｏｒｅ 法を、サンプル間の距離の測定法にはシィティーブロック距離を、距離の定義には重心法を用いた。得られた樹形図を図３及び４に示す。縦軸の患者検体の分類（図３）は発現比率を用いた場合とわずかに異なるものの、１０人の胃癌患者は、やはり癌の深達度が低くリンパ管侵襲の見られない初期胃癌患者とリンパ管侵襲及び／又は転位の見られる中程度から高度の進行胃癌検体の２グループに大きく分かれた。さらに後者は転移の見られないＭＫ８０及びＭＫ９６のサブクラスターを含み、転移検体がこのサブクラスターに複雑に絡んでいた（図４）。
【０１２７】
癌組織の発現強度を用いた場合と、癌組織との対照正常組織の相対発現比率を用いた場合のＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ が似通っておりかつ胃癌の悪性度と相関していることから、これら１０５遺伝子の癌組織における発現変化は個体差による変化ではなく、胃癌の悪性化に関連していると判断された。
【０１２８】
実施例２
（１）ＤＮＡアレイの作製
実施例１で選択された１０５種類の遺伝子について、実施例１記載の方法により、約１００ヌクレオチド長〜約１キロヌクレオチド長）のｃＤＮＡ断片を調製し、該ＤＮＡ断片をスポットしたＤＮＡアレイを作製した。
【０１２９】
（２）標識ｃＤＮＡとＤＮＡアレイとのハイブリダイゼーション
実施例２で調製した蛍光標識ｃＤＮＡを使用し、上記（１）で作成したＤＮＡアレイとのハイブリダイゼーションを実施した。ハイブリダイゼーションの条件は実施例１と同様の条件とした。
【０１３０】
ハイブリダイゼーションの結果に基づき、得られた１０５遺伝子についての蛍光シグナルのうち、表４に示す２０種の遺伝子を選択し、該２０種の遺伝子について、実施例１と同様にＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ を実施した。
【０１３１】
【表４】

【０１３２】
なお、癌組織の対照正常組織に対する各遺伝子の〔相対発現比率の自然対数変換値〕を用い、ｎｏｒｍａｌｉｚａｔｉｏｎ 無し、サンプル間の距離の測定はユークリッド距離、距離の定義は重心法により解析した。得られた樹形図を図５及び６に示す。その結果、１０人の胃癌患者は非転移群と転移群の２グループに分かれることがわかる（図５）。また、非転移群は、初期胃癌患者（ＭＫ３８とＭＫ１６８ ）と、リンパ管侵襲の見られるＭＫ８０及びＭＫ９６とにサブクラスタリングされることがわかる。転移群のサブクラスターを構成する検体はいずれも広範囲のリンパ節に転移が認められた高進行癌の検体であることがわかる。このように、表４に示す２０個の遺伝子の組み合わせで、悪性度と密接に関連したクラスタリングが可能であることが示される。本実験よりこれら２０種の遺伝子の胃癌の悪性度への関与が再確認された。
【０１３３】
実施例３
（１）標識ｃＤＮＡとＤＮＡアレイとのハイブリダイゼーション
手術時の臨床病理データでは予後の予想のつきにくい悪性度レベルは、転移が認められないが、癌が胃の筋層に達した深達度ｍｐの胃癌である。臨床病理的にこのレベルに分類される胃癌患者ＭＫ１１５ の胃癌組織及び対照正常胃組織から、実施例１記載の方法でｍＲＮＡを精製し、さらに蛍光標識ｃＤＮＡを調製した。この蛍光標識ｃＤＮＡと、実施例１で調製した蛍光標識ｃＤＮＡとを使用して実施例２（１）で作成されたＤＮＡマイクロアレイとのハイブリダイゼーションを実施した。なお、ハイブリダイゼーションの条件は、実施例１と同様の条件とした。
【０１３４】
前記胃癌患者ＭＫ１１５ の臨床病理データは、次の通りである：
進行度Ｉ、分化度ｍｏｄ 、深達度ｍｐ、リンパ節転移０ 、リンパ節侵襲０ 、静脈侵襲０ 、腹膜転移０ 、肝転移０ 、浸潤増殖様式α。
【０１３５】
ハイブリダイゼーションの結果、得られた１０５遺伝子についての蛍光シグナルのうち、表５に示す遺伝子８種を選択し、癌組織の対照正常組織に対する相対発現比率を求めた。癌組織における発現が対照正常組織の２倍以上上昇している遺伝子、すなわち相対発現比率が２以上の遺伝子の相対発現比率の総和又は相対発現比率が２以上の遺伝子数にて悪性度を判定した。
【０１３６】
標準癌検体として実施例１及び２で用いられた癌検体１０検体と上記検体を比較した。表５に結果を示す。
【０１３７】
【表５】

【０１３８】
胃癌患者ＭＫ１１５ はリンパ管侵襲、血管侵襲も全く見られない進行度Ｉｂにもかかわらず選択された８遺伝子の相対発現比率の総和と相対発現比率が２以上の遺伝子数はともに大きく進行癌と同レベルであった。同様に、相対発現比率が２以上である遺伝子の相対発現比率の総和も進行癌と同レベルであった。そこで、この患者の予後を調べてみると３年９ヶ月後にガンが再発し死亡していた。通常、リンパ管侵襲、血管侵襲も見られない進行度Ｉの患者がこのような早期に再発することはまれであり、３年後の癌の再発が既に手術時の遺伝子の発現変化で予想できることは意義深い。
【０１３９】
ここで選択した８遺伝子は、転移や悪性度との相関が示唆されている遺伝子が多いにもかかわらず、転移、非転移という観点で見ても該遺伝子の発現は、検体ごとに異なり均一ではない。例えば、マトリリシン（ＭＭＰ−７）は、胃癌の悪性との関連性が高いとの報告〔セノタＡ．ら、クリニカル＆エクスペリメンタルメタスタシス（Ｃｌｉｎｉｃａｌ ＆ Ｅｘｐｅｒｉｍｅｎｔａｌ Ｍｅｔａｓｔａｓｉｓ）、第１６巻、第３１３−３２１頁（１９９８）〕があるが、今回測定した１１人の患者のうち４人で過剰発現していたに過ぎない。しかも、悪性度の高い転移検体でも半数で発現増加が見られるのみである。この結果より明らかなように、単一の遺伝子のみでは精度の高い悪性度判定は不可能であり、悪性度と関連する遺伝子の複数の組み合わせによって、予後の判定が可能になる。
【０１４０】
このように本発明は、従来の病理診断では予測不可能であった癌の悪性度や予後の診断に有効であることが明らかとなった。
【０１４１】
実施例４ 悪性度の指標となる遺伝子の検索
実施例１−（２）記載の胃癌患者由来の試料（胃癌組織と対照正常胃組織）のうちの８種と、新たに１４人の患者から得られた試料とを使用し、実施例１と同様にｍＲＮＡを抽出、精製した。さらに、得られたｍＲＮＡを鋳型とし、蛍光標識ｃＤＮＡを調製した。これらの標識ｃＤＮＡについて、実施例２−（１）で作成したＤＮＡアレイ、ならびに市販のＤＮＡアレイ（Ｉｎｔｅｌｌｉｇｅｎｅ Ｃａｎｃｅｒ Ｃｈｉｐ 、宝酒造社製）を使用したハイブリダイゼーションを実施し、その結果を実施例１−（３）と同様に解析した。
【０１４２】
上記のハイブリダイゼーションの結果、胃癌の悪性度に関与するサブクラスタリンググループに属する７０種の遺伝子が見出された。このサブクラスタリンググループに属する遺伝子をさらに詳しく観察すると、悪性度が上がるにつれ大きく発現が増加する遺伝子；悪性度が上がるにつれ発現が増加する傾向はあるものの、同一進行レベルにある検体間にばらつきがある遺伝子；悪性度が上がるにつれ遺伝子発現が下がる遺伝子などが含まれていることがわかった。そこで、これらの遺伝子を重み付け係数を用いて重み付けすることにより、悪性化への寄与率を反映したポイント診断が可能になるよう試みた。
【０１４３】
悪性度に相関してその発現量が上昇する遺伝子について、その上昇の度合いに応じて５点〜１点の重み付け係数を配点し、また、逆に悪性度が上がるにつれ遺伝子発現量が下がる遺伝子には−１または−３の配点を行なった。対照試料に対する被検試料における遺伝子の相対発現比率の対数に、上記重み付け係数を掛け合せ、７０種の遺伝子のポイントの総和を求めた。
【０１４４】
癌の深達度や進行度といった悪性化因子で判定される悪性度に比例してポイントの総和が上昇し、さらに５年以内に再発及び／又は死亡した検体が１００ 点以上のポイントとなるまで、個々の遺伝子の重み付け係数の配点のやり直しを繰り返した。
【０１４５】
上記の７０遺伝子の名称と、上記の作業の結果、最終的に得られた重み付け係数を表６及び表７に示す。
【０１４６】
【表６】

【０１４７】
【表７】

【０１４８】
また、今回の遺伝子発現量の測定に使用された２２種の試料の臨床病理データ、ならびに各試料の遺伝子発現量の変化について、上記の重み付け係数を加味したポイント合計を表８に示す。なお、表８中の各項目は、前記表３と同様である。
【０１４９】
【表８】

