| 摘要: | 研究背景：乳癌為全世界女性盛行率最高的癌症，早期乳癌的預後顯
著高於晚期乳癌，因此乳癌的偵測對於病人預後極其重要。乳癌的發
生與基因體及表觀基因體的變異緊密相關，特定的DNA 甲基化及體
細胞突變可用於癌症的偵測。
研究目的：台灣民眾乳癌篩檢率普遍偏低，因此如何建構出簡便且高
靈敏性的乳癌篩檢方式為改善乳癌預後的重要議題，也是本論文的研
究目的。
研究方法：本研究以專一性的引子(primer)及探針(probe)在血漿中偵
測癌細胞游離DNA 之甲基化變異及體細胞突變。此檢測方式以抽血
進行，普遍大眾的接受度較高；使用qPCR 平台，相較於其他乳房檢
測方式有較高的檢測客觀性，最終希望應用於臨床偵測。
研究結果：本研究利用TCGA 資料庫篩選出22 個體細胞突變位點及
12 個DNA 甲基化變異之基因(ADCY4、ADAMTS20、C1orf114、ZNF177、
ITPRIPL1、ENPP2、SLC7A4、LOC643719、GCM2、TTBK1、TSTD1、
LNX1)，此外，由於RASSF1 在過去研究已知在乳癌組織中高度甲基
化，因此也會將此基因納入檢測。針對DNA 甲基化部分，將13 個候
選基因在55 對本土乳癌病人組織中做驗證後，剃除在腫瘤組織無明
顯變異之基因TTBK1，進一步在7 位乳癌病人血漿中進行ZNF177、
VII
ITPRIPL1、ENPP2、RASSF1、GCM2 的檢測，ITPRIPL1 及ENPP2 在
七位病人皆有檢測出異常甲基化，ZNF177 及GCM2 則有五位病人被
檢測出，RASSF1 有一位病人被檢測出。
進一步分析12 個異常甲基化候選基因與臨床參數之相關性，多
個基因的甲基化程度與雌激素受體及人類表皮生長因子受體2 的表
現呈顯著相關；而SLC7A4、ITPRIPL1 及ENPP2 之DNA 甲基化與
RNA 表現量有高度負相關；進一步發現SLC7A4 與ITPRIPL1 的DNA
甲基化程度及RNA 表現量皆與乳癌病人的預後有顯著相關，由於
SLC7A4 在亞洲人種高甲基化比例最高，因此後續針對SLC7A4 做進
一步的研究。
SLC7A4 在TCGA 資料庫及本土病人中皆發現與HER2 表現顯著
相關， 因此將MDA-MB-231 大量表現SLC7A4 基因後發現
ERBB2(HER2)的RNA 表現量降低。SLC7A4 與ERBB2 以現有文獻並
無直接關聯性，須進一步探討其中的機轉。
結論：本研究篩選出12 個甲基化變異之基因與22 個突變位點，未來
會在本土病人進一步驗證候選的基因突變位點；甲基化部分已在本土
病人組織及血漿中驗證，未來會持續增加樣本數，並在健康人的血漿
中進行檢測，希望可以建立可信的檢測技術並供臨床用於乳癌之偵測。
Background and purpose: Breast cancer is the most common female cancer among the world. There is strong evidence that delayed diagnosis of breast cancer is associated with poor prognosis. Early detection remains the primary defense available to patients in preventing the development of life-threatening breast cancer. However, conventional screening tests still have a range of known limitations.
Alterations of genetic and epigenetic features can provide important insights into the natural history of breast cancer. Nevertheless, a reproducible blood test for diagnosis of breast cancer has yet to be successfully developed into a routine clinical test. The aim of this study is based on aberrant DNA methylation and somatic mutation to establishing a new breast cancer early detection system to enhance breast cancer screening rate and improve prognosis.
Method and result: To search for valid biomarkers, we characterize DNA methylation patterns in breast cancer genome based on Illumina HumanMethylation450K microarray from TCGA database and screening out 10 hyper-methylated genes(ADCY4, C1orf114, GCM2, ZNF177, TTBK1, ADAMTS20, ENPP2, SLC7A4, ITPRIPL1, LOC643719) and 2 hypo-methylated genes(LNX1, TSTD1). RASSF1, as a well-known methylated gene in breast cancer, was put into our multi-gene panel to compare with others. We used qMSP (quantitative methylation-specific polymerase chain reaction) to examine tumor and adjacent normal tissue from 55 breast cancer patients for the presence of aberrant methylation in 13 candidate genes. We observed the some trend as TCGA database except TTBK1. Further, 7 plasma samples of breast cancer patients were detected and 7 samples are all methylated in ITPRIPL1 and ENPP2, 5 out of 7
samples are methylated in ZNF177 and GCM2, 1 out of 7 samples is methylated in RASSF1. In the part of somatic mutation, we screened out 22 mutation sites form TCGA database. We will design specific primer to detect somatic mutation in breast cancer and adjacent normal tissue.
Correlation between methylated biomarkers and clinical parameters was further discussed. The RNA expression of SLC7A4, one of candidate genes, was detected by RT real-time PCR.
There are more amount of Asian patients with SLC7A4 hypermethylation and SLC7A4 are significantly associated with poor prognosis.
Therefore, more sturdy will focus on SLC7A4. Negative correlation was observed between SLC7A4 DNA methylation and RNA expression level. We further demonstrated that promoter DNA methylation in SLC7A4 regulates its RNA expression by decitabine-induced demethylation assay.
Clinically, patients with SLC7A4 low expression highly correlated with those with HER2 positive. Little is known about relationship between SLC7A4 and ERBB2 (HER2), however, SLC7A4 overexpression in MDAMB- 231 lead to decrease ERBB2 expression. The relationship between
SLC7A4 and ERBB2 should be further discussed.
Conclusion: Aberrant DNA methylation in tumor tissue offer prognostic and predictive information. Our findings provide 12 methylated biomarkers for breast cancer detection and SLC7A4 methylation status as potential prognostic biomarkers in breast cancer. We will further detect the methylated biomarkers in plasma to construct a valid multi-gene panel kit for breast cancer detection. |
