摘要
乳癌至今仍然是目前世界上女性最常見的癌症，而乳癌的早期診斷是乳癌治療的目標之一，研究也證實早期診斷不僅可降低乳癌發生率及死亡率，也可以大大的改善治療的選擇，更可以提高治癒的成功率。因此本研究最主要的目的是想研發全新的乳癌診斷分子標記及技術平台，做為乳癌診斷的輔助工具，希望藉此可提升早期乳癌的檢出率及降低其死亡率。許多的研究學者都認為，乳癌是一個全身性的疾病(systemic disease)，因為在早期或腫瘤很小的乳癌其癌細胞就可能會散佈到血流或淋巴系統中。所以本研究一開始即利用“即時定量聚合酶鏈鎖反應技術”來檢測乳癌病人組織及周邊血液中的癌細胞，而所使用的分子標記包括: CK-19、CEA、hMAM、c-Met及Her2/neu等五個。結果在病人血液中這五個分子標記過度表現的比例分別是80%、56%、60%、39%及31%。此外，當這五個標記中有三個有過度表現時，則靈敏度仍可維持在80%，但特異性可提升至81%。而雖然這樣的組合，於乳癌的診斷上不管是靈敏度或特異性均優於目前臨床上所使用的腫瘤標記，但仍有其缺點。最主的原因是此技術侷限在一個檢體一次只能偵測一個標記，因此若要同時檢測多個標記則會相當的耗費時間。有鑑於此，我們利用本實驗室成功研發的呈色型晶片，做為乳癌檢測的技術平台並與即時定量聚合酶鏈鎖反應技術做比較，以評估兩種檢測方法對於臨床上乳癌患者的診斷價值。所使用的分子標記也是 CK-19、CEA、hMAM、c-Met及Her2/neu等五個。而實驗結果發現，呈色型晶片每1毫升只要有5個癌細胞即可偵測出來。且經由ROC curve及linear regression的生物統計分析，發現呈色型晶片與即時定量聚合酶鏈鎖反應技術所得到的結果是相當一致的(r=0.979, p<0.001)。此外，我們也發現訊息核糖核酸(mRNA)的過度表現與乳癌患者的tumor size(p=0.008)及TNM stage(p=0.013)等臨床病理資料有明顯的關聯性。而不同於即時定量聚合酶鏈鎖反應，呈色型晶片可以同時獲得多個分子標記的資訊，且具有簡單、快速、適合用於大量篩檢等優點。而為了開發出具有台灣本土性乳癌的分子標記，我們利用微矩陣列技術分析，三對不同型態的乳癌組織(分別為Infiltrating lobular cancer、Metaplastic carcinoma lobular carcinoma、Infiltrating ductal cancer)，並從中篩選出87個在三組實驗中均有3倍以上高度表現之基因，接著利用生物資訊軟體來作進一步統整分析，發現其中有20個基因與乳癌有關，然後以臨床檢體作測試，證實其中有4個基因(PTTG1, Survivin, UbcH10, TK1)在乳癌組織中明顯的過度表現且與細胞的分裂調控或腫瘤的形成有關。而我們也更進一步利用呈色型晶片，結合這4個分子標記以檢測乳癌病人血液中的循環性癌細胞。經由統計分析結果PTTG1, Survivin, UbcH10, and TK1的靈敏度及特異性分別介於76%~85%和75%~79%之間。而當4個分子標記中有3個過度表現就判定為陽性時，則其靈敏度及特異性就分別可達到86%及88%，且準確度有93%，而這個結果明顯優於先前所使用的5個分子標記。此外，我們也發現乳癌患者的tumor size(p=0.008)、histologic grade(p=0.038)、lymph node metastasis(p=0.001)及TNM stage(p=0.014)等臨床病理資料與整組的標記有明顯的相關性。雖然呈色型晶片可應用於臨床癌症診斷上，但由於呈色法的操作與判讀易受到人為誤差所影響，因此在靈敏度及特異性仍有提升的空間。所以，我們更進一步的研發冷光型晶片作為乳癌檢測的技術平台，而所使用的分子標記，包括PTTG1、Survivin、UbcH10及TK1等。結果證實，冷光型晶片的確比呈色型晶片更準確，提升了檢測的靈敏度及特異性( 92 % , 93 %)。藉由本研究我們(一) 開發新的分子標記及(二)研發冷光型晶片，希望可以提升乳癌診斷的準確度，以掌握最的佳治療時機，而這對於乳癌之早期診斷及術後追蹤治療，都將會是突破性的進展。
並列摘要
Early detection is regarded as the best defense of medical intervention against breast cancer. Studies have reported that early diagnosis is one of the most effective ways to reduce incidence and death rates of breast cancer. A case of early diagnosed breast cancer may not only have more choices of medical therapy, but a better chance to be cured successfully as well. The purpose of this research is to develop a novel molecular marker and technological platform for breast cancer diagnosis, in the hope of improving early detection and reducing mortality through breast cancer. Many researchers believe that breast cancer is a systemic disease, because breast cancer cells may start to disseminate and be released into the bloodstream and lymph system in an early stage, or when the tumor nodules remain small. First, we used real-time quantitative PCR (Q-PCR) to detect the expression of mRNA markers in the peripheral blood and tissue of breast cancer patients. CK-19, CEA, hMAM, c-Met, and Her2/neu were chosen as candidate markers. The frequency of CK-19, CEA, hMAM, c-Met, and Her2/neu mRNA overexpression was 80%, 56%, 60%, 39%, and 31% respectively. In addition, a combined panel of 5 mRNA markers was demonstrated to achieve sensitivity of 80% and specificity of 81% for breast cancer detection. Despite the sensitivity and specificity of Q-PCR being much higher than those of tumor markers, this technique has the limitation of being able to monitor only one or a few markers for each specimen in a single test. Examinations containing quantities of samples or molecular markers will consume much time and effort. For this reason, we exploited membrane array, previously established in our laboratory, as the technological platform for breast cancer diagnosis. We further compared the clinical diagnostic value of real-time PCR and colorimetric membrane array. We also used CK-19, CEA, hMAM, c-Met, and Her2/neu as candidate markers. We found that the five-RNA marker membrane array method could positively identify circulating breast cancer cells at a density as low as 5 cells per 1ml of blood. In addition, data obtained from