作者：Jacquelyn K. Beals, PhD
出處：WebMD醫學新聞
November 7, 2008 — 有一篇新研究首度發表microRNAs(miRNAs)在食道癌敏感性方面的證據。本研究登載於11月的癌症預防研究期刊，選出26個miRNAs基因或者影響miRNA生源的41個單核苷酸多態性(SNPs)，研究它們與食道癌敏感性的關聯。
miRNAs的基因造成1%至5%的人類基因，678個人類miRNAs已經被確認且列入miRNA登記表中；與訊息RNA不同的是，mRNA的三個密碼子轉錄成蛋白質的氨基酸序列，miRNAs則是調節許多人類基因的表現。初級轉錄物(pri-miRNAs)在細胞核中裂開而產生中間的miRNA前驅物，之後移動到細胞質，在該處後續處理而獲得成熟的miRNAs。
之前已經指出miRNAs的表現與膀胱癌及食道癌等人類癌症的因果關係、類別、惡化或者預後有關；目前的案例控制研究檢視基因中的SNPs是否與miRNA密碼或者其與食道癌敏感性的過程有關。
這些案例病患(n = 346)是新診斷有食道癌的白人，控制病患(n = 346)則是性別、種族與年紀(差距5歲之內)和案例病患相仿的健康成人；從周邊血液獲得DNA。41個可能的功能性SNPs包括了10個位在8 pri-miRNA基因的SNPs，7 個位在7 pre-miRNA基因的SNPs，與miRNA過程有關的11個基因的24個SNPs；分析確認有7個SNPs與食道癌敏感性有顯著關聯。
資深作者、德州大學安德森癌症中心癌症預防中心的Betty B. Marcus Chair小組召集人、流行病學教授Xifeng Wu博士以電子郵件向Medscape Pathology & Lab Medicine表示，每個SNP有2個對偶基因 —一般對偶基因與變異的對偶基因— 造成了每個SNP的3種基因型：一般同合子、異合子、變異同合子。她解釋，如果變異基因型可降低風險，那麼一般基因型即會增加風險。我們的目標是確認發生癌症風險最高的人，之後針對他們進行篩檢、密集監控或者進行化學預防。
最強的風險關聯存在於mir423基因pre-miRNA部份的SNP (rs6505162) (P < .0001; 勝算比 [OR]為0.64; 95% 信心區間 [CI]為0.51 – 0.80)；就如同OR所指出的，此變項代表食道癌風險相當顯著降低。
當分析rs6505162與抽菸、年紀或者性別等流行病學因素之間的關係時，男性的保護效果最顯著(P < .0001;女性人數太少而無法進行適當分析)；64歲以下者的保護效果也很明顯(P < .001)，但是64歲以上者則沒有。未曾抽菸者(P = .001)與曾經抽菸者(一生抽菸少於100根; P = .013)也都有。
令人驚訝的是，位於mir196a-2基因的SNP (rs11614913)在未曾抽菸者的食道癌風險(P = .006; OR, 2.52; 95% CI, 1.30 – 4.88)，是曾經抽菸者(P = .166; OR, 1.41; 95% CI, 0.87 – 2.28)的將近2倍；除了rs6505162之外，SNPs 與食道癌增加風險顯著相關，OR 值範圍從1.58-11.27。
研究者略為增加到 7個與食道癌有關的SNPs，以納入另外 4個處於顯著邊緣的SNPs，與評估多風險相關SNPs的累積效果；相較於具有2個SNPs風險以下者，有3個風險SNPs者的風險加倍(OR, 2.00; 95% CI, 1.31 – 3.08)，4個SNPs以上者的風險達3倍(OR, 3.14; 95% CI, 2.03 –4.85)。如同文章內所載，這些發現指出，在路徑基礎之關聯研究上採取多基因方式的重要性，以確認可作為癌症風險預測因子的基因變項特徵。
Wu博士指出，癌症發生是一種多步驟、多基因的過程，單一基因/SNP的效果預計很小；單一候選基因方法一次考量1個基因/SNP，會錯過一些小效果，因為需要大型樣本，因此，缺乏足夠統計強度來偵測小關聯。即便偵測到，單一基因/SNP的小效果，其臨床與公共衛生相關性有限。不過，路徑基礎的方法或許擴大了個體基因/SNP的效果，促進預測強度到可以臨床運用的程度。
Medscape Pathology & Lab Medicine以電子郵件聯絡加州大學洛杉磯分校公共衛生學院、分子流行病學訓練與研究計畫主任、流行病學教授Zuo-Feng Zhang博士，請他提出看法。
Zhang博士表示，因為我們處於全基因圖關聯的時代，我們略過了單一基因/ SNP的方法；因為有超過35,000基因與超過800-1000萬個SNPs，在癌症發生過程中考量可能的基因與基因相互關係，單一基因/ SNP的方法很難釐清基因的效應。
Zhang 博士結論表示，相對的，全基因圖關聯的研究不只是假設，其重要性在於，以假設的進行路徑為基礎的系統性方式，找到可能的功能性SNPs。
Wu博士宣稱沒有相關資金上的往來。Zhang 博士沒有利益衝突或者相關財經關係，但是指出與Wu博士合作另外一個研究，不過也沒有財經關係。
Susceptibility to Esophageal Cancer Influenced by Genes Related to MicroRNAs
By Jacquelyn K. Beals, PhD
Medscape Medical News
November 7, 2008 — A new study has presented the first evidence of the roles played by microRNAs (miRNAs) in esophageal cancer susceptibility. A study appearing in the November issue of Cancer Prevention Research selected 41 single nucleotide polymorphisms (SNPs) in 26 genes that code for miRNAs or affect miRNA biogenesis and investigated their association with esophageal cancer susceptibility.
