甲基化去氧核醣核酸作為亞洲肺癌病患診斷及疾病監測生物標記之研究

Abstract

肺癌於台灣及世界上許多國家皆高居十大癌症死因第一名，五年存活率不到百分之二十; 其中，早期診斷的病患五年存活率可達一半以上，晚期發現的病患其五年存活率則驟降至小於百分之五。然而，肺癌病患只有約百分之十五於第一期發現癌症，超過一半的病患在發現罹患肺癌時，病程發展都已至末期。現今之肺癌診斷方式主要仰賴影像技術，病理檢查及分子診斷。在所有診斷方式中，分子診斷是依據去氧核醣核酸、核醣核酸或蛋白質的異常來區分癌細胞及正常細胞，這些異常包含基因(例如：突變、基因轉置等)及表基因的變化 (例如：組蛋白修飾及去氧核醣核酸甲基化)。去氧核醣核酸之異常甲基化相當穩定且廣泛存在於各種癌症當中，因此可用來當做基因突變之外之癌症特異性生物標記。過去有許多研究使用去氧核醣核酸之異常甲基化當作肺癌生物標記，然而這些甲基化基因皆來自西方國家之病人族群，而亞洲病患肺癌具有不同之成因及分子特徵，故 所提出之甲基化標記並不一定適用於亞洲病患。本研究中，我們使用美國癌症基因體圖譜計畫的全基因體甲基化資料庫，使用生物資訊分析篩選出於肺微低癌組織中表現高度甲基化，且於正常肺組織中具低甲基化的特定基因。我們發現了四個基因之最佳組合，並進一步使用定量甲基化聚合酶連鎖反應偵測此四基因組合於肺癌細胞株、肺癌組織及健康人周邊血液單核球細胞中之甲基化程度。結果發現，此四基因甲基化組合之敏感度高達95.8%，特異度達90%，而其陽性預測值為92.5%，陰性預測值為92.3%。除肺癌組織檢驗之外，我們亦嘗試使用微滴數字甲基化聚合酶連鎖反應偵測血液中之甲基化循環游離核酸，並於數個血清檢體中測試成功。同時，我們於台大醫院啟動肺癌病人之收案，於診斷時及治療中每三個月收集血清檢體，收案仍在進行中。總結而言，我們所研發之四基因甲基化組合具有極佳潛力，可據以發展協助診斷肺癌病患之血液檢測工具。

Lung cancer causes the most cancer-related deaths all over the world. The five-year survival rate of lung cancer patients is over fifty percent with early-staged diseases, but it plummets to less than five percent at advanced stages. Only about 15% of lung cancer patients were diagnosed at early stages while more than half of patients were in late stages when diagnosed. The diagnosis of lung cancer is mainly based on imaging, pathological examination and molecular diagnosis. Among these modalities, molecular diagnosis relies on abnormalities on nucleic acid or protein levels that can distinguish between cancer and normal cells, such as genetic (i.e. mutations, translocations, etc) or epigenetic variations (i.e. histone modifications and DNA methylation). Abnormal DNA methylation is relatively stable and ubiquitously present in virtually all types of cancer; therefore, it can serve as alternative cancer-specific biomarkers as opposed to genetic mutations. For the past decades, many studies have explored the potentials of abnormal DNA methylation as clinically-applicable biomarkers in lung cancer. However, most methylated markers were identified from patient cohorts in which Asian populations were under-represented. Given the fact that Asian lung cancers differ from their counterparts in western countries in terms of etiologies and molecular characteristics, whether methylated biomarkers proposed in previous studies can be used in Asian lung cancer patients are questionable. In this study, we utilized genome-wide methylation data from the database of the Cancer Genome Atlas (TCGA) in the United States, and performed bioinformatics analysis to identify candidate genes that harbor high levels of DNA methylation in the promoter regions in lung cancer and low levels in normal lung tissues. With this approach, we selected for a best combination of four genes. Then, we performed quantitative methylation-specific PCR (qMSP) to measure methylation levels of this four-gene combination in lung cancer cell lines, lung cancer tissues, and peripheral blood mononuclear cells from healthy volunteers. Our results showed that the four-gene combination had sensitivity of 95.8% and specificity of 90%. The positive predictive value of this gene set is 92.5% and the negative predictive value is 92.3%. In addition to tissue-based assays, to further explore the applicability of this four-gene combination in blood-based assays, we attempted to establish a platform using droplet digital MSP (ddMSP) to identify circulated methylated DNAs in the blood, and have successfully implemented the assays on a few plasma samples as a test. We have also started recruited lung cancer patients and obtain their plasmas at diagnosis, and every three months during treatment at National Taiwan University Hospital. The patient enrollment is still ongoing. In summary, DNA methylation of our 4-gene combination has a promising potential to server as a basis for developing blood-based diagnostic assays in Asian lung cancer patients.