AKR1C3於口腔鱗狀細胞癌中之表現

Abstract

人類醛固酮還原酶家族1成員C3 (Aldo-keto reductase family 1, member C3, AKR1C3)為醛酮還原酶(AKR)家族一員。醛酮還原酶(AKR)家族以NADPH為輔酶將來自食物和細胞內的脂肪類醛、酮和芳香類物質，還原成相對應的醇。AKR1C3蛋白分子量為36853Da，最早在human immature myeloid cell line被分離出來。AKR1C3在人的組織中表達位置，包括腎上腺、腦、腎、肝、肺、乳腺、胎盤、膽囊、小腸、十二指腸、結腸、脾、前列腺、和睾丸。最近研究指出，AKR1C3的表現與頭頸部癌症、食道癌、乳癌、非小細胞肺癌、肝癌、前列腺癌、子宮內膜癌的發生及發展有關。但並無文獻提到專門針對OSCC患者，其癌組織中AKR1C3蛋白值濃度，與OSCC患者預後之相關性。因此，本研究進一步檢測OSCC患者癌組織中AKR1C3蛋白值濃度是否和OSCC患者之腫瘤發展、復發及存活有關。 本研究利用口腔癌病患之組織病理切片進行免疫組織化學染色法(immunohistochemistry, IHC)，測量93例OSCC癌組織及30例正常口腔黏膜 (normal mucosa) 中AKR1C3蛋白表現量。AKR1C3蛋白表現量以染色記分(LSs，定義為染色強度與染色指標的乘積)定義，染色記分越高表示OSCC組織中AKR1C3蛋白表現量愈多。ANOVA、卡方檢定(Chi-square test)、Kaplan-Meier 存活率方法及Cox proportional hazard regression model，來分析癌組織中AKR1C3蛋白表現量與OSCC患者臨床病理參數及存活率之相關性，並試圖尋找預測存活時間的獨立預後因子。 本研究發現，AKR1C3蛋白的染色記分(labeling score, LS)較高和患者有較大腫瘤 (P = 0.040)、有較高臨床分期 (P = 0.009) 、較差的分化 (P <0.001)、有淋巴血管的侵犯(P = 0.019)、有飲食酒精的習慣 (P <0.001)及有嚼食檳榔之口腔習慣 (P = 0.014) ，有統計學上有意義之相關。Kaplan-Meier存活分析發現， AKR1C3 labeling score> 58比AKR1C3 labeling score ≤ 58之OSCC患者，有較差的5年總存活率 (log-rank test, P = 0.011)。以Cox regression model 進行多變數分析發現，有飲食酒精的習慣(P =0.018)、AKR1C3 labeling score> 58 (P = 0.004)為影響OSCC患者存活時間的獨立預測因子。 本研究發現，OSCC患者組織中AKR1C3蛋白質表現量可以預測OSCC之腫瘤大小、臨床分期及其5年總存活率；同時，組織中AKR1B10蛋白質表現量與OSCC患者嚼食檳榔及飲酒之口腔習慣呈現正相關。因此OSCC患者組織中AKR1C3蛋白質表現量可以當作OSCC患者預後的指標。

Aldo-keto reductase family 1 member C3 (AKR1C3) is one of the aldo-keto reductase superfamily. AKR1C3 can efficiently reduce aliphatic and aromatic aldehydes, but less active on hexoses. AKR1C3 was originally identified by Nagase from human immature myeloid cell lines in 1995. AKR1C3 is a cytosolic reductase with 36853 Da of a molecular weight. In normal human tissues, AKR1B10 is primarily expressed in epithelial tissues of brain, lung, breast, small intestine, duodenum, liver, gallbladder, spleen, adrenal gland, kidney, colon, prostate, testis, and placenta. Elevated expression of AKR1C3 is found in a wide variety of human tumors, such as head-and-neck, esophageal, non-small cell lung, breast, liver, prostate, and endometrial cancers. However, no specific report has been reported on AKR1C3 protein expression in OSCCs. In this study, we investigated whether AKR1C3 expression in OSCCs can be correlated with clinicopathological parameters, and used as a prognostic biomarker. In this study, we used immunohistochemical staining (IHC) to detect the AKR1C3 protein levels in 93 OSCCs and 30 normal oral mucosa (NOM). The leveling score (LS) of AKR1C3 protein expression was evaluated semi-quantitatively based on the proportion of positive-staining cells to the total number of counted cancer or epithelial cells (labeling index) and the staining intensity. The correlation between AKR1C3 LSs and clinicopathological parameters of OSCC patients was analyzed by Student’s t-test, ANOVA, Kaplan-Meier and Multivariate Cox regression analyses. In this study, we found that the AKR1C3 labeling score (LS) for OSCCs (69±45) was significantly higher than that for NOM (10±14, p < 0.001). High expression of AKR1C3 was significantly correlated with the large tumor size (p = 0.040), advanced TNM stage (p = 0.009), poorly differentiated (p <0.001), lymphovascular invasion (p = 0.019), alcohol consumption ((p =0.001), and areca quid chewing habit (p = 0.014). A multivariate analysis identified that higher AKR1C3 LS > 58 (p = 0.004) was significantly correlated with 5-year overall survival. This study found that AKR1C3 protein level in OSCC tissues could be used to predict the tumor size, clinical stage and the prognosis of OSCC patients. Therefore, we conclude that AKR1C3 protein level in OSCC tissues may be a valuable biomarker and molecular therapeutic target for OSCC patients in Taiwan.