| 摘要: | 背景：許多國家甲狀腺癌發生率逐年上升，尤以韓國為最，臺灣也是其一，已有許多研究探討了潛在的危險因子，但結果仍有爭議。在臺灣，探討甲狀腺癌危險因子的研究皆使用全民健康保險資料庫，該資料庫缺乏重要的人口學相關資訊和環境因素，且大多分析不同疾病與甲狀腺癌之關係。故此研究欲使用臺灣人體生物資料庫，提供較全面的潛在危險因子，以探討臺灣人群的甲狀腺癌危險因子。方法：本研究為橫斷面研究，使用的臺灣人體生物資料庫成立於2012年，為以社區族群為基礎的長期追蹤之大型世代研究資料庫，包含超過20萬名20歲及以上的本國籍參與者。初始納入所有參與者共189,138名，排除了除甲狀腺癌外有任何癌症病史共1,782名，再排除分析危險因子有遺漏值者共65,572名，最終共121,784名納入分析。以單變項羅吉斯迴歸來估計與各種危險因子相關的勝算比（odds ratio，OR），並以多變項羅吉斯迴歸校正干擾因子並建立風險預測模型。利用自助抽樣法進行內部驗證並繪製列線圖。根據2013年臺灣人口的性別及年齡別分佈比例，我們以頻率配對方式選取病例組人數20倍的人數為對照組進行敏感度分析，也進行女性的次族群分析。結果：最終分析之甲狀腺癌病例為148名，對照組為121,636名。多變項羅吉斯回歸分析中顯示甲狀腺癌的顯著危險因子包括女性（OR=2.645，95%CI=1.741-4.018）、年齡（OR=1.023，95%CI=1.010-1.030）和一等親甲狀腺癌家族史（OR=15.246，95%CI=8.570-27.123）。特別的是B型肝炎病毒（HBV）感染也會增加甲狀腺癌的危險（OR=1.793，95%CI=1.138-2.826）。在女性次族群的多變項羅吉斯回歸分析中顯示除了年齡、一等親甲狀腺癌家族史及HBV感染之外，患有子宮內膜異位症的女性罹患甲狀腺癌的危險也會增加（OR=1.929，95%CI=1.231-3.023）。而於配對的女性次族群分析則發現除子宮內膜異位外，患有子宮肌瘤的女性風險也會增加（OR=1.538，95%CI=1.039-2.277）。結論：除了一等親家族史為最明確的危險因子之外，此研究還顯示HBV感染、子宮內膜異位症及子宮肌瘤是甲狀腺癌的潛在危險因子。因資料庫之限制，進一步需要世代研究以及較多的甲狀腺癌病例來驗證此研究結果。
Background: The incidence of thyroid cancer is increasing year by year in many countries, especially in South Korea, and as well as in Taiwan. Numerous studies have explored potential risk factors, though results remain controversial. In Taiwan, studies investigating risk factors for thyroid cancer all use the National Health Insurance database, which lacks crucial demographic information and environmental factors, and mostly analyzes the relationship between different diseases and thyroid cancer. Therefore, this study intends to use the Taiwan Biobank to provide a more comprehensive set of potential risk factors to identify risk factors for thyroid cancer in the Taiwanese population. Methods: This study is a cross-sectional study using the Taiwan Biobank, which was established in 2012. It is a large-scale cohort database with long-term follow-up based on community population, encompassing over 200,000 participants aged 20 and above. Initially, 189,138 participants were included. A total of 1,782 participants with a history of any cancer other than thyroid cancer were excluded. Additionally, 65,572 participants with missing values in risk factors were excluded from the analysis. Finally, a total of 121,784 participants were included in the analysis. Univariate logistic regression was used to estimate the odds ratio (OR) associated with various risk factors, and multivariate logistic regression was used to adjust confounders and establish a risk prediction model. The bootstrap method was used for internal validation and a nomogram was produced. Based on the gender and age-specific distribution of Taiwan's population in 2013, we used a frequency matching method to select 20 times the number of cases in the control group for sensitivity analysis, and also conducted a subgroup analysis of women. Result: The final analysis included 148 thyroid cancer cases and 121,636 controls. Multivariate logistic regression analysis showed that significant risk factors for thyroid cancer include female gender (OR=2.645, 95%CI=1.741-4.018), age (OR=1.023, 95%CI=1.010-1.030) and first-degree relatives family history of thyroid cancer (OR=15.246, 95%CI=8.570-27.123). In particular, hepatitis B virus (HBV) infection also increases the risk of thyroid cancer (OR=1.793, 95%CI=1.138-2.826). Multivariable logistic regression analysis of female subgroup showed that in addition to age, family history of thyroid cancer in first-degree relatives, and HBV infection, women with endometriosis were also at increased risk of thyroid cancer (OR=1.929, 95%CI=1.231-3.023). Analysis of matched female subgroup found that in addition to endometriosis, women with uterine fibroids were also at increased risk (OR=1.538, 95%CI=1.039-2.277). Conclusion: In addition to first-degree relative family history being the well-established risk factor, this study also suggests that HBV infection, endometriosis and uterine fibroids are potential risk factors for thyroid cancer. Due to limitations of the database, further longitudinal studies and more thyroid cancer patients are needed to validate our findings. |