Genes that code for miRNAs make up 1% to 5% of human genes; 678 human miRNAs have already been identified and listed in the miRNA registry. Unlike messenger RNA, whose triplet codons are translated into the amino acid sequence of proteins, miRNAs regulate the expression of many human genes. The primary transcripts (pri-miRNAs) are cleaved in the cell nucleus to produce an intermediate miRNA precursor (pre-miRNA), which then moves to the cytoplasm, where further processing yields mature miRNAs.
Expression of miRNAs has been previously associated with the causation, type, progression, or prognosis of various human cancers, including bladder and esophageal cancers. The current case-control study tested whether SNPs in genes that code for miRNA or its processing affect esophageal cancer susceptibility.
Case patients (n = 346) were white patients newly diagnosed with esophageal cancer. Control patients (n = 346) were healthy individuals matched to case patients by sex, ethnicity, and age (within 5 years). DNA was extracted from peripheral blood. The 41 potentially functional SNPs included 10 SNPs located in 8 pri-miRNA genes, 7 SNPs in 7 pre-miRNA genes, and 24 SNPs in 11 genes involved in miRNA processing. The analysis identified 7 SNPs significantly associated with esophageal cancer susceptibility.
Senior author Xifeng Wu, MD, PhD, professor of epidemiology and Betty B. Marcus Chair in Cancer Prevention, M.D. Anderson Cancer Center, University of Texas, Houston, told Medscape Pathology and Lab Medicine via email: "There are 2 alleles at each SNP — common allele and variant allele — which result in 3 genotypes for each SNP: common homozygote, heterozygote, and variant homozygote. If the variant genotype is risk-reducing, then the common genotype would be risk-increasing," he explained. "Our purpose is to identify individuals with the highest risk of developing cancer, who may then be targeted for screening, intensive monitoring, or chemoprevention."
The strongest risk association was for a SNP (rs6505162) in the pre-miRNA portion of gene mir423 (P < .0001; odds ratio [OR], 0.64; 95% confidence interval [CI], 0.51 – 0.80). As indicated by the OR, this variant conveyed a significantly lower risk for esophageal cancer.
When rs6505162 was analyzed in terms of its interaction with epidemiological factors such as smoking, age, or sex, the protective effect was significant among men (P < .0001; there were too few women for adequate analysis). The protective effect was also significant for individuals aged 64 years or younger (P < .001) but not for those older than 64 years, and for never-smokers (P = .001) as well as ever-smokers (smoked more than 100 cigarettes in their lifetime; P = .013).
Surprisingly, a SNP (rs11614913) in gene mir196a-2 was associated with nearly twice the risk for esophageal cancer in never-smokers (P = .006; OR, 2.52; 95% CI, 1.30 – 4.88) as in ever-smokers (P = .166; OR, 1.41; 95% CI, 0.87 – 2.28). With the exception of rs6505162, the SNPs significantly associated with esophageal cancer carried increased risk, with OR values ranging from 1.58 to 11.27.
The investigators slightly expanded the pool of 7 SNPs associated with esophageal cancer to include 4 additional SNPs with borderline significance and also evaluated the cumulative effect of multiple risk-associated SNPs. Compared with individuals with 2 or fewer risk SNPs, 3 risk SNPs doubled the risk (OR, 2.00; 95% CI, 1.31 – 3.08), and 4 or more risk SNPs tripled the risk (OR, 3.14; 95% CI, 2.03 – 4.85). As the article states, "These findings highlighted the importance of taking a multi-genic approach in pathway-based association studies to identify signatures of genetic variations as predictors of cancer risk."
Dr. Wu added: "Cancer development is a multistep, multigenetic process, and the effect of single gene/SNP is expected to be very small. A candidate gene approach that considers 1 gene/SNP at a time may miss small effects due to the requirement of large sample size, and hence sufficient statistical power to detect small associations. Even if detected, the small effect of 1 gene/SNP has limited clinical or public health relevance." However, "a pathway-based approach...may amplify the effects of individual gene/SNP and enhance the predictive power to a level that may become clinically applicable."
Medscape Pathology and Lab Medicine contacted Zuo-Feng Zhang, MD, PhD, professor of epidemiology and director of the Molecular Epidemiology Training and Research Program, University of California–Los Angeles School of Public Health, who commented by email.
"We have passed [the] time for the 1 gene/1 SNP approaches because we are in the era of the whole-genome association," said Dr. Zhang. "Because there are over 35,000 genes and over 8 to 10 million SNPs, and considering possible gene-gene interactions in the cancer development process, it is very hard to figure out the genetic effect through the single gene/1 SNP approach."
In contrast, "the whole-genome association studies are...not hypothesis-driven. It would be of importance to have the hypothesis-driven pathway-based systematic approach for potentially functional SNPs," concluded Dr. Zhang.
Dr. Wu has disclosed no relevant financial relationships. Dr. Zhang has no conflicts of interest or relevant financial relationships, but noted that he has collaborated with Dr. Wu on another study without any financial relationship.
Cancer Prev Res. 2008;1(6):460–469.
[ 本帖最後由 goodcat1111 於 2008-11-18 10:30 編輯 ] |
|