real-time Q-PCR and membrane array were subjected to linear regression analysis, revealing that there was a high degree of correlation between the results of these two methods (r=0.979, p<0.0001). The analysis of correlation between the outcome of membrane array and clinicopathological characteristics indicated that overexpression of markers was significantly correlated with tumor size (p=0.008) and TNM stage (0.013). In contrast to Q-PCR, membrane array could not only analyze simultaneously the expression of multiple mRNA markers, but was an easier operation with this high-throughput technique. In order to develop specific molecular markers of breast cancer in Taiwan, we used microarray to analyze three pairs of normal and breast cancer tissue samples (including infiltrating lobular cancer、metaplastic carcinoma lobular carcinoma、and infiltrating ductal cancer). We found that 87 genes showed up-regulation (ratio more than three) in all three pairs of tissue. Then we combined several bioinformatic software programs to analysis the gene profile from the microarray experiment, and picked up 20 target genes that are meaningful in breast cancer. Of those, PTTG1, Survivin, UbcH10, and TK1 were significantly overexpressed in breast cancer tissues and related to regulation of cell cycle or carcinogenesis. We further employed colorimetric membrane array to detect PTTG1, Survivin, UbcH10, and TK1 expressed in the peripheral blood of female patients with breast cancer. Statistical analyses of data indicated that the sensitivity for the markers ranged between 76% and 85% and the specificity between 75% and 79%. For the panel of the four mRNA markers, the sensitivity, specificity, and accuracy were elevated up to 86%, 88%, and 93% respectively. This result obviously surpassed our previous study (CK-19, CEA, hMAM, c-Met, and Her2/neu). Furthermore, it was found that patient clinicopathological characteristics tumor size (p=0.006), histologic grade (p=0.012), lymph node metastasis (p=0.001), and TNM stage (p=0.006) were significantly correlated with the positive detection rate of the multi-marker panel. Although colorimetric membrane array can significantly reduce the difficulty and cost of experiments and help to increase its popularity and application in the medical marketplace, the operation and reading of colorimetric methods are easily affected by artifacts. Therefore, we are starting to develop another chemiluminescent gene chip to detect circulating tumor cells in peripheral blood. We also used PTTG1, Survivin, UbcH10, and TK1 as candidate markers. The sensitivity and specificity of luminescent gene chips were 92% and 93% respectively in detecting circulating tumor cells in peripheral blood of breast cancer patients. The initial results showed that luminescent gene chips can increase the accuracy of diagnosis as a whole. In the present study, we develop (1) novel molecular marker and (2) innovative chemiluminescent membrane array for breast cancer diagnosis. Hopefully, the goal of early breast cancer detection can be achieved through this study, which will in turn improve the efficacy of therapies managing this malignancy. This technique also has a potential for application in monitoring disease progression of breast cancer. We believe that this technique will become an invaluable tool in the fight against breast cancer.
參考文獻
延伸閱讀
- 楊淯翔（2008）。乳癌新腫瘤標記之研發〔碩士論文，高雄醫學大學〕。華藝線上圖書館。https://doi.org/10.6832/KMU.2008.00047
- Cheng, F. T. F., Chou, Y. H., Chen, Y. K., & Hsi, S. C. (2007). 轉移性乳癌之節拍式化學療法：初期報告. Formosan Journal of Surgery, 40(2), 75-80. https://doi.org/10.29833/FJS.200704.0003
- 曾令民（2008）。乳癌標靶治療。中華民國癌症醫學會雜誌，24(2)，132-138。https://doi.org/10.6588/JCOS.2008.24.2.9
- 沈志陽（2008）。乳癌的分子生物學與分子遺傳學。中華民國癌症醫學會雜誌，24(2)，94-97。https://doi.org/10.6588/JCOS.2008.24.2.2
- 郭文宏（2013）。Hormonal Therapy in Breast Cancer。台灣醫學，17(4)，397-403。https://doi.org/10.6320/FJM.2013.17(4).07