惡性攝護腺癌病人尿液中生物標記之探討與其作為預後指標之應用

Abstract

攝護腺癌 (prostate cancer, PCa) 在已開發國家中是一個常好發於男性的惡性腫瘤，也包括台灣。然而，攝護腺癌不是一個致死率高的癌症，並非所有的攝護腺癌病人最終是死於癌症本身。因此，如何區分出致死性與非致死性的攝護腺癌，在臨床上是一項重要的課題。對於非致死性攝護腺癌病人，採取主動監測 (active surveillance) 的方式能夠降低過度治療 (over-treatment) 的問題，也可以避免積極性治療造成的副作用並使病人有較好的生活品質。在本篇研究中，我們考量到前列腺液能夠分泌到尿液中的特性，因此提出在致死性與非致死性攝護腺癌病人的尿液之中，存有能夠作為鑑別這兩組病人的生物標記 (biomarker)。為了找出這些生物標記蛋白，我們利用液相色譜與串聯質譜分析 (Liquid chromatography tandem mass spectrometry, LC-MS/MS) 對病人尿液中的蛋白進行蛋白質身分鑑 定，並利用正交偏最小平方判別分析法 (Orthogonal partial least squares discriminant analysis, OPLS-DA) 對資料進行降維分析，且鑑別出兩組病人尿液之中有顯著差異的蛋白圖譜。分析之後，我們找出五個在致死性攝護腺癌病人尿液中有顯著下降的蛋白生物標記 (significant PCa biomarkers, SPBs)。在診斷及區分致死性與非致死性攝護腺癌病人方面，SPBs在接收操作特徵圖 (receiver operating characteristic curve, ROC curve) 中的曲線下面積 (area under curve, AUC) 可以達到 0.933，而敏感度與特異性則分別可以達到88%與89%，優於現行血液中攝護腺癌生物標記-攝護腺特異性抗原 (prostate specific antigen, PSA) 之表現。此外，SPBs也能夠增加PSA診斷的準確度，當合併SPBs與PSA時，AUC能夠提升至0.967。我們另外也探討了SPBs作為預後指標的可能性，結果發現在接受特定治療如雄性素剝奪療法 (androgen deprivation therapy, ADT)、男性荷爾蒙抑制劑療法 (Abiraterone及Enzalutamide) 以及全身性化療的病人之中，SPBs也具有作為病人預後指標的潛力。總結來說，我們的結果支持SPBs能夠在臨床上作為鑑別致死性與非致死性攝護腺癌病人的尿液中生物標記。另外，SPBs也能夠做為治療中癌症病人的預後指標，在療程中監控病人對藥物治療之反應，並能夠及時調整及改善對於病人的治療方針。

Prostate cancer (PCa) is a common male malignancy in developed countries including Taiwan. However, not all PCa leads to cancer death. Stratification of PCa patients who are lethal or non-lethal will be an important issue to prevent over-treatment and keep life of quality of patients. In this study, we took advantage of the fact that the secretory fluid from prostate has a direct access to urine, and proposed that the protein profiles of lethal PCa patients’ urine may be distinct from those in the non-lethal PCa patients. We then used LC-MS/MS analysis and a supervised approach [Orthogonal Partial Least Squares Discriminant Analysis (OPLS-DA)] for the urine sample analyses, and isolated five significant PCa biomarkers (SPBs) with a dramatic decrease in the lethal PCa patient urine samples. In ROC analysis, the prediction ability of the SPBs in lethal PCa and nonlethal PCa patients revealed the AUC value of 0.933, with 0.88 for sensitivity and 0.89 for specificity. Moreover, SPBs could enhance the accuracy of serum PSA in diagnosis, and the AUC value when combined with serum PSA level was up to 0.967. Extensively, SPBs also had the potentials to be the prediction markers for the PCa patients under clinical treatments including androgen deprivation therapy (ADT), abiraterone, enzalutamide and systemic chemotherapy. In conclusion, our results suggest that SPBs exhibit a potential to be clinical urinary biomarkers to distinguish lethal PCa from nonlethal PCa. Moreover, SPBs might be potentially used as a prognostic marker to monitor the treatment responses during the therapeutic courses and give PCa patients more appropriate therapies.