【０１５０】
表８に示されるように、上記の手法で算出されたポイントと、検体試料の病理学的な所見から判定される癌悪性度との間には明確な相関があり、当該ポイントによって癌の悪性度を評価することが可能であることがわかる。
【０１５１】
さらに、実施例３に記載の胃癌患者ＭＫ１１５ 由来の試料について上記の７０種の遺伝子の発現量の変化を調べ、ポイントを算出したところ、１３８．６のポイントが得られた。すなわち、臨床的には悪性度が低く診断される胃癌においても、上記の方法により正確に悪性度が判定できることが明らかとなった。
【０１５２】
【発明の効果】
本発明の癌、特に胃癌の悪性度の評価方法、癌、特に胃癌の悪性度を評価するためのアレイ及び癌、特に胃癌の悪性度を評価するためのキットにより、癌、特に胃癌の悪性化に関与する多数の遺伝子の発現変化を迅速、かつ高感度に系統的に測定でき、これらの遺伝子発現変動のパターン解析により、癌、特に胃癌の悪性度の判定、又は癌、特に胃癌の検出をすることができるとい優れた効果を奏する。
【図面の簡単な説明】
【図１】図１は、１０５遺伝子の〔相対発現比率の自然対数変換値〕を用いたＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により得られた胃癌患者の分類の樹形図を示す。
【図２】図２は、１０５遺伝子の〔相対発現比率の自然対数変換値〕を用いたＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により得られた各胃癌患者における各遺伝子の発現パターンの樹形図を示す。
【図３】図３は、１０５遺伝子の発現強度を用いたＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により得られた胃癌患者の分類の樹形図を示す。
【図４】図４は、１０５遺伝子の発現強度を用いたＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により得られた各胃癌患者における各遺伝子の発現パターンの樹形図を示す。
【図５】図５は、２０遺伝子の〔相対発現比率の自然対数変換値〕を用いたＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により得られた胃癌患者の分類の樹形図を示す。
【図６】図６は、２０遺伝子の〔相対発現比率の自然対数変換値〕を用いたＨｉｅｒａｒｃｈｉｃａｌ ｃｌｕｓｔｅｒｉｎｇ により得られた各胃癌患者における各遺伝子の発現パターンの樹形図を示す。[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method for evaluating malignancy of cancer, particularly gastric cancer, which can easily and quickly evaluate malignancy of cancer with high reliability, and an array and a kit used for the method.
[0002]
[Prior art]
Cancer has been the leading cause of death in Japan since 1981, with gastric cancer being the most frequent cancer. In recent years, it has been clarified that a multi-stage carcinogenesis mechanism exists until normal cells become cancerous [Feelon E. et al. R. (Fearon, ER) et al., Cell, Vol. 61, pp. 759-767 (1990); (Sugimura, T.), Science, Vol. 258, pp. 603-607 (1992)]. Specifically, canceration of normal cells requires accumulation of multiple gene abnormalities including DNA repair genes, tumor suppressor genes, and oncogenes. In general, gene instability and inactivation of tumor suppressor genes are considered to be involved in the development of cancer. In addition, activation of oncogenes and / or overexpression of growth factors is considered to be involved in cancer progression and malignancy. Furthermore, genes encoding enzymes that degrade extracellular matrix molecules and genes encoding proteins that regulate cell motility or adhesion are thought to be involved in metastasis and invasion.
[0003]
Gene instability includes gene instability and chromosome-level instability associated with abnormalities in the DNA mismatch repair system. As an example of the former, the chain length of a simply repeated sequence present in the genome is different between a cancerous part and a non-cancerous part of the same individual (microsatellite instability) [Sybodo S. et al. N. (Thibodeau, SN) et al., Science, Vol. 260, pp. 816-819 (1993)], and examples of the latter include rearrangement between chromosomes. The rearrangement between the chromosomes causes a protein that is not found in normal cells to be expressed, and even a protein that is expressed in normal cells may affect the expression level. In fact, in human chronic myeloid leukemia, fusion between the bcr gene and the c-abl gene occurs due to interchromosomal rearrangement, and the expression of hybrid mRNA transcribed from the bcr-abl fusion gene, which is not present in normal cells, was confirmed. ing. Furthermore, it has been confirmed that leukemia develops when the bcr-abl fusion gene is introduced into animals [Watson, JD, et al., Molecular Biology of Recombinant DNA, Second Edition; 309 (1992)].
[0004]
Examples of the inactivation of the tumor suppressor gene include inactivation of the p53 gene. The cause of inactivation is thought to be due to deletions in the gene or point mutations occurring in specific parts of the coding region [Negro J. et al. M. (Nigro, JM) et al., Nature, 342, 705-708 (1989), Malkin, D., et al., Science, 250, 1233-1238 ( 1990)]. Also, p53 gene deletions and point mutations have been observed in many types of cancer. For example, in gastric cancer, the p53 gene deletion and point mutation are observed in 60% or more cases of early cancer [Yokozaki H. et al. (Yoko zaki, H.) et al., Journal of Cancer Research and Clinical Oncology, Vol. 119, pp. 67-70 (1992)].
[0005]
On the other hand, the p16 / MTS1 gene is known as a gene that is inactivated by homo-deletion. Specifically, high-frequency homo-deletion has been observed in glioma, pancreatic cancer, bladder cancer, etc. [Cairns P. et al. (Cairns, P.) et al., Nature Genetics, Vol. 11, pp. 210-212 (1995)]. The p16 protein regulates the cell cycle. It has been suggested that the abnormal expression of p16 is involved in canceration of cells [Okamoto A. et al. (Okamoto, A.) et al., Proceedings of the National Academy of Sciences of the United States, Vol. 19, Vol. 9, pp. 991-94, Amer.
[0006]
Activation of an oncogene includes, for example, insertion mutation of a virus near an oncogene or rearrangement between chromosomes. For example, viral insertion mutations have been identified in chicken lymphoma caused by the avian leukemia virus avian leukosis virus (ALV). In this case, ALV DNA is inserted in the vicinity of c-myc which is a kind of gene, and normal c-myc is overexpressed by a strong enhancer and promoter of the virus, or a new gene which is partially different from the normal gene. It was confirmed that the sequence was expressed. In addition, in some types of human B cell tumors, c-myc, one of the oncogenes, is placed under a strong immunoglobulin transcription signal due to interchromosomal rearrangement, and its expression level increases. Has been confirmed. In this case, the expression of the c-myc expression product in the cancer cells was not different from that expressed in normal cells, and it is considered that the canceration was caused by an increase in the expression level of c-myc. [Watson, JD, et al., Molecular Biology of Recombinant DNA, Second Edition; Maruzen Co., Ltd., pp. 305-308 (1992)].
[0007]
Examples of overexpression of a growth factor include overexpression of C-Met encoding a hepatocyte growth factor receptor. This abnormal expression of C-Met is observed in the expression of mRNA having a length of 6.0 kilobases which is not observed in normal mucosa from the early stage of gastric cancer development [Kunias H. et al. (Kuniyasu, H.), et al., International Journal of Cancer, Vol. 55, pp. 72-75 (1993)], gene amplification is frequently observed in metastatic cancer, and gene amplification and cancer Correlation with malignancy has been confirmed [Knyas H. et al. Et al., Biochemical and Biophysical Research Communications, Vol. 189, pp. 227-232 (1992)].
[0008]
As examples in which a correlation has been confirmed between the genetic abnormality and the malignancy of cancer, in addition to the above-mentioned c-Met, amplification and / or overexpression of the oncogene C-erbB2 gene in breast cancer, ovarian cancer, gastric cancer, uterine cancer and the like [Light C. (Wright, C.), et al., Cancer Research, 49, 2087-2090 (1989); (Saffari, B.), et al., Cancer Research, Vol. 55, pp. 5693-5698 (1995)], amplification of the oncogene K-sam gene and / or its amplification in poorly differentiated adenocarcinoma, one tissue type of gastric cancer. Overexpression [Tahara E. et al. (Tahara, E.) et al., Gastric Cancer, Tokyo, Springer-Verlag, 1993, pp. 209-217, and the like, respectively.
[0009]
In order for cancer cells to invade tissues and metastasize, extracellular matrix must be degraded, and various enzymes belonging to the matrix metalloproteinase (MMP) family, which are typical degrading enzymes, are overproduced compared to normal cells. [Goldberg GI. (Goldberg GI.) Et al., Cancer Treatment Research, Vol. 53, pp. 701-708 (1991)]. In gastric cancer, it has been reported that the mRNA of matrilysin (MMP-7), which is relatively specifically expressed by cancer cells, was overexpressed in 60% or more of the cases [Senota A. et al. Et al., Clinical & Experimental Metastasis, Vol. 16, pp. 313-321 (1998)].
[0010]
Malignant tumors cannot grow unless tumor blood vessels are formed, and can induce accelerated growth by inducing angiogenesis. Tumor angiogenesis is controlled by the expression balance of various regulatory factors, angiogenesis promoting factors and suppressors, and the most important factor is vascular endothelial cell growth factor VEGF. A positive correlation has been observed between VEGF expression and enhanced angiogenesis in various solid tumors [Shibuya M et al., Advanced Cancer Research, 67, 281 (1995)], and gastric cancer. It has been reported that the prognosis of VEGF-overexpressing tumors is poor in esophageal cancer, colorectal cancer and the like [Toy M et al., Molecular Medicine, Vol. 35, No. 1206-1215 (1998)].
[0011]
There has recently been the first report that cancer classification, which was previously only possible by pathological diagnosis, has been made possible by simultaneously measuring the expression of a large number of genes using a DNA microarray. That is, Golab T. R. Et al. Screened 6817 genes of humans immobilized on a DNA microarray and showed that classification of acute leukemia (ALL or AML) was possible mainly by changing the expression level of 50 genes [Science (Science) 286, pp. 531-537 (1999)].
[0012]
As described above, at the level of each gene, information on genes involved in the development and progression of cancer and information on the gene abnormality are increasing. However, since the mechanism of carcinogenesis is multi-stage and requires the accumulation of multiple mutations, the determination of a single gene or a random combination of a small number of genes is still sufficient for practical use in determining cancer. At present, diagnosis and prognosis have not been determined.
[0013]
In fact, even today, diagnosis of cancer and determination of the degree of progress and differentiation are all performed by pathological diagnosis. However, some pathologically similar cancers exhibiting the same degree of progress and differentiation have good prognosis cases, while there are cases of high malignancy that relapse or metastasize early. In some cases, there are many cases showing susceptibility to anticancer drugs and radiation irradiation, while others showing resistance. Therefore, at present, it is impossible to predict these before treatment or early after surgery.
[0014]
[Problems to be solved by the invention]
Predicting the diagnosis, progression, differentiation, etc. of cancer makes it possible to avoid unnecessary treatment, minimizes the burden on patients, and greatly contributes to setting treatment policies suitable for individual patients. . It is also thought that this will lead to a reduction in medical expenses.
[0015]
Therefore, a first object of the present invention is to provide a gene that can be an index for determining the degree of malignancy of cancer, particularly gastric cancer, which has a large number of patients, and in particular, a large number of genes whose expression status changes as the cells become cancerous and progress. It is an object of the present invention to provide a method for evaluating cancer malignancy or a method for detecting cancer based on analysis of the expression pattern of the gene, which can clarify, easily and quickly evaluate malignancy of cancer with high reliability. A second object of the present invention is to provide an array or a kit used for evaluating the malignancy of the above-mentioned cancer or detecting cancer cells.
[0016]
[Means for Solving the Problems]
In order to achieve the above object, the present inventors not only individually compare the intracellular expression levels of a large number of genes in cancer tissues of gastric cancer patients and control normal tissues, but also analyze the expression patterns of gastric cancers by analyzing gene expression patterns. Numerous genes involved in transformation were found. The present inventors have devised an appropriate technique for analyzing the pattern of gene expression using a combination of these genes, and have found that it is possible to evaluate the degree of malignancy of cancer. Thus, the present invention has been completed.
[0017]
That is, the gist of the present invention is:
[1] The following steps:
(1) measuring, in a test sample, the expression levels of at least eight genes whose expression levels change depending on the degree of malignancy of gastric cancer; and
(2) a step of evaluating the degree of malignancy of gastric cancer based on the expression level of each gene obtained in (1)
Evaluation method of malignancy of gastric cancer, characterized by performing
[2] Evaluation of gastric cancer malignancy, in which nucleic acids or fragments thereof corresponding to at least eight types of genes whose expression levels change according to the gastric cancer malignancy are immobilized in defined regions on a support, respectively. An array for
[3] Evaluating gastric cancer malignancy comprising primers and / or probes for detecting mRNA or fragments thereof expressed from at least eight genes whose expression level changes depending on the gastric cancer malignancy Kits for doing
[4] A kit for evaluating the malignancy of gastric cancer, comprising an antibody or a fragment thereof to a polypeptide or a fragment thereof encoded by at least eight genes whose expression level changes according to the malignancy of gastric cancer ,
About.
[0018]
BEST MODE FOR CARRYING OUT THE INVENTION
As used herein, the term “malignancy of cancer” is determined by the degree of invasion and the presence or absence of metastasis as primary factors, and as a result, refers to a factor that affects the 5-year survival rate after treatment such as surgery.
[0019]
Specifically, cancer remains in the mucosa and submucosa, and cancer without metastasis is determined to be low-grade cancer, and a 5-year survival rate is predicted to be high (90% or more). On the other hand, if the cancer invades deeply from the submucosal layer to the muscular layer, the subserosa layer, and the serosal surface, and the cancer has metastasized lymphatic, hematogenous, invasive, or disseminated, the malignancy is determined to be high and the patient survives for 5 years The rate is low. In particular, when disseminated metastasis to the peritoneum, liver metastasis, or metastasis to distant lymph nodes is observed, or when cancer invasion extends to other organs, the prognosis is very poor and the 5-year survival rate is low (20% or less). ). In addition, the presence or absence of sensitivity to an anticancer agent or radiation is raised as a secondary factor, the degree of malignancy is low if there is sensitivity to the anticancer agent or radiation, and the degree of malignancy is high if there is resistance Judge.
[0020]
1. Method for evaluating malignancy of cancer of the present invention, particularly gastric cancer
The method for evaluating the malignancy of cancer, particularly gastric cancer according to the present invention comprises the following steps:
(1) measuring, in a test sample, the expression levels of at least eight genes whose expression levels change depending on the degree of malignancy of cancer, particularly gastric cancer, and
(2) a step of evaluating the malignancy of cancer, particularly gastric cancer, based on the expression level of each gene obtained in (1)
Is one feature. According to the evaluation method of the present invention, the expression levels of at least eight genes whose expression levels change according to the degree of malignancy of gastric cancer described below are measured easily and quickly with high reliability. It has an excellent effect of being able to evaluate the degree of malignancy.
[0021]
Examples of the test sample include a sample derived from a living body such as blood; urine; feces; and a tissue extracted by a surgical technique. In the method for evaluating the malignancy of cancer, particularly gastric cancer, of the present invention, a lesion tissue collected with a biopsy forceps is desirable from the viewpoint of preoperative diagnosis. In this specification, an individual serving as a supply source of the test sample may be referred to as a “sample”.
[0022]
The `` gene whose expression level changes in accordance with the degree of malignancy of gastric cancer '' used in the evaluation method of the present invention is a gene that can serve as an index of canceration, an evaluation of the degree of malignancy of cancer. A gene whose expression level changes, in other words, a gene whose expression is significantly induced or suppressed.
[0023]
Such genes can be analyzed, for example, by analyzing the copy number of genes in the genome and the pattern of chromosomal rearrangement, comparing the expression levels of gene products in normal cells and cancerous cells, and determining the differences between the two cells. It can be detected by specifying a certain thing. Examples of the gene product include mRNA transcribed from the gene and a protein that is a translation product. In the detection of the gene used in the present invention, it is efficient to use, as an index, mRNA, for which various techniques have been developed for analysis with the advance of gene manipulation technology.
[0024]
Techniques for confirming changes in gene expression using mRNA as an index include Northern hybridization, RT-PCR, subtraction, differential display, and the like. These methods are appropriately selected and used in the present invention. Can be found. Further, as a method for simultaneously detecting changes in the expression of a large number of genes, such as hundreds or thousands, a method using a DNA array (DNA chip hybridization analysis, DNA macroarray hybridization analysis, etc.) is known.
[0025]
The “gene whose expression level changes according to the degree of malignancy of gastric cancer” can be obtained by using a DNA array on which a nucleic acid corresponding to a human-derived gene or a fragment thereof is immobilized. For example, a DNA microarray in which a fragment of a gene that is suggested to be related to cancer is immobilized is commercially available (IntelliGene Human Cancer CHIP, manufactured by Takara Shuzo Co., Ltd.). Can be found. The “gene whose expression level changes according to the degree of malignancy of gastric cancer” can be obtained, for example, by the following gene selection method.
[0026]
Gene selection method :
For example, mRNA is prepared from a cancer lesion tissue collected from a cancer patient, and a reverse transcription reaction is performed using the mRNA as a template. At this time, a labeled cDNA can be obtained by using, for example, a primer or a labeled nucleotide with an appropriate label. The labeling method is not particularly limited, and examples thereof include a compound containing a radioisotope, a fluorescent substance, and a label using a ligand such as biotin.
[0027]
Next, hybridization is performed between the labeled cDNA and a DNA array on which a nucleic acid corresponding to an appropriate gene or a fragment thereof is immobilized. Hybridization may be performed by a known method, and the conditions may be appropriately selected as appropriate for the DNA array or labeled cDNA to be used. For example, it can be carried out under the conditions described in Molecular Cloning, A Laboratory Manual, Second Edition, pp. 9.52-9.55 (1989).
[0028]
Further, under the same hybridization conditions as described above, nucleic acid derived from a control sample (a sample derived from a healthy person, a sample derived from a non-lesion site, etc.) is hybridized with the DNA array. These human samples take time from collection to measurement of the gene expression level, and mRNA may be degraded by the action of RNase. In order to measure the difference in gene expression between the test sample and the control sample, it is necessary to correct the amounts of both mRNAs using a standard gene having relatively little variation in expression. Furthermore, when competitive hybridization is performed on one DNA array using two types of fluorescence described below, it is necessary to correct the difference in intensity between the two types of fluorescent substances. The nucleic acid used for the purpose of such correction includes a nucleic acid derived from a non-lesion site; a housekeeping gene [for example, a glyceraldehyde 3-phosphate dehydrogenase (GAPD) gene, a cyclophilin gene, a β-actin gene, an α- Tubulin gene, phospholipase A2 gene, etc.], and a plasmid pUC18 as a negative control for confirming non-specific hybridization.
[0029]
Then, by comparing the results of the hybridization with the sample derived from the cancer lesion and the results of the hybridization with the control sample, genes having different expression levels can be detected in both cases. Specifically, for an array hybridized with the nucleic acid sample labeled by the above method, appropriate signals are detected according to the labeling method used, and a sample derived from a cancer lesion site of each gene on the array is detected. And the amount of expression in a control sample can be compared. Preferably, when using a multi-wavelength detection fluorescence analyzer that can detect a plurality of labels, for example, two types of fluorescence, the difference between the gene expression level in a sample derived from a cancer lesion and the gene expression level in a control sample is the same. Comparison can be performed by competitive hybridization on a DNA array. For example, a sample derived from a cancer lesion is fluorescently labeled with Cy5-dUTP, and a control nucleic acid sample is fluorescently labeled with Cy3-dUTP. By mixing the two in equal amounts and performing hybridization with the DNA array, the difference in the gene expression level between the two can be detected as the difference in signal color and fluorescence intensity.
[0030]
Genes having a significant difference in signal intensity obtained in this way are genes whose expression levels change with the transformation of cells into cancer, and are genes that can be used as indicators of canceration and cancer malignancy. The “gene whose expression level changes in accordance with the degree of malignancy of gastric cancer” used in the present invention is preferably an index as a gene having a larger expression level difference.
[0031]
The criteria for selecting a target gene in the gene selection method described above are as follows:
(I) the expression intensity in the test sample or control sample is at least twice the expression intensity of the negative control gene;
(II) High correlation with cancer malignancy
Is mentioned. A specific sorting method is described in Example 1.
[0032]
The gene obtained as described above may be any gene whose expression changes as the cells become cancerous. For example, (1) a group of genes involved in oncogenes and tumor suppression (oncogenes & tumor suppressors); (2) a group of genes involved in nuclear receptor or nuclear receptor transcription coupling, (3) a group of genes involved in kinase-type transmission, (4) a group of genes involved in receptor-type kinase, (5) an intermediate filament marker (intermediate) (6) genes involved in cell cycle and growth regulators, (7) genes involved in apoptosis, (8) DNA damage reaction, repair , Re Genes involved in DNA damage response, repair & recombination, (9) genes involved in receptors, (10) genes involved in cell death & development regulators, and (10) genes involved in cell death & development regulators (11) Genes involved in cell adhesion, motility and invasion (cell adhesion, mobility & invasion), (12) Genes involved in promotion of angiogenesis (angiogenesis regulators), (13) Genes involved in cell invasion (invasion regulators) Genes involved in (14) genes involved in cell-cell interactions, (15) Rho Genes involved in family GTPases and their regulators (Rho family small GTPases & theil regulator), (16) Genes involved in growth factors & cytokines, (17) Genes involved in energy metabolism And the like. More specifically, Tables 1 and 2 below:
[0033]
[Table 1]

[0034]
[Table 2]

