APC、K-ras、p53遺傳基因及環境因子與大腸直腸癌之相關

Abstract

前言 以基因標記結合環境危險因子作為大腸直腸癌社區性篩檢必須要有健康風險評估的預測模型。 目的 建立以社區為主的病例對照研究，並以二階段抽樣設計量化三個主要突變基因及環境因子，對於大腸直腸癌、小型腺腫及大型腺腫發生的危險性。同時評估基因與環境因子之間的交互作用。 方法 資料來源為台灣基隆地區社區篩檢計畫，自1999年至2004年間使用糞便潛血反應檢查作為篩檢工具，進行社區性大腸直腸癌篩檢。共有53,949位民眾參加篩檢計畫，經由篩檢發現97位大腸直腸癌及589位腺腫病患。2001年共有20,951位民眾參加篩檢，隨機挑選20位大腸直腸癌、83位腺腫病患以及100位控制組作為分析糞便DNA三種主要基因APC、K-ras及p53的突變。藉由問卷收集相關環境危險因子的資訊。利用邏輯式迴歸模型估計調整後之odds ratio及95%信賴區間，由主要研究族群的資訊與二階段抽樣方法調整odds ratio，藉由量化模式產生收受者操作特徵(Receiver operating characteristics, ROC)曲線。 結果 在環境因子方面，與大腸直腸癌發展有相關的為飲酒習慣 (aOR=1.40, 95% CI: 1.14-1.72)、運動習慣(aOR=0.71, 95% CI: 0.62-0.82)、家族大腸直腸癌病史(aOR=2.09, 95% CI: 1.21-3.60)、肉食攝取量(aOR=1.11, 95% CI: 0.87-1.41)、魚類海鮮攝取量(aOR=1.39, 95% CI: 1.09-1.76)、奶類製品攝取量(aOR=1.10, 95% CI: 0.91-1.34)、蔬菜攝取量(aOR=1.11, 95% CI: 1.01-1.21)與水果攝取量(aOR=0.91, 95% CI: 0.85-0.97)。研究對象若APC基因有發生突變現象，發生大腸直腸癌的危險性為APC基因正常的20倍(95%CI: 6.00-69.65)。相對的K-ras基因突變odds ratios為56.89 (95%CI: 7.45-434.51)，p53基因突變odds ratios為62.61 (95%CI: 14.39-269.06)。而三種基因中任何一種有突變發生大腸直腸癌的危險性是三種基因均為正常的94倍。只有水果攝取量與APC基因之間有顯著意義。本研究利用上述三種基因突變的資料結合環境因子，如水果攝取及飲酒習慣等，所建構出來的預測模式對於預測大腸腺腫的發生具有極佳的鑑別力。 結論 本研究利用族群資料量化環境因子與基因突變於大腸直腸癌發展的影響。肉類與酒類過量攝取或是較少攝取水果、缺乏運動等都會增加大腸直腸癌發展的危險性。APC、K-ras及p53為大腸直腸癌發展很強的預測因子。結合環境及基因兩方的資訊，可用作大腸直腸癌社區性篩檢的篩檢工具。

Introduction Predictive model for health risk assessment may be required before using genetic markers in combination with environmental risk factors as screening tool for population-based screening for CRC. Aim We conducted a population-based case-control study with two-stage sampling design to quantify the effect of three genetic major genes and environmental factors on occurrence of CRC and small to large adenomatous polyps. The interaction between gene mutations and environmental factors is also assessed. Methods Data used were derived from a community-based screening program in Keelung, Taiwan within which a population-based screening for colorectal neoplasm with fecal occult blood test (FOBT) was conducted between 1999 and 2004. A total of 53,949 participants were enrolled with the ascertainment of 97 colorectal cancer and 589 adenomatous polyps. Of 20,951 participants in 2001, we randomly selected 20 cancers, 83 adenomatous polyps, and 100 controls for stool DNA analysis on three major mutation genes, APC, K-ras, and p53. Information on environmental risk factors was collected from questionnaire. Logistic regression model was used to estimate adjusted odds ratios and their 95% confidence interval with two-stage method for adjusting odds ratio using information from main study. Receiver operating characteristics (ROC) curve was plotted. Results For environmental factors, variables that were related to the risk for developing colorectal neoplasm included alcohol drinking (aOR=1.40, 95% CI: 1.14-1.72), exercise frequency (aOR=0.71, 95% CI: 0.62-0.82), family history of CRC (aOR=2.09, 95% CI: 1.21-3.60), intake of meat (aOR=1.11, 95% CI: 0.87-1.41), intake of seafood (aOR=1.39, 95% CI: 1.09-1.76), intake of milk (aOR=1.10, 95% CI: 0.91-1.34), intake of vegetables (aOR=1.11, 95% CI: 1.01-1.21), and intake of fruits (aOR=0.91, 95% CI: 0.85-0.97). Subject carried with APC gene mutation had 20-fold risk (95%CI: 6.00-69.65) for the development of colorectal neoplasm compared with those in the absence of APC gene mutation. The corresponding adjusted odds ratios were 56.89 (95%CI: 7.45-434.51) for K-ras gene mutation and 62.61 (95%CI: 14.39-269.06) for p53 gene mutation. Those who were carried with either three of gene mutations had 94-fold risk for colorectal neoplasm compares with those in the absence of these gene mutations. Only the interaction between intake of fruit and APC gene was statistically significant. Information on three mutation genes in combination with environmental factors such as intake of fruit and alcohol drinking has been demonstrated to have a good discriminative ability to predict colorectal neoplasm. Conclusions The impact of environmental factors and three major genes on the risk for developing colorectal neoplasm have been demonstrated. High consumption of meat and alcohol, less intake of fruit, and physical inactivity may increase the risk for developing colorectal neoplasm. APC, K-ras, and P53 are strong predictor for occurrence of colorectal neoplasm. Information from both finding can be used as a screening tool for population-based screening for colorectal neoplasm.