Environmental factors known to affect breast cancer risk do not appear to interact with common gene variants that convey a small increased risk of breast cancer, according to a study published online June 2 in the Lancet.

Breast cancer risk is related to both environmental factors and genetic factors, but little is known about how the two types of factors interact. This study examined 10 environmental factors – all of them reproductive, behavioral, or anthropometric factors known to raise breast cancer risk – to assess whether they influenced the small risks conveyed by 12 common genetic variants.

”And the answer is that they do not,” said Dr. Ruth C. Travis of the cancer epidemiology unit, University of Oxford (England), and her associates. This lack of interaction means that environmental and genetic factors increase breast cancer risk independently, the investigators noted.

The study assessed 12 single nucleotide polymorphisms (SNPs)that are carried by many women but convey only a small risk of breast cancer; it did not include the much rarer susceptibility genes such as BRCA1 and BRCA2 that convey a high risk of breast cancer but are carried by relatively few women.

The investigators used data collected in the Million Women Study to examine this relationship. In an editorial comment accompanying this report, Dr. Steven A. Narod of Women’s College Research Institute, Toronto, explained, “The premise is that, given a large enough sample, if the impact of an environmental factor is conditional on a woman's genotype, the statistical term for the gene-environment interaction should be formally significant.”

Findings from this type of analysis could conceivably improve clinicians’ ability to predict breast cancer risk, provide clues about as yet undiscovered risk factors for the disease, or lead to identification of new candidate genes, Dr. Narod said (Lancet 2010 June 2 [doi:10.1016/S0140-6736(10)60876-8]).

The Million Women Study collected data on 1.3 million middle-aged women in the United Kingdom in 1996-2001 and followed them to track the development of cancer. A subgroup of these subjects participated in a genetic susceptibility study, donating blood samples for genotyping.

In their analysis, Dr. Travis and her colleagues included data on 7,610 of these women who developed incident breast cancer and 10,196 controls who did not. The mean age at diagnosis was 60 years.

The investigators found no interactions between the 12 common SNPs in the DNA of these subjects and any of 10 environmental risk factors (age at menarche, parity, age at first birth, breastfeeding status, menopausal status, age at menopause, use of hormone therapy, body mass index, height, and alcohol consumption).

In the 120 separate statistical comparisons for gene-environment interaction, only 4 yielded a result approaching statistical significance. The strongest interaction was between the CASP8 gene’s rs1045485 SNP and alcohol consumption, with relative risk for breast cancer higher in women who reported taking one or more alcohol drinks per day.

Similarly, the interactions between the TNRC9 gene’s SNP rs3803662 and age at menopause, the 2q35 gene’s SNP rs13387042 and age at menarche, and the 5q gene’s SNP rs-30099 and use of hormone therapy were of borderline significance.. “However, in view of the large number of tests done, these could well be chance findings,” Dr. Travis and her colleagues said (Lancet 2010 June 2 [doi:10.1016/s0140-6736(10)60636-8]).

“Our findings are largely hypothesis generating,” the investigators said.

“Tens of thousands of cases will be needed to assess reliably a comprehensive range of biologically plausible gene-environment interactions,” they added.

Dr. Narod questioned whether this approach “has been oversold,” however, “if attention is limited to the gene variants that have come out of the genome-wide association studies so far and to the known risk factors for breast cancer, we are unlikely to make much progress. The challenge of breast cancer prevention remains and new approaches are needed,” he concluded.

This study was funded by Cancer Research UK and the U.K. Medical Research Council. No financial conflicts of interest by the investigators were reported.

Copyright (c) 2009 Elsevier Global Medical News. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.

《柳叶刀》(The Lancet)6月2日在线发表的一项研究表明，已知会影响乳腺癌风险的环境因素与可能使乳腺癌风险小幅增加的常见基因变异体之间似乎不存在相互作用。

 

虽然乳腺癌风险与环境因素和遗传因素都相关，但少有数据提示这两类因素之间是如何相互作用的。该研究共调查了10种环境因素，评价其是否会对12种常见基因变异体所引起的乳腺癌风险小幅增加产生影响，所评价的环境因素均为已知会导致乳腺癌风险增加的生殖、行为或人体测量学因素。

 

英国牛津大学癌症流行病学中心的Ruth C. Travis博士及其同事说：“研究所得出的答案是，这两者之间不存在相互作用。”研究者指出，不存在相互作用意味着环境因素和遗传因素分别独立地使乳腺癌风险增加。

 

该研究共评价了12种单核苷酸多态性(SNP)，许多妇女都携带有这些SNP，但它们只会导致乳腺癌风险小幅增加；研究并没有纳入那些非常罕见的易感基因，如BRCA1和BRCA2，这些基因会导致乳腺癌风险明显增加，但只有较少的妇女携带。

 

研究者是基于百万妇女研究所收集到的数据来评估这两者之间的相关性。加拿大多伦多女子学院研究所的Steven A. Narod博士在随刊编者按中解释道：“前提是基于一个足够大的样本，如果环境因素对妇女基因型的影响是条件性的，那么基因-环境相互作用的统计学术语应该是有正式的统计学意义。”

 

Narod博士说，这类分析的结果应该有助于提高临床医生预测乳腺癌风险的能力，为该病至今尚未发现的危险因素提供线索，或有助于发现新的候选基因(Lancet 2010 June 2 [doi:10.1016/S0140-6736(10)60876-8])。

 

百万妇女研究收集了1996~2001年英国130万例中年妇女的数据，并对癌症的发病情况进行了追踪。其中一个亚组参与了一项遗传易感性研究，捐献了血液样本以供基因分型分析。

 

Travis博士及其同事将其中7,610例出现了偶发性乳腺癌以及10,196例没有出现乳腺癌的对照者的数据纳入了分析。确诊时的平均年龄为60岁。

 

研究者发现，这些受试者的DNA中的12种常见SNP与10种环境危险因素(月经初潮年龄、产次、初产年龄、哺乳状态、绝经状态、绝经年龄、激素疗法的使用、体重指数、身高以及酒精摄入量)之间均不存在相互作用。

 

在这120项基因-环境相互作用的统计学比较中，仅4项得出的结果接近统计学显著性。CASP8基因的rs1045485 SNP与酒精摄入量之间的相互作用最强，每天饮用一瓶或一瓶以上酒精饮料的妇女罹患乳腺癌的相对风险较高。

 

同样，TNRC9基因的SNP rs3803662与绝经年龄、2q35基因的SNP rs13387042与月经初潮年龄、5q 基因的SNP rs-30099与激素疗法的使用之间的相互作用具有临界统计学意义。Travis博士及其同事说：“但考虑到所开展的统计学检验数量很大，共进行了120项分析，因此这4项的结果很可能只是偶然发现。” (Lancet 2010 June 2 [doi:10.1016/s0140-6736(10)60636-8])。

 

研究者说：“我们的研究结果在很大程度上属于推断假设。”

 

研究者补充道：“还需要成千上万的病例来对有生物学可能的基因-环境相互作用进行全面且可靠的评价。”

 

Narod博士质疑这种方法是否“已经过时”，不过，“如果我们仍然只关注到目前为止全基因组关联研究所发现的基因变异体以及已知的乳腺癌危险因素，那么我们很难取得重大进展。关于乳腺癌预防的挑战依然存在，需要开创出新的研究途径，”他总结道。

 

该研究受英国癌症研究院和英国医学研究理事会资助。研究者声明无相关经济利益冲突。

 

爱思唯尔  版权所有