[0035]
The expression levels of 105 genes listed in Table 6 and 51 genes further listed in Tables 6 and 7 described below can be measured to evaluate the malignancy of cancer.
[0036]
In the method for evaluating the malignancy of cancer, particularly gastric cancer, of the present invention, from the viewpoint of obtaining an accurate evaluation, among the above genes, the genes whose expression levels are measured in the present invention are at least eight, and preferably at least eight. The number is preferably 20 or more, more preferably 30 or more, still more preferably 40 or more, and particularly preferably 50 or more. Ease of operability in the evaluation method, in particular, gene expression by Hierarchical clustering described later From the viewpoint of easiness of operation when performing expression pattern analysis, it is desirable that the number be 100 or less.
[0037]
As a specific example of the “gene whose expression level changes according to the degree of malignancy of gastric cancer”, preferably, an apoptosis inhibitor survivin (apoptosis inhibitor survivin (Genebank registration number: U75285)) gene,
Type I collagen α-2 [collagen type I α-2 (Genebank accession number: J03464)] gene,
Type III collagen pro-α-1 chain [collagen type III pro-alpha-1 (Genebank accession number: X14420)] gene,
A fibronectin precursor (FN) [fibronectin precursor (FN) (Genebank accession number: X02761)] gene,
MMP-1; gastrointestinal collagenase [matrix metalloproteinase 1; intestinal collagenase (Genebank accession number: M13509)] gene,
MMP-7: uterine matrilysin [matrix metalloproteinase 7; matrilysin, uterine (Genebank accession number: Z11887)] gene,
U-plasminogen activator (EC 3.4.21.73); U-plasminogen activator (UPA); plasmin activator;
Urokinase-type plasminogen activator receptor; GPI-anchored from precursor (U-PAR); monocyte activation antigen gene;
Alpha-2-macroglobulin [alpha-2-macroglobulin (alpha-2-M) (Genebank accession number: M11313)] gene,
β-type platelet-derived growth factor receptor precursor (beta platelet-derived grown factor receptor precursor (PDGFR-beta); CD140B antigen (Genebank registration number: J03278)) gene,
BIGH3 (Genebank accession number: M77349) gene,
XVIII type collagen α-1 chain [collagen type XVIII alpha-1 (Genebank accession number: AF0188081)] gene,
A growth hormone-dependent insulin-like insulin-like growth factor-binding protein (Genebank registration number: M35878)] gene,
Hepatoma-derived growth factor (HDGF) (Genebank accession number: D16431) gene,
An insulin-like growth factor binding protein 2 (IGFBp2) (Genebank accession number: X16302)] gene,
Type IV collagenase [type IV collagenase (Genebank accession number: M55593)] gene,
spac precursor (osteonectin) [sparc precursor (secrete protein acidic and rich in cysteine; osteonectin) (ON); basement membrane protein BM-40;
A thrombospondin 2 precursor [thrombospondin 2 precursor (Genebank accession number: L12350)] gene,
Chondroitin sulfate proteoglycan core protein 2 (verifiedcan sulfate protein proteoglycan core protein 2);
Embopplakin (Genebank accession number: U53786) gene,
Integrin β-4 chain [integrin, beta 4] (Genebank accession number: X53587) gene,
Vimentin (Genebank accession number: Z19554) gene,
c-fms proto-oncogene [genebank registration number: X03663],
Proliferating cell nuclear antigen (Genebank accession number: M15796) gene,
Leukocyte antigen-related protein (Genebank accession number: Y00815) gene,
Interleukin 8 [interleukin 8] (Genebank accession number: M26383) gene,
GDP dissociation inhibitor (GDI) [GDP dissociation inhibitor (GDI)] (Genebank accession number: L20688) gene,
A brain-specific tubulin α chain [Tubulin, alpha, brain-specific] (Genebank accession number: X01703) gene,
Cell cycle control gene cdc2 (cell division cycle 2, G1 to S phase & G2 to M phase) [cell division cycle 2, G1 to S and G2 to M] (Genebank registration number: X05360),
Tissue-type plasminogen activator [plasminogen activator, tissue] gene bank accession number: M15518] gene,
Stromal cell-derived factor 1 (Gene bank accession number: U16752) gene,
Matrix metalloproteinase 10 (stromelysin 2) (Genebank accession number: X07820) gene,
Matrix metalloproteinase 15 (membrane-inserted) (Genebank accession number: Z48482) gene,
Antioxidant protein [antioxidant protein 1] (Genebank accession number: D49396) gene,
Laminin α-4 chain [laminin, alpha 4] (Genebank accession number: S78569) gene,
The ubiquitin-conjugating enzyme E2M (homologous to yeast UBC12), which is homologous to yeast UBC12; Genebank accession number: AF057599;
Retinoblastoma-binding protein 6 (Genebank accession number: X85133) gene,
K-sum [K-sam] (Genebank accession number: M87770) gene,
Cathepsin B (genebank accession number: L22569) gene,
Signal transduction & transcriptional activator 1; 91 kD [signal transducer and activator of transscription 1, 91 kD] (Genebank accession number: M97935) gene,
Carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reacting) (carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reactive antigen)
antigen)] (Genebank accession number: M18728) gene,
Heparan sulfate proteoglycan (gene bank accession number: M85289) gene,
v-Ki-ras2 Kirsten rat sarcoma virus oncogene homolog [v-Ki-ras2 Kirsten rat sarcoma 2 viral oncogene homolog] (Genebank accession number: M54968) gene,
Ras homolog gene family, factor D [ras homolog gene family, member D] (Genebank accession number: D85815) gene,
Hexabrachion (tenascin C, cytotactin) (Genebank accession number: X78565) gene,
Cyclin D2 [cyclin D2] (Genebank accession number: D13639) gene,
Metalloproteinase 3 tissue inhibitory factor [tissue inhibitor of metalloproteinase 3 (Sorsby fundus dystrophy, pseudoinflammation)] (Genebank registration number: AL023282) gene,
Laminin receptor 1, ribosomal protein SA (laminin receptor 1 (67 kD, ribosomal protein SA)) (Genebank accession number: U43901) gene,
Lactate dehydrogenase A (Genebank accession number: X02152) gene,
Keratin 18 (Genebank accession number: M26326) gene,
Ia-related invariant gamma chain [Ia-associated invariant gamma-chain] (Genebank accession number: AH001484) gene,
CD72 antigen [CD72 antigen] (Genebank accession number: M54992) gene,
Connective tissue growth factor (Genebank accession number: X78947) gene,
Matrix metalloproteinase 3 (stromelysin 1, progelatinase) (Genebank accession number: X05232) gene,
Cyclin-dependent kinase 4 (Genebank accession number: U37022) gene,
A human SB class II histocompatibility antigen [Human mRNA for SB class II histocompatibility antigen alpha-chain] (Genebank accession number: X03100) gene;
Transforming growth factor β-3 chain [transforming growth factor, beta 3] (Genebank accession number: X14885) gene,
Thyroid hormone receptor interacting factor 6 [thyroid hormone receptor interacting]
protein 6] (Genebank accession number: AJ001902) gene,
Platelet / endothelial cell adhesion molecule (CD31 antigen) (Genebank accession number: L34657) gene,
Profilin 1 (Genebank accession number: J03191) gene,
GTP-binding factor [GTP-binding protein] (Genebank accession number: AF120334) gene,
Homo sapiens clone 24703 β-tubulin mRNA [Homo sapiens clone 24703 beta-tubulin mRNA, complete cds] (Genebank accession number: AF070600) gene,
A Rho-associated coiled-coil containing protein kinase 1 (Genebank accession number: U43195) gene,
Type IV collagen, α-1 chain [collagen, type IV, alpha 1] (Genebank accession number: M26576) gene,
Topoisomerase (DNA) I (gene bank accession number: M60706) gene,
Kunitz type 1 serine proteinase inhibitor, [serine protease inhibitor, Kunitz type 1] (Genebank accession number: AB000095) gene,
Rho GTPase activating protein 1 [Rho GTPase activating protein 1] (Genebank accession number: U02570) gene,
Nidogen (gene bank accession number: M30269) gene,
Cyclin-dependent kinase 5, regulatory factor 1 [cyclin-dependent kinase 5, regulatory subunit 1 (p35)] (Genebank accession number: X80343) gene,
CD9 antigen [CD9 antigen] (Genebank accession number: M38690) gene,
Aortic-type smooth muscle α-actin gene, exon 9 (aortic-type smooth muscle alpha-actin gene, exon 9 (Genebank registration number: M33216)),
Bone morphogenetic protein 1 (BMp1) (Genebank accession number: U50330) gene,
Type VI collagen α-3 chain [collagen type VI alpha-3 (Genebank accession number: X52022)] gene,
A cytokine humig (interkine-gamma-induced monokine (MIG) (Genebank accession number: X72755)) gene,
Integrin alpha-3 chain [integrin alpha-3 chain (Genebank accession number: M59911)] gene,
An interferon-inducible gene family (Genebank accession number: X57351) gene,
A metastasis-associated MTA1 [metastasis-associated MTA1 (Genebank accession number: U35113)] gene,
Notch2 Notch homolog 3 (Notch2 Notch homolog 3 (Genebank accession number: U97669)) gene, and
Proto-oncogene rhoA multidrug resistance protein (GTP-binding protein (rhoA) (Genebank accession number: L25080)) at least 8 genes selected from the group consisting of each gene; Preferably, at least 20 genes are used. In addition, specific examples of at least 8 types of genes and at least 20 types of genes are not particularly limited, and include genes described in Table 5 of Example 3 and genes described in Table 4 of Example 2 below. Can be
[0038]
The gene is also useful as an indicator for detecting cancer or cancer cells. Therefore, the method for evaluating the malignancy of cancer, particularly gastric cancer, according to the present invention also enables detection of cancer cells in cancer, particularly gastric cancer. Such a method for detecting cancer or cancer cells is also included in the scope of the present invention.
[0039]
In the method for detecting cancer or cancer cells, the expression of many of the selected genes fluctuates due to malignant transformation, and further, the variation patterns differ due to malignant transformation. The determination of malignancy can be performed at the same time.
[0040]
In the method of the present invention for evaluating the malignancy of cancer, particularly gastric cancer, the expression level of a gene can be measured from the amount of mRNA transcribed from the gene or the amount of polypeptide encoded by the gene.
[0041]
The mRNA amount is measured by various known methods, for example, a hybridization method represented by a Northern hybridization method, an in situ hybridization method, a dot blot hybridization method, a method using a DNA array, and the like. , A nucleic acid amplification method represented by an RT-PCR method and the like can be used. From the viewpoint of obtaining sufficient mRNA detection sensitivity, a hybridization method or a nucleic acid amplification method is preferable, and more specifically, a nucleic acid or a fragment thereof corresponding to a gene whose expression level is to be measured is defined on a support. A hybridization method using an array (for example, a DNA array) immobilized on the fixed region, and an RT-PCR method are preferable. From the viewpoint of simultaneously measuring the expression of many genes, a hybridization method using a DNA array is particularly preferable.
[0042]
As used herein, the term “DNA array” refers to a DNA array in which a gene or a DNA fragment derived from the gene is fixed to a predetermined region on a support, and includes, for example, a DNA chip. A gene or a DNA fragment derived from the gene which is fixed on a support at a high density is also called a DNA microarray.
[0043]
The support for the DNA array is not particularly limited as long as it can be used for hybridization, and usually a slide glass, a silicon chip, a nitrocellulose or nylon membrane or the like is used. The gene or its fragment immobilized on the support is not particularly limited, and examples thereof include a genomic DNA library and a cDNA library, and DNA amplified by PCR using these as a template.
[0044]
By using such a DNA array, the amounts of various types of nucleic acid molecules contained in a nucleic acid sample can be measured simultaneously. In addition, there is an advantage that a small amount of nucleic acid sample can be measured. For example, the mRNA expressed in the sample is detected simultaneously by labeling the mRNA in the sample or preparing a labeled cDNA using the mRNA as a template and performing hybridization between this and a DNA array. Then, the expression level can be measured.
[0045]
As a labeling substance used for labeling, a substance such as a radioisotope, a fluorescent substance, a chemiluminescent substance, and a substance having a luminophore can be used. For example, examples of the fluorescent substance include Cy2, FluorX, Cy3, Cy3.5, Cy5, Cy5.5, Cy7, fluorescein isothiocyanate (FITC), Texas Red, and rhodamine. From the viewpoint of simultaneous detection, it is desirable to label a test sample and a sample used as a control with different fluorescent substances using two or more types of fluorescent substances. Here, the labeling of the sample is carried out by labeling mRNA in the sample, cDNA derived from the mRNA, or nucleic acid transcribed or amplified from the cDNA.
[0046]
The method for detecting the label can be appropriately selected depending on the type of the labeling substance used. For example, when Cy3 and Cy5 are used as labeling substances, Cy3 can be detected by scanning at a wavelength of 532 nm, and Cy5 can be detected by scanning at a wavelength of 635 nm. The strength of the label is used as an index of the expression level of the gene.
[0047]
When the mRNA amount is measured by a nucleic acid amplification method, particularly an RT-PCR method, for example, a competitive PCR method, a TaqMan method [for example, Linda G. Lee et al. (Linda G. Lee), Nucleic Acids Research, Vol. See pages 3761-1376 (1993) and the like].
[0048]
The amount of the polypeptide encoded by the gene can be measured using an antibody against the polypeptide or a fragment thereof. Specifically, for example, it can be carried out by a conventional ELISA method using a labeled antibody.
[0049]
In the evaluation method of the present invention, in step (2), the gene expression level is analyzed by comparing the gene expression level in a test sample with the gene expression level in a control sample, whereby the malignancy of cancer, particularly gastric cancer, is analyzed. The degree can be evaluated. Here, the control sample refers to a sample derived from a healthy person or a sample derived from a non-lesion site.
[0050]
Analysis of gene expression can be performed by pattern analysis. Such pattern analysis can be performed by clustering analysis using the expression intensity signal of each gene in the test sample or the relative expression ratio of each gene in the test sample relative to the control sample.
[0051]
That is, clustering analysis is performed by a computer using the obtained expression intensity of each gene. The data used for the analysis include the expression intensity signal of each gene in the test sample or the relative expression ratio of each gene in the test sample relative to the control sample.
[0052]
In step (2), a process of correcting the measured value of the expression level of each gene obtained in step (1) can be further performed.
[0053]
The relative expression ratio refers to a value represented by [gene expression amount in test sample / gene expression amount in control sample]. The clustering method is not particularly limited, and includes, for example, Hierarchical clustering, Neural network clustering, and the like.
[0054]
Hereinafter, Hierarchical clustering that is relatively frequently used in the biotechnology field will be described.
[0055]
Hierarchical clustering means that when data measured for each sample is arranged in multidimensional coordinates, the distance between the data is found in order from the closest sample, and finally similar samples are arranged side by side. The feature is that the results can be represented by a tree diagram. There are various methods for measuring the distance between samples, and the Euclidean distance, Mahalanobis' generalized distance, and the like are often used. In the process of creating a tree diagram, there are various ways of defining the mutual distance of a set of individual samples, and the method is not particularly limited, but mainly includes a shortest distance method, a longest distance method, a group average method, and a center of gravity method. And Ward's method.
[0056]
When using the expression intensity signal of each gene in a test sample, for example, cancer tissue, a value obtained by subtracting the background signal from the spot signal of each gene (this value is referred to as the expression intensity signal) is input to the computer. Since the distribution of the signal intensity at this time differs for each array, it is preferable to perform normalization. The normalization method includes Z-score, mean deviation, mean absolute deviation, and median absolute deviation, and is appropriately selected. Genes to be clustered are selected from the genes listed in Tables 1 and 2, and the higher the number, the more accurate the classification. Specifically, even with a combination of at least eight kinds of restricted genes, a limited malignancy factor can be determined.
[0057]
Next, by performing Hierarchical clustering analysis on the expression intensity signal after the normalization, clustering of a sample, for example, a gastric cancer sample and gene clustering are performed. As a result, classification between specimens on the dendrogram, that is, between specimens with similar gene expression patterns, showed the degree of malignancy such as cancer invasion, lymph node metastasis, peritoneal metastasis, and liver metastasis in pathological diagnosis. Metastatic specimens with high commonality and high malignancy, and early cancer specimens with low invasiveness were placed at the farthest distance. By using this method, it is possible to determine the degree of malignancy of an unknown sample only by collecting a small amount of cancer tissue by biopsy before surgery. That is, mRNA is purified from a cancer tissue, and hybridization is performed using, for example, a probe that has been subjected to one-color fluorescence labeling to determine the expression intensity signal of each gene. By performing a clustering analysis together with the expression intensity signal and the expression intensity signal of a gastric cancer sample that is a standard whose malignancy has already been determined, it is determined which malignancy level the unknown sample is closer to the standard sample. In conventional postoperative pathological diagnosis, if the cancer invasion spreads from the mucous membrane to the muscular layer, is the cancer with a relatively good prognosis or a malignant cancer that may recur and / or metastasize in the future? The prognosis could not be determined because it was located at the bifurcation point, but by performing the clustering analysis of the present invention, the specimen of unknown malignancy was a low-grade early cancer or a metastatic high-grade cancer specimen Which is closer to the distance, the prognosis can be predicted.
[0058]
On the other hand, the relative expression ratio of each gene in a test sample (eg, gastric cancer tissue) with respect to a control sample (eg, control normal tissue), that is, [gene expression amount in test sample / gene expression amount in control sample] (specifically, In the case of using [gene expression level in cancer tissue / gene expression level in control normal tissue]), clustering of gastric cancer specimens yields almost the same results as those using cancer tissue expression intensity signals. Can be When the relative expression ratio is used, there is an advantage that a gene expression profile excluding individual differences in gene expression can be seen. Gene clustering is classified for each gene whose expression changes due to canceration and malignancy are similar. By analyzing the correlation between the expression fluctuation and the malignancy factor in pathological diagnosis, genes involved in malignancy are identified. It is possible to specify. As a result, the association between the 105 genes shown in Tables 1 and 2 and the 51 genes shown in Tables 6 and 7, and the malignancy, that is, the correlation with the invasion level indicating the degree of cancer invasion was found. Was done.
[0059]
The result obtained by the above analysis is output to another software package such as a printer, a display, or a graphic software for a display. Such an output is particularly advantageous when the result obtained by the evaluation method of the present invention is used for diagnosis or the like.
[0060]
Furthermore, the inventors have found that a simpler diagnosis of malignancy is possible by applying the classification of genes correlated with malignancy found by the above method. In other words, it is possible to evaluate the malignancy of cancer, especially stomach cancer, by evaluating the level of gene expression fluctuation using the number of genes whose expression fluctuates and the ratio of expression fluctuation as an index without performing complex clustering analysis by computer. Was found. Such an embodiment is also included in the present invention.
[0061]
Here, the “variation level of gene expression” is expressed by (1) [the relative expression ratio of the gene whose expression level is increased in the test sample compared to the control sample or the sum of the absolute values of the logarithms of the relative expression ratio. (2) [The value indicated by the sum of the points of each gene in the test sample, with the gene whose expression level increased in the test sample as one point compared to the control sample] can do. Specifically, for example, when the relative expression ratio of the gene whose expression level is increased in the test sample as compared with the control sample is 2 or more, one point is obtained, and the sum of the points of each gene in the test sample is determined. From the indicated values, the level of fluctuation in the expression of the gene can be evaluated.
[0062]
As used herein, the term “variation in gene expression” refers to a variation of at least two times, that is, a relative expression ratio ([gene expression amount in test sample / gene expression amount in control sample]) of 2 or more or 1/2 or less. The case of Here, the case where the relative expression ratio is 2 or more is a gene whose expression is increased, and the case where the relative expression ratio is 1/2 or less is an index of a gene whose expression is decreased. In (2) above, the number of expression-variable genes in a group consisting of eight or more genes selected from a group of genes that greatly correlate with each factor of malignancy is determined by comparing the number of malignancy and prognosis to the cancer tissue that is already known. It can be evaluated by comparing with the number of genes whose expression is varied.
[0063]
In the value of (1), the value of [the amount of gene expression in the test sample / the amount of gene expression in the control sample] is indicated by a logarithm, and the absolute value is used to evaluate the rate of increase and decrease in expression equally. can do. The log value at that time is the natural logarithm log ₂ , Absolute log ₁₀ May be represented by any of
[0064]
When assessing malignancy using the number of genes whose expression fluctuates and the percentage of expression fluctuation as indices, it is highly reliable to judge based on the intensity signal (background value) of the negative control gene mounted on the same chip. It is possible to That is, when both the expression intensity signal in the test sample and the expression intensity signal in the control sample are smaller than twice the background value, the reliability of the fluctuation is low. The reliability is high, and if any of them is four times or more, the reliability is even higher. When the reliability is added to the malignancy determination, the reliability of the evaluation is increased. That is, when either the expression intensity signal of the cancer or the expression intensity signal of the control normal tissue is at least twice the intensity signal of the negative control gene, it can be evaluated by the sum of the number of genes whose expression fluctuates and the ratio of the expression fluctuation. desirable.
[0065]
In the present invention, the analysis of the gene expression level is performed by comparing the expression levels of at least eight genes whose expression levels change depending on the degree of malignancy of gastric cancer with the expression level of the housekeeping gene in the test sample. Can be.
[0066]
In the method for evaluating the malignancy of cancer, particularly gastric cancer, of the present invention, in step (2), the gene expression level is analyzed by comparing the gene expression level in a test sample with the gene expression level in a control sample, The malignancy of cancer, particularly gastric cancer, may be evaluated based on the obtained analysis result of the gene expression amount and the weighting coefficient for each gene. In other words, the evaluation is performed by summing the values indicated by the logarithms of the relative expression ratios of the genes in the test sample with respect to the control samples by weighting factors that take into account the contribution of each gene to malignancy. Can be. In the present invention, the degree to which the variation in the expression of each gene contributes to malignant transformation is not uniform. Therefore, by weighting the variation in the expression of the gene according to the degree, it is possible to perform an evaluation that reflects the malignant transformation more.
[0067]
In the above embodiment, the weighting coefficient of a gene having a large contribution to malignancy is set high, and the weighting coefficient of a gene having a small contribution is set small. When the gene whose expression is increased in the test sample is inversely correlated with malignancy as compared with the control sample, the weighting coefficient becomes a negative value.
[0068]
The stage of assigning the weighting coefficients is not particularly limited, but is usually assigned in 2 to 10 stages.
[0069]
For example, in Tables 6 and 7 described in Examples below, for 70 types of genes, weighting coefficients having seven levels of -3, -1, 1, 2, 3, 4, and 5 are exemplified. , May be changed in five or three stages.
[0070]
As a specific evaluation method using a weighting coefficient, for example, for a gene having a weighting coefficient of 5 whose expression level is twice as high as that of a control sample in a test sample, 0 is a natural logarithm of 2 which is an expression ratio. The value of 3.465 obtained by multiplying .693 by 5 of the weighting factor is the point of the gene. Similarly, the points of all genes are obtained, and the degree of malignancy is evaluated by the sum of the points.
[0071]
The above-mentioned weighting coefficient is set to a value that can examine the level of gene expression variation using clinicopathological data and a specimen with a known prognosis, and appropriately indicate the degree of malignancy.
[0072]
2. Array of the invention
The array of the present invention for evaluating the malignancy of cancer, particularly gastric cancer, which is used to evaluate malignant transformation of cancer, particularly of gastric cancer, has an expression level of at least 8 which varies depending on the malignancy of cancer, particularly gastric cancer. An array in which nucleic acids (eg, DNA, etc.) corresponding to species genes or fragments thereof are immobilized in defined regions on a support, respectively. In the array of the present invention, the above-mentioned 1. Since the nucleic acids or fragments thereof corresponding to at least eight genes whose expression levels change depending on the malignancy of cancer, particularly gastric cancer, are fixed, the malignancy of cancer, particularly gastric cancer, can be easily determined. It has an excellent property that it can be evaluated by simple operations.
[0073]
Here, "immobilized at a predetermined position" means that the position where the nucleic acid corresponding to each gene or the fragment thereof is immobilized is predetermined on the support. That is, when such an array is used, it is possible to know from the position of the detected signal which gene nucleic acid or its fragment is derived from.
[0074]
The support used in the array of the present invention is not particularly limited as long as it can be used for hybridization, and usually a slide glass, a silicon chip, a nitrocellulose or nylon membrane or the like is used. More preferably, any material may be used as long as it is nonporous and has a structure with a smooth surface, and is not particularly limited. For example, glass such as a slide glass can be suitably used. The surface of the support may be any as long as it can immobilize single-stranded DNA by a covalent bond or a non-covalent bond, and preferably has a hydrophilic or hydrophobic functional group on the surface of the support. Although there is no particular limitation, for example, those having a hydroxyl group, an amino group, a thiol group, an aldehyde group, a carboxyl group, an acyl group or the like can be suitably used. These functional groups may be present as surface characteristics of the support itself, or may be introduced by surface treatment. Examples of such surface-treated products include those obtained by treating glass with a commercially available silane coupling agent such as aminoalkylsilane, and those treated with a polycation such as polylysine or polyethyleneimine. Some of the slide glasses subjected to these treatments are commercially available.
[0075]
In the array of the present invention, the nucleic acid or a fragment thereof may be single-stranded or double-stranded. For example, a DNA array in which the nucleic acid or a fragment thereof is aligned and immobilized as two denatured strands on a support, or a DNA array in which at least a part of the immobilized DNA is single-stranded DNA may be used. . The array of the present invention may be a DNA array in which double-stranded DNA is denatured and aligned on the same support and spotted.
[0076]
Further, in the array of the present invention, the density of the nucleic acid of the gene to be immobilized or the fragment thereof on the support is not particularly limited. For example, a high-density array may be used, and 100 dots / cm. ² An array on which nucleic acids, particularly DNAs, are immobilized at the above density can be suitably used.
[0077]
The nucleic acid or its fragment immobilized on the support is not particularly limited, and may be any of a polynucleotide and an oligonucleotide. There is no particular limitation on the production method, and it may be chemically synthesized, isolated and purified from natural nucleic acids, enzymatically synthesized, or a combination thereof.
[0078]
The nucleic acid or its fragment immobilized on the support is not particularly limited, and is, for example, a double-stranded polynucleotide having a chain length of 50 bases or more or a derivative thereof, and is composed of PCR (polymerase chain reaction). Those prepared by enzymatic amplification by a method or the like are denatured at the time of immobilization of the DNA to the immobilization support, and single-stranded DNA or a derivative thereof can be suitably used. The derivative may be modified as long as it can be immobilized on the surface of the support, and is not particularly limited. Examples of the derivative include an amino group and a thiol group at the 5 ′ end of DNA. And DNA into which the functional group is introduced.
[0079]
For example, DNA amplified by PCR or the like using a genomic DNA library or a cDNA library as a template can be used. The array can be prepared by immobilizing a nucleic acid corresponding to the above gene or a fragment thereof on a support having an amino group introduced therein, for example, by a known method. In addition, by performing the above-mentioned immobilization operation using a DNA array producing apparatus, for example, a DNA chip producing apparatus manufactured by GMS, an array of the present invention in which nucleic acids corresponding to genes are aligned and immobilized is produced. Can be.
[0080]
When a nucleic acid fragment is immobilized on an array, the chain length of the fragment is not particularly limited, and is preferably, for example, about 100 bases to about 1 kilobase, and is shorter than the chain length. Alternatively, a long one may be used as long as it specifically hybridizes with a nucleic acid derived from a test sample in hybridization.
[0081]
The gene used in the array of the present invention may be any gene whose expression level changes depending on the malignancy of cancer, particularly gastric cancer, and is not particularly limited. And the like, for example, genes encoding oncogenes, tumor suppressor genes, growth factors, metastasis / invasion factors, angiogenesis factors, or proteins involved in energy metabolism. Furthermore, for example, expression is detected by differential display (DD) from total RNA extracted from cancer tissues and control normal tissues of gastric cancer patients of various malignancies whose functions are unknown but different in malignancy of gastric cancer. You may find genes that decrease or increase and immobilize them. Further, 1. All the genes whose expression is fluctuated in the course of malignant transformation of gastric cancer, which are found by the gene selection method described in the section, are all available, and the malignancy of gastric cancer, in particular, all nucleic acids corresponding to such genes are immobilized. An evaluation DNA array can be prepared.
[0082]
Specifically, 1. And at least 8 genes, preferably at least 20 genes selected from the group consisting of the genes listed in the specific examples of "genes whose expression level changes according to the degree of malignancy of gastric cancer".
[0083]
The arrays of the present invention can also be used to detect cancer cells, especially in gastric cancer.
[0084]
3. Kit for evaluating the malignancy of cancer of the present invention, particularly gastric cancer
The kit for evaluating the malignancy of the cancer of the present invention, particularly gastric cancer, includes: (1) mRNA or mRNA expressed from at least eight genes whose expression level changes depending on the malignancy of gastric cancer, depending on the detection target; A kit containing a primer and / or a probe for detecting the fragment, and (2) an antibody against a polypeptide or a fragment thereof encoded by at least eight genes whose expression level changes depending on the malignancy of gastric cancer. A kit containing the fragment is included.
[0085]
The kit of the present invention comprises (1) a primer and / or a probe for detecting mRNA or a fragment thereof expressed from at least eight genes whose expression level changes depending on the grade of gastric cancer, or (2) gastric cancer Since it contains an antibody or a fragment thereof to a polypeptide encoded by at least eight kinds of genes or fragments thereof whose expression level changes depending on the grade of malignancy, it is used in the method for evaluating the grade of malignancy of the cancer, particularly gastric cancer. Can be operated more easily and quickly. Further, the kit of the present invention can also be used for detecting cancer cells in cancer, particularly gastric cancer.
[0086]
The kit of (1) is a nucleic acid amplification method, specifically, an RT-PCR method, for mRNA or a fragment thereof expressed from at least eight genes whose expression level changes depending on the malignancy of cancer, particularly gastric cancer. Primers and / or probes for detection.
[0087]
The primers and / or probes hybridize under stringent conditions to at least eight genes whose expression levels change depending on the malignancy of cancer, particularly gastric cancer, or to a nucleic acid having a nucleotide sequence complementary to the gene. What is necessary is soybean.
[0088]
The “stringent conditions” are not particularly limited, but include, for example, a solution containing 6 × SSC, 0.5% SDS, 5 × Denhardt, 100 μg / ml herring sperm DNA, [the primer and / or the probe Tm-25 ° C.].
[0089]
The base sequence of the above primer may be any sequence that can specifically amplify the nucleic acid corresponding to the gene under the reaction conditions of a conventional nucleic acid amplification method, particularly RT-PCR.
[0090]
On the other hand, the nucleotide sequence of the probe may be any nucleic acid sequence capable of hybridizing to the nucleic acid corresponding to the gene under the stringent conditions.
[0091]
The Tm of the probe or primer is, for example, the following formula:
Tm = 81.5-16.6 (log ₁₀ [Na ⁺ ]) + 0.41 (% G + C)-(600 / N)
(Where N is the length of the probe or primer and% G + C is the content of guanine and cytosine residues in the probe or primer)
Required by
[0092]
When the chain length of the probe or the primer is shorter than 18 bases, Tm is, for example, the product of the content of A + T (adenine + thymine) residue and 2 ° C. and the content of G + C residue and 4 ° C. It can be estimated from the sum of the products [(A + T) × 2 + (G + C) × 4].
[0093]
The chain length of the probe is not particularly limited, but is preferably 15 bases or more, and more preferably 18 bases or more, from the viewpoint of preventing non-specific hybridization.
[0094]
The chain length of the primer is not particularly limited, but is, for example, 15 to 40 bases, and preferably 17 to 30 bases.
[0095]
As the “at least eight genes whose expression levels change depending on the degree of malignancy of gastric cancer”, preferably, the “1. And at least 8 genes, more preferably at least 20 genes selected from the group consisting of the genes listed in the specific examples of "genes whose expression level changes according to the degree of malignancy of gastric cancer" .
[0096]
In the kit of the above item (2), examples of the polypeptide include polypeptides encoded by at least eight types of genes whose expression levels change depending on the degree of malignancy of the gastric cancer. Specifically, 1. And at least eight, more preferably at least twenty, selected from the group consisting of the polypeptides encoded by the genes listed in the specific examples of “genes whose expression level changes according to the degree of malignancy of gastric cancer”. Species polypeptides.
[0097]
The antibody is not particularly limited as long as it has an ability to specifically bind to the polypeptide, and may be either a polyclonal antibody or a monoclonal antibody. Furthermore, an antibody or an antibody derivative modified by a known technique, for example, a humanized antibody, a Fab fragment, a single-chain antibody, or the like can also be used. The antibody can be obtained, for example, in 1992, published by John Wiley & Sons, Inc., edited by John E. Coligan, Current Protocols in Immunology (Current Protocols in). Immunology) can be easily prepared by immunizing a rabbit, rat, mouse or the like with all or a part of the polypeptide. The antibody thus obtained is purified and then treated with peptidase or the like to obtain an antibody fragment. Antibodies can also be produced by genetic engineering. Furthermore, the antibody of the present invention or a fragment thereof may be variously modified in order to facilitate detection by an enzyme immunoassay, a fluorescence immunoassay, a luminescence immunoassay, or the like.
[0098]
The above antibodies or fragments thereof include those which can specifically bind to a certain partial fragment of the polypeptide.
[0099]
The kit of the present invention may appropriately contain a detection reagent and the like.
[0100]
In addition, the present invention analyzes the results obtained by the evaluation method and the like by analysis means (for example, an image processing method using a computer) and records the results on a recording medium such as paper; A diagnostic method provided by recording or displaying on a medium or the like can also be provided. Such a diagnostic method is also included in the present invention.
[0101]
【Example】
Hereinafter, the present invention will be described in more detail with reference to examples, but the present invention is not limited to the scope of the examples.
[0102]
Example 1
(1) Preparation of DNA array
A DNA fragment found in the pBluescript phagemid clone of the cancer-related gene described in WO 98/37187, which was found by the differential display (DD) method, was used as a template, and was placed at both ends of the multicloning site of the phagemid. The target cDNA fragment was amplified and recovered by the PCR method using the set primers. Further, glyceraldehyde 3-phosphate dehydrogenase (GAPD) and a cyclophilin gene were prepared as a housekeeping gene, and a plasmid pUC18 was prepared as a negative control, respectively.
[0103]
These DNAs were collected by ethanol precipitation, dissolved in 100 mM carbonate buffer (pH 9.5) to 1 μM, and amino group-introduced slide glass (Sigma) using a DNA chip manufacturing device (GMS). And fixed by UV irradiation. The slide was washed with 0.2% SDS and then with distilled water and dried to obtain a DNA array.
[0104]
(2) Preparation of fluorescently labeled cDNA
From ten gastric cancer patients having different degrees of cancer invasion, metastasis, or progression shown in Table 3, gastric cancer tissues and control normal stomach tissues were separated from tissues removed at the time of cancer removal surgery. Subsequently, total RNA was individually extracted from each tissue by the AGPC (Acid Guanidium Phenol-Chloroform) method. MRNA was purified from each of these total RNAs using Oligotex-MAG mRNA Purification kit (Takara Shuzo).
[0105]
[Table 3]

