循環甲基化去氧核醣核酸作為非小細胞肺癌疾病監測的生物標誌物之研究

Abstract

肺癌目前高居全球所有癌症死亡率的首位，雖然肺癌的治療在過去數十年有顯著的突破，然而肺癌病患的五年存活率仍只有18.6%，反映臨床上對於選擇合適的治療方案及評估治療效果時面臨挑戰。血液中的游離腫瘤DNA (Circulating tumor DNA; ctDNA)是因腫瘤細胞壞死、凋亡或分泌作用所釋放出的DNA，其會進入血液循環，具備與腫瘤細胞相同的突變或DNA甲基化(DNA methylation)特徵。DNA methylation是一種主要的表觀遺傳(Epigenetic)調控機制，在肺癌的致癌過程中，腫瘤抑制基因(Tumor suppressor gene)的啟動子(Promoter)附近的CpG位點區域會發生高度甲基化(Hypermethylation)，引起抑癌基因轉錄失活，從而導致癌細胞的產生，故透過監測ctDNA之異常甲基化程度及濃度水平可用於動態追蹤治療效果與病情進展。 本實驗室過去為了找出肺癌疾病監測之生物標記，使用台大醫院與美國癌症基因體圖譜計畫(The Cancer Genome Altas; TCGA)的資料庫分析得出最佳之基因組合，對於偵測肺癌細胞株、肺癌組織及健康人周邊血液單核球細胞之甲基化程度，敏感度及特異度皆高達9成以上，因此，本研究將延伸至利用微滴式數位核酸偵測系統(Droplet digital PCR; ddPCR)偵測台大醫院肺癌病患血漿中的甲基化ctDNA。結果顯示，病人的ctDNA濃度明顯高於健康受試者的游離DNA濃度，並且癌細胞的ctDNA片段長度平均較一般游離DNA短；通過偵測本四基因組合於肺癌患者的ctDNA及健康人的游離DNA之甲基化程度，發現病患的ctDNA甲基化程度較高，反之於健康人的游離DNA甲基化程度較低，且敏感度及特異度皆超過8成以上，證明本基因組合的確具有肺癌特異性；除此之外，我們利用測得之肺癌患者甲基化ctDNA與臨床資料比對分析，不僅發現ctDNA甲基化程度確實與病患的治療反應和病情發展有明顯關係，而且透過本候選基因組合測得病患ctDNA甲基化程度上升，能夠比影像學檢查更早發現病情的惡化，具有追蹤與預測病情發展的作用，因此可知本基因組合極具發展成為肺癌疾病監測之生物標記的潛力。未來我們將會分析更多肺癌患者的ctDNA和健康受試者的游離DNA樣本，進一步優化此候選基因組合，期望透過本研究的執行，發展可監控肺癌病情與復發之生物標誌物，以建立低侵入性的肺癌疾病監測方法，未來或可供醫師進行臨床治療之參考。

Lung cancer is the leading cause of cancer deaths worldwide. Despite the advancement of treatment for lung cancer, the overall 5-year survival rate for all types of lung cancer patients is still around 18.6% due to drug resistance or disease progression under treatment. Therefore, developing biomarkers for disease monitoring and drug response evaluation in a timely fashion will facilitate adjusting treatment plan to improve patients’ clinical outcome. Circulating tumor DNA (ctDNA) is released into bloodstream by necrosis, apoptosis or secretion of circulating tumor cells (CTCs). ctDNA fragments contain the same genetic alterations and epigenetic changes as those found in the primary tumors. DNA methylation is a major mechanism of epigenetic regulation. Hypermethylation, which often occurs in the CpG islands near the promoter of tumor suppressor genes, is a hallmark of carcinogenic process. As a result, use of methylated ctDNA as a blood-based biomarker represents a promising strategy for disease monitoring. In order to identify candidate methylated genes as biomarkers for lung cancer, our lab has found a four-gene combination from the lung cancer methylation database of National Taiwan University Hospital (NTUH) and the Cancer Genome Altas (TCGA). Moreover, we have previously validated the methylation levels of four probes in primary lung cancer tissues, lung cancer cell lines and peripheral blood mononuclear cells (PBMC) from healthy volunteers. The sensitivity and specificity of the four-gene combination are as high as 90%. In this thesis study, we examined methylation levels of the four gene panel in plasma cfDNA from non-small cell lung cancer (NSCLC) patients in NTUH by using methylation-specific droplet digital PCR (MS-ddPCR). The results showed that the concentrations of ctDNA in patients were significantly higher than those in healthy subjects, and the size of ctDNA fragments originated from tumor cells were shorter than cfDNA fragments originated from normal cells. We also found that our four-gene combination had higher levels of ctDNA methylation in lung cancer patients as compared with healthy volunteers, the gene panel yielded a sensitivity of 82.19% and specificity of 85%, suggesting that this four-gene panel was highly cancer-specific. In addition, we explored the clinical applicability of this combination by longitudinally monitoring the dynamics of ctDNA methylation levels and their relationships with treatment response in lung cancer patients. We discovered a significant association between ctDNA methylation levels and tumor burden dynamics, the changes in ctDNA methylation levels may predict disease progression following initial favorable therapeutic responses, which was even earlier than radiological assessment in the clinic. In summary, our gene panel showed the potential clinical utility as biomarkers for disease monitoring in patients with NSCLC. In the future, we will study more samples from patients and healthy volunteers to further optimize this 4-probe combination. This study provides a foundation for developing minimally-invasive blood-based assays for monitoring treatment response and disease progression in lung cancer patients.