台灣鼻咽癌之環境與遺傳危險因子之流行病學研究：Epstein-Barr病毒、抽菸、家族傾向及基因多形性

Abstract

本論文將進行以下三個子研究探討EB病毒、抽菸、鼻咽癌家族史及DNA修補基因、細胞週期控制基因及凋亡基因多形性與鼻咽癌危險性之相關： 研究一 EB病毒及抽菸與鼻咽癌之長期追蹤研究 為了瞭解EB病毒與抽菸對於鼻咽癌之獨立作用，本研究於1984至1986年間，於台灣地區6個鄉鎮市區針對9622名男性居民，進行收案追蹤研究。採用Cox氏比例危害複迴歸模型來估計相對危險性及95%信賴區間。結果顯示，收案當時EB病毒抗體效價與鼻咽癌的危險性呈現劑量效應關係。在調整年齡與累積抽菸量的干擾作用後，抗EB病毒外鞘抗原IgA抗體（VCA IgA）低效價與高效價，發生鼻咽癌的相對危險性，分別為陰性者的9.5倍以及21.4倍，而抗EB病毒DNA分解酶抗體（DNase）低效價與高效價，其相對危險性分別為陰性者的1.6倍以及16.0倍，累積抽菸量超過30包年，其相對危險性則為累積抽菸量小於30包年的3.0倍。 研究二 鼻咽癌家族史與EB病毒和鼻咽癌之相關 本研究欲探討，在控制了EB病毒後，鼻咽癌多發性家族（家族內有2位以上鼻咽癌病人）成員是否有較高的機會罹患鼻咽癌。因此藉由社區族群為基礎之世代追蹤研究以及鼻咽癌多發性家族成員之追蹤研究，來比較罹患鼻咽癌的危險性。採用Cox氏比例危害複迴歸模型來估計相對危險性及95%信賴區間。結果顯示，在追蹤了1,019名多發性家族男性親屬以及9,622名社區男性居民後，多發性家族男性成員相對於社區男性居民罹患鼻咽癌的危險性為13.9倍，即使調整EB病毒與抽菸習慣後，相對危險性仍然有8.3倍。多發性家族成員，且抗EB病毒外鞘抗原IA抗體（VCA IgA）或DNA分解酶抗體（DNase）任一為陽性者，其相對危險性更高達37倍。 研究三 DNA修補、細胞週期控制及凋亡基因多形性與鼻咽癌之相關 本研究以病例對照研究法，分析334名鼻咽癌病例及283名經性別、年齡及居住地區配對之社區對照，探討17個DNA修補、細胞週期控制及凋亡基因，共31個基因多形性與鼻咽癌之相關。採用非條件邏輯迴歸模型來估計年齡、性別、種族及教育程度調整後之危險對比值與95%信賴區間。針對單一基因多形性以及半形分析，更進一步分析基因與環境之交互作用，及基因劑量效應。結果顯示，BRCA2 N372H（對比值為1.4）、BRCA2 IVS24+5557 C>A（對比值為1.5）、EXO1 T439M（對比值為0.5）、hMLH1 -93 A>G（對比值為0.6）以及CASP9 F136F（對比值為0.5）均與罹患鼻咽癌的危險性有顯著相關。帶有越多BRCA2、EXO1、hMLH1及CASP9等危險基因，得到鼻咽癌的風險則越高。本研究並未發現基因與環境有交互作用。

Three studies were conducted to investigate the associations of Epstein-Barr virus (EBV), cigarette smoking, family history of nasopharyngeal carcinoma (NPC), and polymorphisms of DNA repair and cell cycle control/apoptosis genes with the risk of developing NPC. Study 1. Epstein-Barr Virus and Cigarette Smoking on Nasopharyngeal Carcinoma: A Follow-up Study This study aimed to assess independent effects of Epstein-Barr virus (EBV) and cigarette smoking on nasopharyngeal carcinoma (NPC), which have never been assessed in long-term follow-up studies. A cohort of 9,622 men was enrolled from 1984 to 1986. Cox’s proportional hazard regression analysis was used to estimate multivariate-adjusted hazard ratio (HRadj) with its 95% confidence interval (CI). Increasing serum levels of anti-EBV VCA IgA and DNase were significantly associated with NPC risk in a dose-response relationship. The HRadj of developing NPC for low and high antibody levels compared with seronegatives was 9.5 and 21.4, respectively, for anti-EBV VCA IgA; and 1.6 and 16.0, respectively, for anti-EBV DNase. The HRadj was 3.0 for ≥30 pack-years of cumulative cigarette smoking compared with <30 pack-years as the referent. Study 2. Associations of Family History and Epstein-Barr Virus Infection Seromarkers with Risk of Nasopharyngeal Carcinoma This study examined the impact of family history (defined as 2+ NPC cases in a family) on NPC risk after adjustment for anti-EBV seromarkers. We evaluated the longitudinal risk of NPC among males from NPC multiplex families against that observed among a second group of males from a community-based cohort study in Taiwan. Cox’s proportional hazard regression analysis was used to estimate multivariate-adjusted hazard ratio (HRadj) with 95% confidence interval (CI). In total, 1,019 males from the NPC multiplex family cohort and 9,622 males from the community cohort were in the analyses. Compared with participants in the community cohort, the HRadj of developing NPC was 13.9 for participants in the multiplex family cohort. The increased NPC risk for members of NPC multiplex families remained significantly elevated after adjustment for anti-EBV seromarkers and cigarette smoking habit (HRadj: 8.3). The HRadj (95% CI) was 37.0 for participants who were anti-EBV-seropositive and reported a family history of NPC compared with EBV-seronegatives individuals without such a history. Study 3. Associations of Polymorphisms of DNA Repair Genes and Cell Cycle Control/Apoptosis Genes with Nasopharyngeal Carcinoma Risk We conducted a community-based case-control study to investigate the genotypes of 334 NPC cases and 283 healthy community controls matched by sex, age, and residence. This study focused on effects of 31 genetic polymorphisms of 17 DNA repair and cell cycle control/apoptosis genes on the development of nasopharyngeal carcinoma (NPC). We calculated odds ratios (ORs) and 95% confidence intervals (CIs) for the association of single nucleotide polymorphism (SNPs) and haplotype with NPC. Moreover, gene-environment interaction and gene dosage effect were also evaluated. The results showed a highly significant increased risk for NPC was associated with the BRCA2 N372H (OR = 1.4) and BRCA2 IVS24+5557 A>C (OR = 1.5). The EXO1 T439M, hMLH1 -93A>C, and CASP9 F136F had a 0.5-fold, 0.6-fold, and 0.5-fold reduced risk for NPC, respectively. In cumulative genetic risk analyses, individuals with four risk genotype (OR = 1.3), five (OR = 2.2), six (OR = 2.3) exhibited increased risks of NPC compared the number less than 3 risk genotypes as the reference, with a significant gene dosage effect. No significant interaction between gene and environment was found.