[0106]
Each item in Table 3 shows the following findings:
Degree of progress:
The stage I is an early stage cancer in which the degree of invasion of the cancer is shallow and no metastasis is observed.
Progression II refers to a state where the cancer has invaded and spread to nearby lymph nodes.
Progression III refers to a condition that has spread to distant lymph nodes.
Progression IV refers to a state in which the disease has progressed further, spread to distant lymph nodes, and has metastasized to the peritoneum and liver.
[0107]
Differentiation degree:
As general types, papillary adenocarcinoma (pap), highly differentiated adenocarcinoma (tub 1), moderately differentiated adenocarcinoma (tub 2), poorly differentiated adenocarcinoma (por), signet-ring cell carcinoma (sig), and mucinous carcinoma (muc) are categorized
[0108]
Depth:
It is expressed by the deepest layer of the stomach wall that has reached the cancer invasion. Is expressed.
[0109]
Lymph node metastasis (represents the degree of metastasis of the lymph node group found histologically):
No metastasis is observed (N ₀ )
Metastasis is observed only in group 1 lymph nodes (N ₁ )
Metastases are observed in the lymph nodes of the second group, but not in the third and fourth groups (N ₂ )
Metastases are found in the lymph nodes of Group 3 but not in Group 4 (N ₃ )
Metastasis to lymph nodes of group 4 (N ₄ )
[0110]
Lymphatic invasion:
No invasion found (0)
Mild invasion (1)
Moderate invasion (2)
Highly invasive (3)
[0111]
Venous invasion:
No invasion found (0)
Mild invasion (1)
Moderate invasion (2)
Highly invasive (3)
[0112]
Peritoneal metastases:
No sowing at all (0)
Seeding in the adjacent peritoneum (1)
Few metastases to distant peritoneum (2)
Numerous metastases to the distant peritoneum (3)
[0113]
Liver metastases:
No liver metastasis (0)
Liver metastasis is observed only in one lobe (1)
Few sporadic metastases on both leaves (2)
Many scattered metastases on both leaves (3)
[0114]
Invasive growth mode:
The cancerous layer shows swelling growth and is distinct from its surroundings (α)
Foci showing invasive growth and indistinct border with surrounding tissue (γ)
Something in between (β)
[0115]
Using the above mRNA as a template, a reverse transcription reaction was performed using Cy3-dUTP (manufactured by Amersham-Pharmacia) for the control normal gastric cancer tissue group and Cy5-dUTP (manufactured by Amersham-Pharmacia) for the gastric cancer tissue group. Then, a fluorescence-labeled cDNA was prepared.
[0116]
The composition of the reaction solution is shown below.
Reaction solution A: about 1 μg of mRNA, 300 pmol of oligo dT primer (manufactured by Takara Shuzo) and water treated with diethylpyrocarbonate (DEPC, manufactured by Nacalai Tesque) to a final concentration of 11.9 μl;
Reaction solution B: 4 μl of 5 × AMV RTase buffer (manufactured by Life Science), 0.1 mM dATP, dCTP, dGTP and 0.065 mM dTTP, 60 U RNase inhibitor (manufactured by Takara Shuzo), 0.035 mM Cy3 Alternatively, Cy5-labeled dUTP (manufactured by Amersham Pharmacia) was mixed to obtain a solution having a final volume of 6.5 μl.
[0117]
After keeping the reaction solution A at 70 ° C. for 5 minutes, it was cooled on an ice bath. Thereafter, the reaction solution A (6.5 μl) and 30 U of AMV RTase (manufactured by Life Science) were added to the reaction solution A to obtain an RT reaction solution. After keeping the RT reaction solution at 55 ° C. for 30 minutes, the temperature was set to 42 ° C. After adding 30 U of AMV RTase to the RT reaction solution to adjust the volume to 20 μl, the obtained mixture was again kept at 42 ° C. for 60 minutes. To the obtained reaction solution, 2.5 μl of a 500 mM EDTA solution and 5 μl of 1 M sodium hydroxide were added, and the mixture was kept at 37 ° C. for 10 minutes to decompose the template RNA. After cooling the reaction solution to room temperature, 12.5 μl of 1 M Tris-hydrochloric acid (pH 7.5) was added. This solution was subjected to gel filtration using Centri-Sep spin column (manufactured by Applied Biosystems). As a result, a Cy3-labeled cDNA solution (about 35 μl) and a Cy5-labeled cDNA (about 35 μl) were obtained.
[0118]
The thus obtained Cy3-labeled cDNA and Cy5-labeled cDNA are mixed so as to form a pair for each same patient, concentrated by ethanol precipitation, and subjected to hybridization buffer (6 × SSC / 0.2% SDS / 5 × Denhardt's solution / 0. It was dissolved in 8 μl of 1 mg / ml salmon sperm DNA) to prepare a fluorescence-labeled cDNA.
[0119]
(3) Hybridization between labeled cDNA and DNA array
Pre-hybridization buffer (6 × SSC / 0.2% SDS / 5 × Denhardt's solution / 1 mg / ml salmon sperm) was added to the DNA array prepared in the above (1) and a commercially available DNA array (Intelligene Cancer Chip, manufactured by Takara Shuzo). DNA) was added dropwise, the area was covered with a cover glass, and the periphery was sealed with a film. After keeping this at room temperature for 2 hours, the cover glass was removed, and the plate was washed with 2 × SSC, then with 0.2 × SSC, and air-dried.
[0120]
After heat denaturation (94 ° C., 3 minutes) of the fluorescence-labeled cDNA prepared in (2), 8 μl of the cDNA was dropped on the pre-hybridized DNA array, covered with a cover glass, and sealed around with a film. After keeping the obtained array at 65 ° C. for 16 hours, the cover glass was removed, and the plate was washed twice in 2 × SSC / 0.2% SDS at 55 ° C. for 30 minutes, and then washed at 65 ° C. for 5 minutes in 1 ×. It was further washed in 0.05 × SSC at room temperature for 10 minutes and air-dried. The obtained array was subjected to a microarray scanner (manufactured by GMS) to measure the fluorescent signal of each spot. The measured signals were analyzed with expression data analysis software Imagegene (manufactured by Biodiscovery), and the expression levels of each gene in cancer tissues and control normal tissues were examined.
[0121]
Relative expression ratio of each gene immobilized on the used DNA array in cancer tissue and control normal tissue ([expression intensity signal of cancer tissue / expression intensity signal of control normal tissue]), that is, [gene in test sample Expression level / gene expression level in control sample] was complicatedly different among the 10 gastric cancer patients. However, some genes showed expression patterns depending on the degree of cancer invasion, the presence or absence of metastasis, or the degree of progression. Through the following steps, genes related to the malignancy of gastric cancer were found from these genes.
[0122]
First, Hierarchical clustering using the clustering software GeneSight (manufactured by BioDiscovery) was performed on the [natural logarithmic conversion value of the relative expression ratio] in the cancer tissues of all the genes used on the DNA array and the control normal tissues. The Euclidean distance was used to measure the distance between each sample, and the centroid method was used to define the distance. In the tree diagram obtained by this analysis, the vertical axis indicates the classification of the specimen, which is large, [Initial stomach cancer specimen], [The cancer has invaded into the muscularis propria or the subsera, but metastasis does not occur. Stomach cancer specimens) and [metastatic specimens]. In addition, the horizontal axis of the dendrogram indicates the classification of the genes, and these are [genes having a low expression intensity ratio in all gastric cancer patient cancer tissues] and [high expression ratio in all gastric cancer patient cancer tissues] It can be seen that the gene groups are classified into three groups: gene groups and gene groups having different expression patterns for each gastric cancer patient.
[0123]
Genes related to the degree of malignancy of gastric cancer were considered to belong to [gene groups having different expression patterns for each gastric cancer patient] among the above three classifications, so the genes of this group were first selected as candidates. From these candidate genes, genes in which the expression intensity signal of the cancer and the expression intensity signal of the control normal tissue were smaller than twice the background value were removed because their expression was low and the evaluation reliability of fluctuation was low. Since candidate gene groups having different expression patterns for each gastric cancer patient are sub-clustered with about 3 to 10 genes as one group, next, the remaining genes are removed from the selected genes for each sub-cluster gene group. Clustering was performed similarly. As a result, a sub-clustering group in which the correlation between the classification between specimens and the malignancy was lost was found. It is thought that the expression of such a gene is closely related to the grade of gastric cancer.
[0124]
The 105 gene names selected through the above steps and their accession numbers are the same as those shown in Tables 1 and 2 above.
[0125]
FIGS. 1 and 2 show dendrograms obtained by Hierarchical clustering using these 105 genes. FIG. 1 shows the classification of gastric cancer patients. FIG. 2 shows a tree diagram of the expression pattern of each gene in each gastric cancer patient. As shown in FIG. 1, it can be seen that there is a large separation between early gastric cancer patients (MK38 and MK168), which have low cancer penetration and no lymphatic invasion, and moderate to advanced advanced gastric cancer patients with invasiveness. . In addition, it can be seen that the latter includes a sample group (MK153, MK44 and MK143) in which cancer has deeply infiltrated and has metastasized to lymph nodes, and subclusters of MK144, MK80 and MK96. Of these, MK96 had relatively low cancer penetration and did not show metastasis, but when the surgical section was observed under a microscope, the cancer was mildly invading the lymphatic vessels. Currently, 4 years and 10 months after the operation, he is still alive. Although no metastasis was observed during the operation of MK80, the cancer had deeply infiltrated, and the cancer was highly invaded into the lymphatic vessels by microscopic observation of the excised surgical section. The cancer has since recurred and died. MK144 was a highly aggressive cancer patient that had metastasized to both lymph nodes and peritoneum.
[0126]
Hierarchical clustering was also performed using the expression intensity of the gene on the DNA array in the cancer tissue as an index. The Z-score method was used to normalize the expression intensity in cancer tissues, the city block distance was used to measure the distance between samples, and the centroid method was used to define the distance. The resulting dendrograms are shown in FIGS. Although the classification of the patient specimens on the vertical axis (FIG. 3) is slightly different from that using the expression ratio, the 10 gastric cancer patients are also the early gastric cancer patients with low cancer penetration and no lymphatic invasion. It was largely divided into two groups of moderate to high advanced gastric cancer specimens with lymphatic invasion and / or metastasis. Furthermore, the latter contained a subcluster of MK80 and MK96 in which metastasis was not observed, and the metastasis sample was complicatedly involved in this subcluster (FIG. 4).
[0127]
Hierarchical clustering when using the expression intensity of the cancer tissue and when using the relative expression ratio of the control normal tissue to the cancer tissue are similar and correlate with the malignancy of gastric cancer. It was determined that the expression change in the cancer tissue was not a change due to individual differences but was related to malignant transformation of gastric cancer.
[0128]
Example 2
(1) Preparation of DNA array
For the 105 types of genes selected in Example 1, cDNA fragments having a length of about 100 nucleotides to about 1 kilonucleotide) were prepared by the method described in Example 1, and a DNA array on which the DNA fragments were spotted was prepared. .
[0129]
(2) Hybridization between labeled cDNA and DNA array
Using the fluorescence-labeled cDNA prepared in Example 2, hybridization with the DNA array prepared in the above (1) was performed. Hybridization conditions were the same as in Example 1.
[0130]
Based on the results of the hybridization, 20 kinds of genes shown in Table 4 were selected from the fluorescent signals of the obtained 105 genes, and Hierarchical clustering was performed on the 20 kinds of genes in the same manner as in Example 1.
[0131]
[Table 4]

