© The Author 2007. Published by Oxford University Press.
CORRESPONDENCE |
Re: Commonly Studied Single-Nucleotide Polymorphisms and Breast Cancer: Results From the Breast Cancer Association Consortium
Affiliations of authors: Department of Cancer Prevention and Control, Roswell Park Cancer Institute, Buffalo, NY (CBA, CCH); Lombardi Comprehensive Cancer Center, Georgetown University, Washington, DC (PGS); Department of Social and Preventive Medicine, University at Buffalo, Buffalo, NY (CBA, JLF)
Correspondence to: Christine B. Ambrosone, PhD, Department of Cancer Prevention and Control, Roswell Park Cancer Institute, Elm & Carlton Streets, Buffalo, NY 14263 (e-mail: christine.ambrosone{at}roswellpark.org).
In a recent pooled analysis of breast cancer data from up to 12 case–control studies from the Breast Cancer Association Consortium (1), five of 16 common single-nucleotide polymorphisms (SNPs) were associated with modest increases in risk. The authors interpreted their findings, with P values ranging from .06 to .0088, as failing to meet a level of statistical significance appropriate to genetic association studies. There were no associations for the other SNPs evaluated.
To us, these analyses and interpretation represent a growing divergence in perspective between investigators approaching disease from a statistical genetics viewpoint and those approaching it from a molecular epidemiologic viewpoint. Carcinogenesis is a dynamic process, usually the result of both exogenous and endogenous exposures, and we would not expect, for most pathways, to observe main effects for common genetic variants without consideration of relevant exposures. Moreover, because of the complex nature of carcinogenesis pathways, it is unlikely that a SNP in one single low-penetrance gene alone would be associated with an increase in cancer risk, without consideration of potential interactions with other polymorphic genes.
Genetic processes, and interactions with exposures, are far from simple. In addition to SNPs modifying risk associated with exposures, the impact of exposures on relationships between genetic variants and phenotype, through enzyme induction or inhibition, must also be considered. For example, effects of dietary exposures on relationships between catalase genotypes and enzyme activity have been demonstrated (2). Although genotype predicted phenotype overall, relationships varied by levels of fruit and vegetable consumption, with genotype/phenotype correspondence only among consumers of low levels of fruits and vegetables, suggesting feedback mechanisms. Similarly, in a study of glutathione peroxidase genotype, erythrocyte activity, and breast cancer risk (3), although genotype predicted glutathione peroxidase activity overall, relationships between genotype and activity were modified by alcohol consumption and smoking, with changes in activity according to levels of exposure varying by genotypes. These data illustrate the dynamic biologic systems in which the effects of SNPs on cancer risk are evaluated. Genotypes are not static variables, and the ultimate impact of SNPs on phenotype, and, more important, on cancer risk, will be modified by exogenous and endogenous exposures.
Accumulating evidence illustrates the importance of exposures for breast cancer risk. It is notable that cancer rates are rising in Japan with westernization and that migrants to the United States from low-risk countries have increased breast cancer risk with subsequent generations (4). Clearly, genes are not changing; rather, lifestyle factors, perhaps interacting with genetics, are likely the cause of these rising cancer rates. The consortia authors noted heterogeneity in results among the pooled studies and commented that it is "likely due to some unexplained artifact" and that "it seems unlikely that there are associations in one population and not others." We would argue that, when examining the effects of genotypes without consideration of other exposures, it is highly likely that results from different populations would in fact differ, if there are important differences in the relevant exposures in the populations. For example, results from assessment of the main effects of alcohol dehydrogenase 1C (ADHC1), which is involved in metabolism of alcohol, would likely vary widely if one population included a large proportion of heavy consumers of alcoholic beverages and the other did not.
In summary, examination of common single SNPs in the absence of data on exposures, and expectations to see main effects, may not be appropriate. We suggest that considerable caution be exercised in the interpretation of main effects of common polymorphisms on risk of cancer. Genome-wide association studies are now yielding results and offer promise for identification of genetic variants that may play a role in cancer risk. However, null results from such studies will only mean that the gene, in and of itself, is not associated with cancer risk and should not rule out further investigation of the effects of the genetic variants in modifying relationships between exposures and cancer risk.
REFERENCES
(1) The Breast Cancer Association Consortium. (2006) Commonly studied single-nucleotide polymorphisms and breast cancer: results from the Breast Cancer Association Consortium. J Natl Cancer Inst 98:1382–96.
(2) Ahn J, Nowell SA, McCann SE, Yu J, Carter L, Lang NP, et al. (2006) Associations between catalase phenotype and genotype: modification by epidemiologic factors. Cancer Epidemiol Biomarkers Prev 15:1217–22.
(3) Ravn-Haren G, Olsen A, Tjonneland A, Dragsted LO, Nexo BA, Wallin H, et al. (2006) Associations between GPX1 Pro198Leu polymorphism, erythrocyte GPX activity, alcohol consumption and breast cancer risk in a prospective cohort study. Carcinogenesis 27:820–5.
(4) Ziegler RG, Hoover RN, Pike MC, Hildesheim A, Nomura AM, West DW, et al. (1993) Migration patterns and breast cancer risk in Asian-American women. J Natl Cancer Inst 85:1819–27.
Response to this Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 488-489.
This article has been cited by other articles:
|
|
 |
|
  P. A. Thompson Counterpoint: Genetic Risk Feedback for Common Disease Time to Test the Waters Cancer Epidemiol. Biomarkers Prev., September 1, 2007; 16(9): 1727 - 1729. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||