Search tips
Search criteria 


Logo of amjepidLink to Publisher's site
Am J Epidemiol. 2010 November 1; 172(9): 998–999.
Published online 2010 September 16. doi:  10.1093/aje/kwq282
PMCID: PMC2962256
Response to Invited Commentary

Miyaki et al. Respond to “Gene × Lifestyle Interactions”

We appreciate the thorough commentary by Drs. Franks and Nettleton (1). The major finding of our paper (2) was that interaction between the cyclin-dependent kinase 5 regulatory subunit-associated protein 1-like 1 (CDKAL1) gene and total energy intake (TEI) affected hemoglobin A1c levels in apparently healthy Japanese men. Given the limited space in which to respond to Franks and Nettleton, we will focus on a few of the more pertinent issues.

First, we are aware of the limitations of our study, including its low statistical power due to the small sample size, and we agree with Franks and Nettleton regarding the consequences of a low-powered study in general. These consequences can only be overcome by selecting a larger cohort. Thus, our paper is, at best, an exploratory analysis rather than a validation analysis of the effect of a well-studied CDKAL1 polymorphism.

Franks and Nettleton raised a question with regard to the accuracy of the food frequency questionnaire (FFQ) used for measuring TEI. We agree that assessing the validity of the dietary intake method is essential. The FFQ we used has been compared with 7-day dietary records and has been demonstrated to have a significant correlation with TEI (3). Thus, we believe this method allowed us to place participants into broad categories along a distribution of TEI (e.g., tertiles, as we used in the study). Body height was also taken into account as a means of reducing measurement bias in the calculation of TEI from FFQ (4). The possible measurement bias was noted as a limitation of the study in our original report (2).

Franks and Nettleton raised an issue regarding consistency, taking Hill's criteria (5) as an example and referring to 2 intervention studies (6, 7) and 1 cohort study (8) that were claimed to not support our data. One of the intervention studies (6), which Franks coauthored, involved a high-risk diabetic population, the majority of which received lifestyle intervention or medication. Significant differences exist in the characteristics of the populations, including mean body mass index, which was 34 in Franks’ group and 23 in our group. The cohort study (8), also coauthored by Franks, accounted for an interaction between the gene and physical activity, dichotomized into active and inactive states by means of a computer-assisted questionnaire. Whether or how TEI was measured in this group does not seem to be explained in the paper (8). Thus, the context and aims of these studies seem to be so remote from ours that we cannot infer any indication of consistency or inconsistency.

In addition, Franks and Nettleton questioned the biologic plausibility of our findings. CDKAL1 is involved in β-cell function, and polymorphisms within the gene have been repeatedly shown to affect blood glucose levels and insulin secretion, but not fasting glucose levels, after a glucose tolerance test (9). Therefore, we think it is reasonable to speculate on and test whether the CDKAL1 gene × TEI interaction correlates with blood glucose levels. Here, we would like to reiterate that although we observed a correlation with hemoglobin A1c, we did not observe a correlation with fasting glucose levels (2).

As for causality, Franks and Nettleton also suggested that the effect of the CDKAL1 variant might be to modify TEI but not blood glucose level. This assertion is logically true, and to ascertain the possibility one needs to stratify the population according to blood glucose level, genotype the participants, and survey TEI under an ad lib diet.

The commentary by Franks and Nettleton provided us with an opportunity to review our own study. We agree that future studies using larger cohorts are needed to validate our novel finding that the CDKAL1 gene × TEI interaction affects blood glucose level. Such studies may highlight other environmental or genetic interaction factors that are prerequisites for this effect.


Author affiliations: Department of Neurology, School of Medicine, Keio University, Tokyo, Japan (Koichi Miyaki, Htay Lwin, Yutaka Tomita, Haruhiko Hoshino, Norihiro Suzuki); and Department of Molecular Epidemiology, Medical Research Institute, Tokyo Medical and Dental University, Tokyo, Japan (Koichi Miyaki, Than Oo, Yixuan Song, Htay Lwin, Masaaki Muramatsu).

Conflict of interest: none declared.


1. Franks PW, Nettleton JA. Invited commentary: gene × lifestyle interactions and complex disease traits—inferring cause and effect from observational data, sine qua non. Am J Epidemiol. 2010;172(9):992–997. [PMC free article] [PubMed]
2. Miyaki K, Oo T, Song Y, et al. Association of a cyclin-dependent kinase 5 regulatory subunit-associated protein 1-like 1 (CDKAL1) polymorphism with elevated hemoglobin A1c levels and the prevalence of metabolic syndrome in Japanese men: interaction with dietary energy intake. Am J Epidemiol. 2010;172(9):985–991. [PubMed]
3. Takahashi K, Yoshimura Y, Kaimoto T, et al. Validation of a food frequency questionnaire based on food groups for estimating individual nutrient intake [in Japanese] Jpn J Nutr. 2001;59(5):221–232.
4. Jakes RW, Day NE, Luben R, et al. Adjusting for energy intake—what measure to use in nutritional epidemiological studies? Int J Epidemiol. 2004;33(6):1382–1386. [PubMed]
5. Hill AB. The environment and disease: association or causation? Proc R Soc Med. 1965;58:295–300. [PMC free article] [PubMed]
6. Moore AF, Jablonski KA, McAteer JB, et al. Extension of type 2 diabetes genome-wide association scan results in the Diabetes Prevention Program. Diabetes. 2008;57(9):2503–2510. [PMC free article] [PubMed]
7. Ruchat SM, Rankinen T, Weisnagel SJ, et al. Improvements in glucose homeostasis in response to regular exercise are influenced by the PPARG Pro12Ala variant: results from the HERITAGE Family Study. Diabetologia. 2010;53(4):679–689. [PMC free article] [PubMed]
8. Brito EC, Lyssenko V, Renström F, et al. Previously associated type 2 diabetes variants may interact with physical activity to modify the risk of impaired glucose regulation and type 2 diabetes: a study of 16,003 Swedish adults. Diabetes. 2009;58(6):1411–1418. [PMC free article] [PubMed]
9. Steinthorsdottir V, Thorleifsson G, Reynisdottir I, et al. A variant in CDKAL1 influences insulin response and risk of type 2 diabetes. Nat Genet. 2007;39(6):770–775. [PubMed]

Articles from American Journal of Epidemiology are provided here courtesy of Oxford University Press