[0132]
In addition, using the [natural logarithmic conversion value of the relative expression ratio] of each gene with respect to the control normal tissue of the cancer tissue, without normalization, the distance between samples was measured by the Euclidean distance, and the definition of the distance was analyzed by the centroid method. The resulting dendrograms are shown in FIGS. As a result, it can be seen that the 10 gastric cancer patients are divided into two groups, a non-metastatic group and a metastatic group (FIG. 5). In addition, the non-metastatic group was found to be sub-clustered into early gastric cancer patients (MK38 and MK168) and MK80 and MK96 in which lymphatic invasion was observed. It can be seen that all the specimens constituting the sub-cluster of the metastasis group were specimens of highly advanced cancer in which metastasis was observed in a wide range of lymph nodes. Thus, it is shown that the combination of the 20 genes shown in Table 4 enables clustering closely related to malignancy. This experiment reconfirmed the involvement of these 20 genes in the malignancy of gastric cancer.
[0133]
Example 3
(1) Hybridization between labeled cDNA and DNA array
The malignancy level at which the prognosis is difficult to predict from the clinicopathological data at the time of surgery is gastric cancer of which the cancer reaches the muscular layer of the stomach, but whose cancer reaches the muscular layer of the stomach. MRNA was purified from gastric cancer tissue of stomach cancer patient MK115 classified as clinicopathologically at this level and control normal stomach tissue by the method described in Example 1, and fluorescent-labeled cDNA was prepared. Using this fluorescently labeled cDNA and the fluorescently labeled cDNA prepared in Example 1, hybridization was performed with the DNA microarray prepared in Example 2 (1). The hybridization conditions were the same as in Example 1.
[0134]
The clinicopathological data of the gastric cancer patient MK115 is as follows:
Progression degree I, differentiation degree mod, invasion degree mp, lymph node metastasis 0, lymph node invasion 0, venous invasion 0, peritoneal metastasis 0, liver metastasis 0, invasive growth mode α.
[0135]
As a result of the hybridization, eight kinds of genes shown in Table 5 were selected from among the obtained fluorescence signals of 105 genes, and the relative expression ratio of cancer tissues to control normal tissues was determined. The malignancy was determined based on the total expression of genes whose relative expression ratio was 2 or more, or the number of genes whose relative expression ratio was 2 or more, that is, genes whose expression in cancer tissues was more than twice as high as those in control normal tissues. .
[0136]
The above samples were compared with the 10 cancer samples used in Examples 1 and 2 as standard cancer samples. Table 5 shows the results.
[0137]
[Table 5]

[0138]
In the gastric cancer patient MK115, the sum of the relative expression ratios of the eight genes selected and the number of genes with relative expression ratios of 2 or more are both large, similar to those of advanced cancer, despite the progression Ib showing no lymphatic invasion or vascular invasion. Level. Similarly, the sum of the relative expression ratios of genes whose relative expression ratio was 2 or more was also at the same level as in advanced cancer. Examining the prognosis of this patient, the cancer recurred and died three years and nine months later. Normally, patients with grade I who do not have lymphatic invasion or vascular invasion rarely recur at such an early stage, and recurrence of cancer three years later can be already predicted by changes in gene expression at the time of surgery. Is significant.
[0139]
Although the eight genes selected here have many genes that are suggested to be correlated with metastasis or malignancy, the expression of the genes differs from sample to sample even from the viewpoint of metastasis and non-metastasis. Absent. For example, matrilysin (MMP-7) has been reported to be highly associated with the malignancy of gastric cancer [Senota A. et al. Et al., Clinical & Experimental Metastasis, Vol. 16, pp. 313-321 (1998)], but overexpressed in 4 of the 11 patients measured this time. It's just Moreover, even in metastatic specimens having a high degree of malignancy, only half of them have increased expression. As is clear from these results, highly accurate determination of malignancy cannot be performed using only a single gene, and prognosis can be determined by a plurality of combinations of genes associated with malignancy.
[0140]
As described above, it has been revealed that the present invention is effective for diagnosis of cancer malignancy and prognosis, which could not be predicted by conventional pathological diagnosis.
[0141]
Example 4 Searching for Genes That Are Indicators of Malignancy
Using eight of the samples (stomach cancer tissue and control normal stomach tissue) derived from gastric cancer patients described in Example 1- (2) and samples newly obtained from 14 patients, Similarly, mRNA was extracted and purified. Further, using the obtained mRNA as a template, a fluorescence-labeled cDNA was prepared. For these labeled cDNAs, hybridization was carried out using the DNA array prepared in Example 2- (1) and a commercially available DNA array (Intelligene Cancer Chip, manufactured by Takara Shuzo Co., Ltd.). Analysis was performed in the same manner as in 3).
[0142]
As a result of the above hybridization, 70 genes belonging to the subclustering group involved in the malignancy of gastric cancer were found. A closer look at the genes belonging to this sub-clustering group reveals that genes whose expression increases significantly with increasing malignancy; although expression tends to increase with increasing malignancy, there is variation among samples at the same progression level Genes: Genes containing genes whose gene expression decreased as malignancy increased were found. Therefore, by weighting these genes using a weighting coefficient, an attempt was made to enable point diagnosis reflecting the contribution rate to malignancy.
[0143]
For genes whose expression level increases in correlation with malignancy, a weighting coefficient of 5 to 1 point is assigned according to the degree of the increase, and conversely, for genes whose gene expression level decreases as the malignancy increases Assigned a score of -1 or -3. The logarithm of the relative expression ratio of the gene in the test sample with respect to the control sample was multiplied by the above-mentioned weighting coefficient to obtain the sum of the points of 70 genes.
[0144]
The sum of points increases in proportion to the grade of malignancy, such as the degree of cancer penetration or progression, and the number of relapsed and / or deceased specimens within 5 years reaches 100 or more The re-assignment of the weighting coefficient of each gene was repeated.
[0145]
Tables 6 and 7 show the names of the 70 genes and the weighting factors finally obtained as a result of the above-mentioned work.
[0146]
[Table 6]

[0147]
[Table 7]

[0148]
Table 8 shows the clinicopathological data of 22 kinds of samples used for the measurement of the gene expression amount and the change of the gene expression amount of each sample in consideration of the above-mentioned weighting factors. Each item in Table 8 is the same as in Table 3 above.
[0149]
[Table 8]

[0150]
As shown in Table 8, there is a clear correlation between the point calculated by the above method and the cancer malignancy determined from the pathological findings of the sample, and the point indicates the cancer malignancy. It can be seen that the degree can be evaluated.
[0151]
Further, for the sample derived from the gastric cancer patient MK115 described in Example 3, changes in the expression levels of the above 70 genes were examined, and the points were calculated. As a result, 138.6 points were obtained. In other words, it has been clarified that the above method can accurately determine the malignancy of gastric cancer that is clinically diagnosed with low malignancy.
[0152]
【The invention's effect】
The cancer of the present invention, particularly the method for evaluating the degree of malignancy of gastric cancer, the array for evaluating the degree of malignancy of cancer, particularly gastric cancer, and the kit for evaluating the degree of malignancy of cancer, particularly gastric cancer, malignant transformation of cancer, particularly gastric cancer Changes in the expression of many genes involved in can be measured quickly and systematically with high sensitivity, and by analyzing the pattern of these gene expression fluctuations, it is possible to determine the malignancy of cancer, especially gastric cancer, or to detect cancer, especially gastric cancer. It has an excellent effect when it can be done.
[Brief description of the drawings]
FIG. 1 shows a tree diagram of a classification of gastric cancer patients obtained by Hierarchical clustering using [natural logarithmic conversion values of relative expression ratios] of 105 genes.
FIG. 2 shows a tree diagram of expression patterns of each gene in each gastric cancer patient obtained by Hierarchical clustering using [natural logarithmic conversion value of relative expression ratio] of 105 genes.
FIG. 3 shows a tree diagram of classification of gastric cancer patients obtained by Hierarchical clustering using the expression intensity of 105 genes.
FIG. 4 shows a tree diagram of the expression pattern of each gene in each gastric cancer patient obtained by Hierarchical clustering using the expression intensity of 105 genes.
FIG. 5 shows a tree diagram of classification of gastric cancer patients obtained by Hierarchical clustering using [natural logarithmic conversion value of relative expression ratio] of 20 genes.
FIG. 6 shows a tree diagram of expression patterns of each gene in each gastric cancer patient obtained by Hierarchical clustering using [natural logarithmic conversion value of relative expression ratio] of 20 genes.

Claims (30)

The following steps:
(1) measuring the expression levels of at least eight genes whose expression levels change in accordance with the degree of malignancy of gastric cancer in the test sample; and (2) measuring the expression levels of each gene obtained in (1). A step of evaluating the degree of malignancy of gastric cancer based on the method. Apoptosis inhibitor survivin (apoptosis inhibitor survivin (Genebank accession number: U75285)) gene,
Type I collagen α-2 [collagen type I α-2 (Genebank accession number: J03464)] gene,
Type III collagen pro-α-1 chain [collagen type III pro-alpha-1 (Genebank accession number: X14420)] gene,
A fibronectin precursor (FN) [fibronectin precursor (FN) (Genebank accession number: X02761)] gene,
MMP-1; gastrointestinal collagenase [matrix metalloproteinase 1; intestinal collagenase (Genebank accession number: M13509)] gene,
MMP-7: uterine matrilysin [matrix metalloproteinase 7; matrilysin, uterine (Genebank accession number: Z11887)] gene,
U-plasminogen activator [urokinase-type plasminogen activator precursor (EC 3.4.2.7.73); U-plasminogen activator (UPA);
Urokinase-type plasminogen activator receptor; GPI-anchored from precursor (U-PAR); monocyte activation antigen gene;
Alpha-2-macroglobulin [alpha-2-macroglobulin (alpha-2-M) (Genebank accession number: M11313)] gene,
β-type platelet-derived growth factor receptor precursor (beta platelet-derived grown factor receptor precursor (PDGFR-beta); CD140B antigen (Genebank registration number: J03278)) gene,
BIGH3 (Genebank accession number: M77349) gene,
XVIII type collagen α-1 chain [collagen type XVIII alpha-1 (Genebank accession number: AF0188081)] gene,
A growth hormone-dependent insulin-like insulin-like growth factor-binding protein (Genebank registration number: M35878)] gene,
Hepatoma-derived growth factor (HDGF) (Genebank accession number: D16431) gene,
An insulin-like growth factor binding protein 2 (IGFBp2) (Genebank accession number: X16302)] gene,
Type IV collagenase [type IV collagenase (Genebank accession number: M55593)] gene,
spac precursor (osteonectin) [sparc precursor (secrete protein acidic and rich in cysteine; osteonectin) (ON); basement membrane protein BM-40;
A thrombospondin 2 precursor [thrombospondin 2 precursor (Genebank accession number: L12350)] gene,
Chondroitin sulfate proteoglycan core protein 2 (verifiedcan sulfate protein proteoglycan core protein 2);
Embopplakin (Genebank accession number: U53786) gene,
Integrin β-4 chain [integrin, beta 4] (Genebank accession number: X53587) gene,
Vimentin (Genebank accession number: Z19554) gene,
c-fms proto-oncogene [genebank registration number: X03663],
Proliferating cell nuclear antigen (Genebank accession number: M15796) gene,
Leukocyte antigen-related protein (Genebank accession number: Y00815) gene,
Interleukin 8 [interleukin 8] (Genebank accession number: M26383) gene,
GDP dissociation inhibitor (GDI) [GDP dissociation inhibitor (GDI)] (Genebank accession number: L20688) gene,
A brain-specific tubulin α chain [Tubulin, alpha, brain-specific] (Genebank accession number: X01703) gene,
Cell cycle control gene cdc2 (cell division cycle 2, G1 to S phase & G2 to M phase) [cell division cycle 2, G1 to S and G2 to M] (Genebank registration number: X05360),
Tissue-type plasminogen activator [plasminogen activator, tissue] gene bank accession number: M15518] gene,
Stromal cell-derived factor 1 (Gene bank accession number: U16752) gene,
Matrix metalloproteinase 10 (stromelysin 2) (Genebank accession number: X07820) gene,
Matrix metalloproteinase 15 (membrane-inserted) (Genebank accession number: Z48482) gene,
Antioxidant protein [antioxidant protein 1] (Genebank accession number: D49396) gene,
Laminin α-4 chain [laminin, alpha 4] (Genebank accession number: S78569) gene,
The ubiquitin-conjugating enzyme E2M (homologous to yeast UBC12), which is homologous to yeast UBC12; Genebank accession number: AF057599;
Retinoblastoma-binding protein 6 (Genebank accession number: X85133) gene,
K-sum [K-sam] (Genebank accession number: M87770) gene,
Cathepsin B (genebank accession number: L22569) gene,
Signal transduction & transcriptional activator 1; 91 kD [signal transducer and activator of transscription 1, 91 kD] (Genebank accession number: M97935) gene,
Carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reacting) (carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reactive antigen)
antigen)] (Genebank accession number: M18728) gene,
Heparan sulfate proteoglycan (gene bank accession number: M85289) gene,
v-Ki-ras2 Kirsten rat sarcoma virus oncogene homolog [v-Ki-ras2 Kirsten rat sarcoma 2 viral oncogene homolog] (Genebank accession number: M54968) gene,
Ras homolog gene family, factor D [ras homolog gene family, member D] (Genebank accession number: D85815) gene,
Hexabrachion (tenascin C, cytotactin) (Genebank accession number: X78565) gene,
Cyclin D2 [cyclin D2] (Genebank accession number: D13639) gene,
Metalloproteinase 3 tissue inhibitory factor [tissue inhibitor of metalloproteinase 3 (Sorsby fundus dystrophy, pseudoinflammation)] (Genebank registration number: AL023282) gene,
Laminin receptor 1, ribosomal protein SA (laminin receptor 1 (67 kD, ribosomal protein SA)) (Genebank accession number: U43901) gene,
Lactate dehydrogenase A (Genebank accession number: X02152) gene,
Keratin 18 (Genebank accession number: M26326) gene,
Ia-related invariant gamma chain [Ia-associated invariant gamma-chain] (Genebank accession number: AH001484) gene,
CD72 antigen [CD72 antigen] (Genebank accession number: M54992) gene,
Connective tissue growth factor (Genebank accession number: X78947) gene,
Matrix metalloproteinase 3 (stromelysin 1, progelatinase) (Genebank accession number: X05232) gene,
Cyclin-dependent kinase 4 (Genebank accession number: U37022) gene,
A human SB class II histocompatibility antigen [Human mRNA for SB class II histocompatibility antigen alpha-chain] (Genebank accession number: X03100) gene;
Transforming growth factor β-3 chain [transforming growth factor, beta 3] (Genebank accession number: X14885) gene,
Thyroid hormone receptor interacting factor 6 [thyroid hormone receptor interacting]
protein 6] (Genebank accession number: AJ001902) gene,
Platelet / endothelial cell adhesion molecule (CD31 antigen) (Genebank accession number: L34657) gene,
Profilin 1 (Genebank accession number: J03191) gene,
GTP-binding factor [GTP-binding protein] (Genebank accession number: AF120334) gene,
Homo sapiens clone 24703 β-tubulin mRNA [Homo sapiens clone 24703 beta-tubulin mRNA, complete cds] (Genebank accession number: AF070600) gene,
A Rho-associated coiled-coil containing protein kinase 1 (Genebank accession number: U43195) gene,
Type IV collagen, α-1 chain [collagen, type IV, alpha 1] (Genebank accession number: M26576) gene,
Topoisomerase (DNA) I (gene bank accession number: M60706) gene,
Kunitz type 1 serine proteinase inhibitor, [serine protease inhibitor, Kunitz type 1] (Genebank accession number: AB000095) gene,
Rho GTPase activating protein 1 [Rho GTPase activating protein 1] (Genebank accession number: U02570) gene,
Nidogen (gene bank accession number: M30269) gene,
Cyclin-dependent kinase 5, regulatory factor 1 [cyclin-dependent kinase 5, regulatory subunit 1 (p35)] (Genebank accession number: X80343) gene,
CD9 antigen [CD9 antigen] (Genebank accession number: M38690) gene,
Aortic-type smooth muscle α-actin gene, exon 9 (aortic-type smooth muscle alpha-actin gene, exon 9 (Genebank registration number: M33216)),
Bone morphogenetic protein 1 (BMp1) (Genebank accession number: U50330) gene,
Type VI collagen α-3 chain [collagen type VI alpha-3 (Genebank accession number: X52022)] gene,
A cytokine humig (interkine-gamma-induced monokine (MIG) (Genebank accession number: X72755)) gene,
Integrin alpha-3 chain [integrin alpha-3 chain (Genebank accession number: M59911)] gene,
An interferon-inducible gene family (Genebank accession number: X57351) gene,
A metastasis-associated MTA1 [metastasis-associated MTA1 (Genebank accession number: U35113)] gene,
Notch2 Notch homolog 3 (Genebank accession number: U97669)] gene, and proto-oncogene rhoA multidrug resistance protein (proto-oncogene rhoA multidrug resistance protein (registered trademark; GTP); No .: L25080)] The evaluation method according to claim 1, wherein the expression levels of at least eight genes selected from the group consisting of genes are measured. The evaluation method according to claim 2, wherein the expression levels of at least 20 types of genes are measured. The evaluation method according to any one of claims 1 to 3, wherein the expression level of the gene is measured from the amount of mRNA transcribed from the gene or the amount of polypeptide encoded by the gene. The evaluation method according to claim 4, wherein the mRNA amount is measured by a hybridization method or a nucleic acid amplification method. 6. The evaluation method according to claim 5, wherein in the hybridization method, an array in which nucleic acids or fragments thereof corresponding to the gene whose expression level is to be measured are immobilized in respective predetermined regions on a support. The evaluation method according to claim 5, wherein the nucleic acid amplification method is an RT-PCR method. The evaluation method according to claim 4, wherein the amount of the polypeptide encoded by the gene is measured using an antibody against the polypeptide or a fragment thereof. 9. The method according to claim 1, wherein in step (2), the gene expression level is analyzed by comparing the gene expression level in the test sample with the gene expression level in the control sample, thereby evaluating the malignancy of gastric cancer. Evaluation method described in Crab. In step (2), the gene expression level is analyzed by comparing the gene expression level in the test sample with the gene expression level in the control sample, and the obtained gene expression level analysis result and a weighting coefficient for each gene are analyzed. The evaluation method according to claim 9, wherein the degree of malignancy of gastric cancer is evaluated by the method. The evaluation method according to claim 9 or 10, wherein the control sample is a sample derived from a healthy person or a sample derived from a non-lesion site. The evaluation method according to any one of claims 9 to 11, wherein the analysis of the expression level of the gene is performed by pattern analysis. The evaluation method according to claim 12, wherein the pattern analysis of the gene expression is performed by clustering analysis using an expression intensity signal of each gene in the test sample or a relative expression ratio of each gene in the test sample with respect to the control sample. 14. The evaluation method according to claim 13, wherein the relative expression ratio is a value represented by [gene expression amount in test sample / gene expression amount in control sample]. 15. The evaluation according to claim 13 or 14, wherein the expression fluctuation level of the gene is evaluated by a value indicated by the absolute value of the logarithm of the relative expression ratio of the gene whose expression level has increased in the test sample as compared to the control sample. Method. When the relative expression ratio of the gene whose expression level was increased in the test sample as compared to the control sample was 2 or more, 1 point was determined. The gene indicated by the sum of the points of each gene in the test sample The evaluation method according to claim 13 or 14, wherein the expression fluctuation level is evaluated. The analysis of the gene expression level is performed by comparing the expression levels of at least eight types of genes whose expression levels change according to the degree of malignancy of gastric cancer with the expression level of a housekeeping gene in a test sample. 7. The evaluation method according to any one of items 1 to 7. 18. The evaluation method according to claim 1, wherein in step (2), the measured value of the expression level of each gene obtained in step (1) is further corrected. A nucleic acid or a fragment thereof corresponding to at least eight genes whose expression level changes according to the grade of gastric cancer is immobilized in a predetermined region on a support, for evaluating the grade of gastric cancer. array. 20. The array of claim 19, wherein the support is a glass slide. Apoptosis inhibitor survivin (apoptosis inhibitor survivin (Genebank accession number: U75285)) gene,
Type I collagen α-2 [collagen type I α-2 (Genebank accession number: J03464)] gene,
Type III collagen pro-α-1 chain [collagen type III pro-alpha-1 (Genebank accession number: X14420)] gene,
A fibronectin precursor (FN) [fibronectin precursor (FN) (Genebank accession number: X02761)] gene,
MMP-1; gastrointestinal collagenase [matrix metalloproteinase 1; intestinal collagenase (Genebank accession number: M13509)] gene,
MMP-7: uterine matrilysin [matrix metalloproteinase 7; matrilysin, uterine (Genebank accession number: Z11887)] gene,
U-plasminogen activator (EC 3.4.21.73); U-plasminogen activator (UPA); plasmin activator;
Urokinase-type plasminogen activator receptor; GPI-anchored from precursor (U-PAR); monocyte activation antigen gene;
Alpha-2-macroglobulin [alpha-2-macroglobulin (alpha-2-M) (Genebank accession number: M11313)] gene,
β-type platelet-derived growth factor receptor precursor (beta platelet-derived grown factor receptor precursor (PDGFR-beta); CD140B antigen (Genebank registration number: J03278)) gene,
BIGH3 (Genebank accession number: M77349) gene,
XVIII type collagen α-1 chain [collagen type XVIII alpha-1 (Genebank accession number: AF0188081)] gene,
A growth hormone-dependent insulin-like insulin-like growth factor-binding protein (Genebank registration number: M35878)] gene,
Hepatoma-derived growth factor (HDGF) (Genebank accession number: D16431) gene,
An insulin-like growth factor binding protein 2 (IGFBp2) (Genebank accession number: X16302)] gene,
Type IV collagenase [type IV collagenase (Genebank accession number: M55593)] gene,
spac precursor (osteonectin) [sparc precursor (secrete protein acidic and rich in cysteine; osteonectin) (ON); basement membrane protein BM-40;
A thrombospondin 2 precursor [thrombospondin 2 precursor (Genebank accession number: L12350)] gene,
Chondroitin sulfate proteoglycan core protein 2 (verifiedcan sulfate protein proteoglycan core protein 2);
Embopplakin (Genebank accession number: U53786) gene,
Integrin β-4 chain [integrin, beta 4] (Genebank accession number: X53587) gene,
Vimentin (Genebank accession number: Z19554) gene,
c-fms proto-oncogene [genebank registration number: X03663],
Proliferating cell nuclear antigen (Genebank accession number: M15796) gene,
Leukocyte antigen-related protein (Genebank accession number: Y00815) gene,
Interleukin 8 [interleukin 8] (Genebank accession number: M26383) gene,
GDP dissociation inhibitor (GDI) [GDP dissociation inhibitor (GDI)] (Genebank accession number: L20688) gene,
A brain-specific tubulin α chain [Tubulin, alpha, brain-specific] (Genebank accession number: X01703) gene,
Cell cycle control gene cdc2 (cell division cycle 2, G1 to S phase & G2 to M phase) [cell division cycle 2, G1 to S and G2 to M] (Genebank registration number: X05360),
Tissue-type plasminogen activator [plasminogen activator, tissue] gene bank accession number: M15518] gene,
Stromal cell-derived factor 1 (Gene bank accession number: U16752) gene,
Matrix metalloproteinase 10 (stromelysin 2) (Genebank accession number: X07820) gene,
Matrix metalloproteinase 15 (membrane-inserted) (Genebank accession number: Z48482) gene,
Antioxidant protein [antioxidant protein 1] (Genebank accession number: D49396) gene,
Laminin α-4 chain [laminin, alpha 4] (Genebank accession number: S78569) gene,
The ubiquitin-conjugating enzyme E2M (homologous to yeast UBC12), which is homologous to yeast UBC12; Genebank accession number: AF057599;
Retinoblastoma-binding protein 6 (Genebank accession number: X85133) gene,
K-sum [K-sam] (Genebank accession number: M87770) gene,
Cathepsin B (genebank accession number: L22569) gene,
Signal transduction & transcriptional activator 1; 91 kD [signal transducer and activator of transscription 1, 91 kD] (Genebank accession number: M97935) gene,
Carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reacting) (carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reactive antigen)
antigen)] (Genebank accession number: M18728) gene,
Heparan sulfate proteoglycan (gene bank accession number: M85289) gene,
v-Ki-ras2 Kirsten rat sarcoma virus oncogene homolog [v-Ki-ras2 Kirsten rat sarcoma 2 viral oncogene homolog] (Genebank accession number: M54968) gene,
Ras homolog gene family, factor D [ras homolog gene family, member D] (Genebank accession number: D85815) gene,
Hexabrachion (tenascin C, cytotactin) (Genebank accession number: X78565) gene,
Cyclin D2 [cyclin D2] (Genebank accession number: D13639) gene,
Metalloproteinase 3 tissue inhibitory factor [tissue inhibitor of metalloproteinase 3 (Sorsby fundus dystrophy, pseudoinflammation)] (Genebank registration number: AL023282) gene,
Laminin receptor 1, ribosomal protein SA (laminin receptor 1 (67 kD, ribosomal protein SA)) (Genebank accession number: U43901) gene,
Lactate dehydrogenase A (Genebank accession number: X02152) gene,
Keratin 18 (Genebank accession number: M26326) gene,
Ia-related invariant gamma chain [Ia-associated invariant gamma-chain] (Genebank accession number: AH001484) gene,
CD72 antigen [CD72 antigen] (Genebank accession number: M54992) gene,
Connective tissue growth factor (Genebank accession number: X78947) gene,
Matrix metalloproteinase 3 (stromelysin 1, progelatinase) (Genebank accession number: X05232) gene,
Cyclin-dependent kinase 4 (Genebank accession number: U37022) gene,
A human SB class II histocompatibility antigen [Human mRNA for SB class II histocompatibility antigen alpha-chain] (Genebank accession number: X03100) gene;
Transforming growth factor β-3 chain [transforming growth factor, beta 3] (Genebank accession number: X14885) gene,
Thyroid hormone receptor interacting protein 6 (Gene bank accession number: AJ001902) gene,
Platelet / endothelial cell adhesion molecule (CD31 antigen) (Genebank accession number: L34657) gene,
Profilin 1 (Genebank accession number: J03191) gene,
GTP-binding factor [GTP-binding protein] (Genebank accession number: AF120334) gene,
Homo sapiens clone 24703 β-tubulin mRNA [Homo sapiens clone 24703 beta-tubulin mRNA, complete cds] (Genebank accession number: AF070600) gene,
A Rho-associated coiled-coil containing protein kinase 1 (Genebank accession number: U43195) gene,
Type IV collagen, α-1 chain [collagen, type IV, alpha 1] (Genebank accession number: M26576) gene,
Topoisomerase (DNA) I (gene bank accession number: M60706) gene,
Kunitz type 1 serine proteinase inhibitor, [serine protease inhibitor, Kunitz type 1] (Genebank accession number: AB000095) gene,
Rho GTPase activating protein 1 [Rho GTPase activating protein 1] (Genebank accession number: U02570) gene,
Nidogen (gene bank accession number: M30269) gene,
Cyclin-dependent kinase 5, regulatory factor 1 [cyclin-dependent kinase 5, regulatory subunit 1 (p35)] (Genebank accession number: X80343) gene,
CD9 antigen [CD9 antigen] (Genebank accession number: M38690) gene,
Aortic-type smooth muscle α-actin gene, exon 9 (aortic-type smooth muscle alpha-actin gene, exon 9 (Genebank registration number: M33216)),
Bone morphogenetic protein 1 (BMp1) (Genebank accession number: U50330) gene,
Type VI collagen α-3 chain [collagen type VI alpha-3 (Genebank accession number: X52022)] gene,
A cytokine humig (interkine-gamma-induced monokine (MIG) (Genebank accession number: X72755)) gene,
Integrin alpha-3 chain [integrin alpha-3 chain (Genebank accession number: M59911)] gene,
An interferon-inducible gene family (Genebank accession number: X57351) gene,
A metastasis-associated MTA1 [metastasis-associated MTA1 (Genebank accession number: U35113)] gene,
Notch2 Notch homolog 3 (Genebank accession number: U97669)] gene, and proto-oncogene rhoA multidrug resistance protein (proto-oncogene rhoA multidrug resistance protein (registered trademark; GTP); The array according to claim 19 or 20, wherein nucleic acids corresponding to at least eight genes selected from the group consisting of genes are fixed. 22. The array according to claim 21, wherein at least 20 genes are fixed. A method for evaluating the malignancy of gastric cancer, comprising a primer and / or a probe for detecting mRNA or a fragment thereof expressed from at least eight genes whose expression level changes according to the malignancy of gastric cancer kit. The kit according to claim 23, wherein the primer and / or the probe is for detection by a nucleic acid amplification method. The kit according to claim 24, wherein the nucleic acid amplification method is an RT-PCR method. A gene whose expression level changes according to the degree of malignancy of gastric cancer,
Apoptosis inhibitor survivin (apoptosis inhibitor survivin (Genebank accession number: U75285)) gene,
Type I collagen α-2 [collagen type I α-2 (Genebank accession number: J03464)] gene,
Type III collagen pro-α-1 chain [collagen type III pro-alpha-1 (Genebank accession number: X14420)] gene,
A fibronectin precursor (FN) [fibronectin precursor (FN) (Genebank accession number: X02761)] gene,
MMP-1; gastrointestinal collagenase [matrix metalloproteinase 1; intestinal collagenase (Genebank accession number: M13509)] gene,
MMP-7: uterine matrilysin [matrix metalloproteinase 7; matrilysin, uterine (Genebank accession number: Z11887)] gene,
U-plasminogen activator (EC 3.4.21.73); U-plasminogen activator (UPA); plasmin activator;
Urokinase-type plasminogen activator receptor; GPI-anchored from precursor (U-PAR); monocyte activation antigen gene;
Alpha-2-macroglobulin [alpha-2-macroglobulin (alpha-2-M) (Genebank accession number: M11313)] gene,
β-type platelet-derived growth factor receptor precursor (beta platelet-derived grown factor receptor precursor (PDGFR-beta); CD140B antigen (Genebank registration number: J03278)) gene,
BIGH3 (Genebank accession number: M77349) gene,
XVIII type collagen α-1 chain [collagen type XVIII alpha-1 (Genebank accession number: AF0188081)] gene,
A growth hormone-dependent insulin-like insulin-like growth factor-binding protein (Genebank registration number: M35878)] gene,
Hepatoma-derived growth factor (HDGF) (Genebank accession number: D16431) gene,
An insulin-like growth factor binding protein 2 (IGFBp2) (Genebank accession number: X16302)] gene,
Type IV collagenase [type IV collagenase (Genebank accession number: M55593)] gene,
spac precursor (osteonectin) [sparc precursor (secrete protein acidic and rich in cysteine; osteonectin) (ON); basement membrane protein BM-40;
A thrombospondin 2 precursor [thrombospondin 2 precursor (Genebank accession number: L12350)] gene,
Chondroitin sulfate proteoglycan core protein 2 (verifiedcan sulfate protein proteoglycan core protein 2);
Embopplakin (Genebank accession number: U53786) gene,
Integrin β-4 chain [integrin, beta 4] (Genebank accession number: X53587) gene,
Vimentin (Genebank accession number: Z19554) gene,
c-fms proto-oncogene [genebank registration number: X03663],
Proliferating cell nuclear antigen (Genebank accession number: M15796) gene,
Leukocyte antigen-related protein (Genebank accession number: Y00815) gene,
Interleukin 8 [interleukin 8] (Genebank accession number: M26383) gene,
GDP dissociation inhibitor (GDI) [GDP dissociation inhibitor (GDI)] (Genebank accession number: L20688) gene,
A brain-specific tubulin α chain [Tubulin, alpha, brain-specific] (Genebank accession number: X01703) gene,
Cell cycle control gene cdc2 (cell division cycle 2, G1 to S phase & G2 to M phase) [cell division cycle 2, G1 to S and G2 to M] (Genebank registration number: X05360),
Tissue-type plasminogen activator [plasminogen activator, tissue] gene bank accession number: M15518] gene,
Stromal cell-derived factor 1 (Gene bank accession number: U16752) gene,
Matrix metalloproteinase 10 (stromelysin 2) (Genebank accession number: X07820) gene,
Matrix metalloproteinase 15 (membrane-inserted) (Genebank accession number: Z48482) gene,
Antioxidant protein [antioxidant protein 1] (Genebank accession number: D49396) gene,
Laminin α-4 chain [laminin, alpha 4] (Genebank accession number: S78569) gene,
The ubiquitin-conjugating enzyme E2M (homologous to yeast UBC12), which is homologous to yeast UBC12; Genebank accession number: AF057599;
Retinoblastoma-binding protein 6 (Genebank accession number: X85133) gene,
K-sum [K-sam] (Genebank accession number: M87770) gene,
Cathepsin B (genebank accession number: L22569) gene,
Signal transduction & transcriptional activator 1; 91 kD [signal transducer and activator of transscription 1, 91 kD] (Genebank accession number: M97935) gene,
Carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reacting) (carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reactive antigen)
antigen)] (Genebank accession number: M18728) gene,
Heparan sulfate proteoglycan (gene bank accession number: M85289) gene,
v-Ki-ras2 Kirsten rat sarcoma virus oncogene homolog [v-Ki-ras2 Kirsten rat sarcoma 2 viral oncogene homolog] (Genebank accession number: M54968) gene,
Ras homolog gene family, factor D [ras homolog gene family, member D] (Genebank accession number: D85815) gene,
Hexabrachion (tenascin C, cytotactin) (Genebank accession number: X78565) gene,
Cyclin D2 [cyclin D2] (Genebank accession number: D13639) gene,
Metalloproteinase 3 tissue inhibitory factor [tissue inhibitor of metalloproteinase 3 (Sorsby fundus dystrophy, pseudoinflammation)] (Genebank registration number: AL023282) gene,
Laminin receptor 1, ribosomal protein SA (laminin receptor 1 (67 kD, ribosomal protein SA)) (Genebank accession number: U43901) gene,
Lactate dehydrogenase A (Genebank accession number: X02152) gene,
Keratin 18 (Genebank accession number: M26326) gene,
Ia-related invariant gamma chain [Ia-associated invariant gamma-chain] (Genebank accession number: AH001484) gene,
CD72 antigen [CD72 antigen] (Genebank accession number: M54992) gene,
Connective tissue growth factor (Genebank accession number: X78947) gene,
Matrix metalloproteinase 3 (stromelysin 1, progelatinase) (Genebank accession number: X05232) gene,
Cyclin-dependent kinase 4 (Genebank accession number: U37022) gene,
A human SB class II histocompatibility antigen [Human mRNA for SB class II histocompatibility antigen alpha-chain] (Genebank accession number: X03100) gene;
Transforming growth factor β-3 chain [transforming growth factor, beta 3] (Genebank accession number: X14885) gene,
Thyroid hormone receptor interacting factor 6 [thyroid hormone receptor interacting]
protein 6] (Genebank accession number: AJ001902) gene,
Platelet / endothelial cell adhesion molecule (CD31 antigen) (Genebank accession number: L34657) gene,
Profilin 1 (Genebank accession number: J03191) gene,
GTP-binding factor [GTP-binding protein] (Genebank accession number: AF120334) gene,
Homo sapiens clone 24703 β-tubulin mRNA [Homo sapiens clone 24703 beta-tubulin mRNA, complete cds] (Genebank accession number: AF070600) gene,
A Rho-associated coiled-coil containing protein kinase 1 (Genebank accession number: U43195) gene,
Type IV collagen, α-1 chain [collagen, type IV, alpha 1] (Genebank accession number: M26576) gene,
Topoisomerase (DNA) I (gene bank accession number: M60706) gene,
Kunitz type 1 serine proteinase inhibitor, [serine protease inhibitor, Kunitz type 1] (Genebank accession number: AB000095) gene,
Rho GTPase activating protein 1 [Rho GTPase activating protein 1] (Genebank accession number: U02570) gene,
Nidogen (gene bank accession number: M30269) gene,
Cyclin-dependent kinase 5, regulatory factor 1 [cyclin-dependent kinase 5, regulatory subunit 1 (p35)] (Genebank accession number: X80343) gene,
CD9 antigen [CD9 antigen] (Genebank accession number: M38690) gene,
Aortic-type smooth muscle α-actin gene, exon 9 (aortic-type smooth muscle alpha-actin gene, exon 9 (Genebank registration number: M33216)),
Bone morphogenetic protein 1 (BMp1) (Genebank accession number: U50330) gene,
Type VI collagen α-3 chain [collagen type VI alpha-3 (Genebank accession number: X52022)] gene,
A cytokine humig (interkine-gamma-induced monokine (MIG) (Genebank accession number: X72755)) gene,
Integrin alpha-3 chain [integrin alpha-3 chain (Genebank accession number: M59911)] gene,
An interferon-inducible gene family (Genebank accession number: X57351) gene,
A metastasis-associated MTA1 [metastasis-associated MTA1 (Genebank accession number: U35113)] gene,
Notch2 Notch homolog 3 (Genebank accession number: U97669)] gene, and proto-oncogene rhoA multidrug resistance protein (proto-oncogene rhoA multidrug resistance protein (registered trademark; GTP); No .: L25080)] The kit according to any one of claims 23 to 25, which is at least 8 species selected from the group consisting of genes. The kit according to claim 26, wherein the genes whose expression levels change according to the degree of malignancy of gastric cancer are at least 20 genes. A kit for evaluating the malignancy of gastric cancer, comprising an antibody or a fragment thereof to a polypeptide encoded by at least eight types of genes or fragments thereof whose expression level changes depending on the malignancy of gastric cancer. The polypeptide is an apoptosis inhibitor survivin [apoptosis inhibitor survivin (Genebank accession number: U75285)] gene,
Type I collagen α-2 [collagen type I α-2 (Genebank accession number: J03464)] gene,
Type III collagen pro-α-1 chain [collagen type III pro-alpha-1 (Genebank accession number: X14420)] gene,
A fibronectin precursor (FN) [fibronectin precursor (FN) (Genebank accession number: X02761)] gene,
MMP-1; gastrointestinal collagenase [matrix metalloproteinase 1; intestinal collagenase (Genebank accession number: M13509)] gene,
MMP-7: uterine matrilysin [matrix metalloproteinase 7; matrilysin, uterine (Genebank accession number: Z11887)] gene,
U-plasminogen activator (EC 3.4.21.73); U-plasminogen activator (UPA); plasmin activator;
Urokinase-type plasminogen activator receptor; GPI-anchored from precursor (U-PAR); monocyte activation antigen gene;
Alpha-2-macroglobulin [alpha-2-macroglobulin (alpha-2-M) (Genebank accession number: M11313)] gene,
β-type platelet-derived growth factor receptor precursor (beta platelet-derived grown factor receptor precursor (PDGFR-beta); CD140B antigen (Genebank registration number: J03278)) gene,
BIGH3 (Genebank accession number: M77349) gene,
XVIII type collagen α-1 chain [collagen type XVIII alpha-1 (Genebank accession number: AF0188081)] gene,
A growth hormone-dependent insulin-like insulin-like growth factor-binding protein (Genebank registration number: M35878)] gene,
Hepatoma-derived growth factor (HDGF) (Genebank accession number: D16431) gene,
An insulin-like growth factor binding protein 2 (IGFBp2) (Genebank accession number: X16302)] gene,
Type IV collagenase [type IV collagenase (Genebank accession number: M55593)] gene,
spac precursor (osteonectin) [sparc precursor (secrete protein acidic and rich in cysteine; osteonectin) (ON); basement membrane protein BM-40;
A thrombospondin 2 precursor [thrombospondin 2 precursor (Genebank accession number: L12350)] gene,
Chondroitin sulfate proteoglycan core protein 2 (verifiedcan sulfate protein proteoglycan core protein 2);
Embopplakin (Genebank accession number: U53786) gene,
Integrin β-4 chain [integrin, beta 4] (Genebank accession number: X53587) gene,
Vimentin (Genebank accession number: Z19554) gene,
c-fms proto-oncogene [genebank registration number: X03663],
Proliferating cell nuclear antigen (Genebank accession number: M15796) gene,
Leukocyte antigen-related protein (Genebank accession number: Y00815) gene,
Interleukin 8 [interleukin 8] (Genebank accession number: M26383) gene,
GDP dissociation inhibitor (GDI) [GDP dissociation inhibitor (GDI)] (Genebank accession number: L20688) gene,
A brain-specific tubulin α chain [Tubulin, alpha, brain-specific] (Genebank accession number: X01703) gene,
Cell cycle control gene cdc2 (cell division cycle 2, G1 to S phase & G2 to M phase) [cell division cycle 2, G1 to S and G2 to M] (Genebank registration number: X05360),
Tissue-type plasminogen activator [plasminogen activator, tissue] gene bank accession number: M15518] gene,
Stromal cell-derived factor 1 (Gene bank accession number: U16752) gene,
Matrix metalloproteinase 10 (stromelysin 2) (Genebank accession number: X07820) gene,
Matrix metalloproteinase 15 (membrane-inserted) (Genebank accession number: Z48482) gene,
Antioxidant protein [antioxidant protein 1] (Genebank accession number: D49396) gene,
Laminin α-4 chain [laminin, alpha 4] (Genebank accession number: S78569) gene,
The ubiquitin-conjugating enzyme E2M (homologous to yeast UBC12), which is homologous to yeast UBC12; Genebank accession number: AF057599;
Retinoblastoma-binding protein 6 (Genebank accession number: X85133) gene,
K-sum [K-sam] (Genebank accession number: M87770) gene,
Cathepsin B (genebank accession number: L22569) gene,
Signal transduction & transcriptional activator 1; 91 kD [signal transducer and activator of transscription 1, 91 kD] (Genebank accession number: M97935) gene,
Carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reacting) (carcinoembryonic antigen-related cell adhesion molecule 6 (non-specific cross-reactive antigen)
antigen)] (Genebank accession number: M18728) gene,
Heparan sulfate proteoglycan (gene bank accession number: M85289) gene,
v-Ki-ras2 Kirsten rat sarcoma virus oncogene homolog [v-Ki-ras2 Kirsten rat sarcoma 2 viral oncogene homolog] (Genebank accession number: M54968) gene,
Ras homolog gene family, factor D [ras homolog gene family, member D] (Genebank accession number: D85815) gene,
Hexabrachion (tenascin C, cytotactin) (Genebank accession number: X78565) gene,
Cyclin D2 [cyclin D2] (Genebank accession number: D13639) gene,
Metalloproteinase 3 tissue inhibitory factor [tissue inhibitor of metalloproteinase 3 (Sorsby fundus dystrophy, pseudoinflammation)] (Genebank registration number: AL023282) gene,
Laminin receptor 1, ribosomal protein SA (laminin receptor 1 (67 kD, ribosomal protein SA)) (Genebank accession number: U43901) gene,
Lactate dehydrogenase A (Genebank accession number: X02152) gene,
Keratin 18 (Genebank accession number: M26326) gene,
Ia-related invariant gamma chain [Ia-associated invariant gamma-chain] (Genebank accession number: AH001484) gene,
CD72 antigen [CD72 antigen] (Genebank accession number: M54992) gene,
Connective tissue growth factor (Genebank accession number: X78947) gene,
Matrix metalloproteinase 3 (stromelysin 1, progelatinase) (Genebank accession number: X05232) gene,
Cyclin-dependent kinase 4 (Genebank accession number: U37022) gene,
A human SB class II histocompatibility antigen [Human mRNA for SB class II histocompatibility antigen alpha-chain] (Genebank accession number: X03100) gene;
Transforming growth factor β-3 chain [transforming growth factor, beta 3] (Genebank accession number: X14885) gene,
Thyroid hormone receptor interacting protein 6 (Gene bank accession number: AJ001902) gene,
Platelet / endothelial cell adhesion molecule (CD31 antigen) (Genebank accession number: L34657) gene, profilin 1 (profilin 1) (Genebank accession number: J03191) gene,
GTP-binding factor [GTP-binding protein] (Genebank accession number: AF120334) gene,
Homo sapiens clone 24703 β-tubulin mRNA [Homo sapiens clone 24703 beta-tubulin mRNA, complete cds] (Genebank accession number: AF070600) gene,
A Rho-associated coiled-coil containing protein kinase 1 (Genebank accession number: U43195) gene,
Type IV collagen, α-1 chain [collagen, type IV, alpha 1] (Genebank accession number: M26576) gene,
Topoisomerase (DNA) I (gene bank accession number: M60706) gene,
Kunitz type 1 serine proteinase inhibitor, [serine protease inhibitor, Kunitz type 1] (Genebank accession number: AB000095) gene,
Rho GTPase activating protein 1 [Rho GTPase activating protein 1] (Genebank accession number: U02570) gene,
Nidogen (gene bank accession number: M30269) gene,
Cyclin-dependent kinase 5, regulatory factor 1 [cyclin-dependent kinase 5, regulatory subunit 1 (p35)] (Genebank accession number: X80343) gene,
CD9 antigen [CD9 antigen] (Genebank accession number: M38690) gene,
Aortic-type smooth muscle α-actin gene, exon 9 (aortic-type smooth muscle alpha-actin gene, exon 9 (Genebank registration number: M33216)),
Bone morphogenetic protein 1 (BMp1) (Genebank accession number: U50330) gene,
Type VI collagen α-3 chain [collagen type VI alpha-3 (Genebank accession number: X52022)] gene,
A cytokine humig (interkine-gamma-induced monokine (MIG) (Genebank accession number: X72755)) gene,
Integrin alpha-3 chain [integrin alpha-3 chain (Genebank accession number: M59911)] gene,
An interferon-inducible gene family (Genebank accession number: X57351) gene,
A metastasis-associated MTA1 [metastasis-associated MTA1 (Genebank accession number: U35113)] gene,
Notch2 Notch homolog 3 (Genebank accession number: U97669)] gene, and proto-oncogene rhoA multidrug resistance protein (proto-oncogene rhoA multidrug resistance protein (registered trademark; GTP); No .: L25080)] The kit according to claim 28, which is at least eight kinds of polypeptides selected from the polypeptides encoded by each of the genes. 30. The kit of claim 29, wherein the polypeptide is at least 20 polypeptides.

Images