Figures
Abstract
Objective
To evaluate whether the genetic susceptibility of T2D was associated with overall survival (OS) and disease-free survival (DFS) outcomes for breast cancer (BC).
Methods
Included in the study were 6346 BC patients who participated in three population-based epidemiological studies of BC and were genotyped with either GWAS or Exome-chip. We constructed a genetic risk score (GRS) for diabetes using risk variants identified from the GWAS catalog (http://genome.gov/gwastudies) that were associated with T2D risk at a minimum significance level of P ≤ 5.0E-8 among Asian population and evaluated its associations with BC outcomes with Cox proportional hazards models.
Results
During a median follow-up of 8.08 years (range, 0.01–16.95 years), 1208 deaths were documented in 6346 BC patients. Overall, the diabetes GRS was not associated with OS and DFS. Analyses stratified by estrogen receptor status (ER) showed that the diabetes GRS was inversely associated with OS among women with ER- but not in women with ER+ breast cancer; the multivariable adjusted HR was 1.38 (95% CI: 1.05–1.82) when comparing the highest to the lowest GRS quartiles. The association of diabetes GRS with OS varied by diabetes status (P for interaction <0.01). In women with history of diabetes, higher diabetes GRS was significantly associated with worse OS, with HR of 2.22 (95% CI: 1.28–3.88) for the highest vs. lowest quartile, particularly among women with an ER- breast cancer, with corresponding HR being 4.59 (95% CI: 1.04–20.28). No significant association between the diabetes GRS and OS was observed across different BMI and PR groups.
Conclusions
Our study suggested that genetic susceptibility of T2D was positively associated with total mortality among women with ER- breast cancer, particularly among subjects with a history of diabetes. Additional studies are warranted to verify the associations and elucidate the underlying biological mechanism.
Citation: Bao P-P, Zhao Z-G, Gao Y-T, Zheng Y, Zhang B, Cai H, et al. (2015) Association of Type 2 Diabetes Genetic Variants with Breast Cancer Survival among Chinese Women. PLoS ONE 10(2): e0117419. https://doi.org/10.1371/journal.pone.0117419
Academic Editor: Xifeng Wu, MD Anderson Cancer Center, UNITED STATES
Received: August 15, 2014; Accepted: December 23, 2014; Published: February 13, 2015
Copyright: © 2015 Bao et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and its Supporting Information files.
Introduction
Breast cancer is the most common cancer diagnosis in women in China and worldwide[1]. Survival after breast cancer diagnosis is influenced by tumor characteristics such as disease stage, tumor grades, hormonal status, and treatment. Recent epidemiological studies, including our own study, the Shanghai Breast Cancer Survival Study (SBCSS) [2], have shown that comorbidities, such as diabetes and hypertension, also influence breast cancer outcomes [3, 4, 5, 6]. It has been reported that 16% to 20% of breast cancer patients had type 2 diabetes (T2D) at time of cancer diagnosis [7, 8] and this comorbidity was inversely associated with breast cancer survival [9, 10, 11, 12, 13]. Studies also suggested that several diabetes-related conditions including insulin resistance, hyperinsulinemia, and chronic inflammation, being associated with the breast cancer outcomes [9, 10, 11, 12, 13, 14, 15]. However, few studies [11, 13] have evaluated the association of diabetes on outcomes of breast cancer by estrogen receptor status, the most important prognostic factor. In addition, studies have shown that under-diagnosis of diabetes is common, ranging from 27% in the US to 60% in China [16, 17]. Furthermore, it has been recently recognized that diabetes treatments, particularly the use of metformin, may mitigate the negative influence of diabetes on mortality, and may even reduce the risk of recurrence [18, 19], making it difficult to interpret the diabetes and breast cancer outcomes associations.
Diabetes has a strong genetic basis. To date, genome-wide association studies (GWAS) have identified multiple genetic variants for T2D [20, 21]. Given the strong relationship between T2D and breast cancer survival, we hypothesized that genetic susceptibility of T2D would also be associated with breast cancer outcomes and evaluated this hypothesis in a large scale study including 6346 breast cancer patients participated in the Shanghai Breast Cancer Study (SBCS), the SBCSS or the Shanghai Women’s Health Study (SWHS).
Materials and Methods
Ethics statement
The study protocols were approved by the Institutional Review Boards of participating institutes, i.e., the Vanderbilt University School of Medicine, Vanderbilt University, Nashville, Tennessee, United States; the Shanghai Cancer Institute, Shanghai, China; and Shanghai Municipal Center for Disease Prevention & Control, Shanghai, China. All participants provided written, informed consent.
Study population
Participants for the current study were breast cancer cases from the SBCS, the SBCSS, and the SWHS [22, 23]. The details of the study design and data collection procedures have been described previously [22, 23, 24, 25]. Briefly, the SBCS is a population-based two-phase (SBCS-I and SBCS-II) case-control study that recruited 3,448 patients between August 1996 and March 1998 for SBCS-I and again between April 2002 and February 2005 for SBCS-II [24], of which 90.6% provided a blood or exfoliated buccal cell sample. The SBCSS is a population-based prospective cohort study and recruited 5042 breast cancer patients between March 2002 and April 2006, of which 98% provided an exfoliated buccal cell sample [25]. The SWHS is a population-based prospective cohort study initiated in March 1997 and completed in May 2000, which focus on the cancer and other chronic disease [26]. All participants were followed for survival status and recurrence through a combination of in-person surveys and record linkages with the Shanghai Vital Statistics Registry. Because of the overlapping recruitment time, 1469 women participated in both the SBCS and SBCSS. After taking these overlaps into consideration and excluding patients without follow-up data, a total of 6346 participants remained in the present study.
Cancer diagnoses, cancer characteristics and treatment information were obtained by medical records review. All anthropometric measurements, diabetes history, and other information were collected by in-person interviews using validated questionnaires. All study interviewers were retired, trained medical professionals.
Genotyping and candidate SNP selection
Genotyping methods and quality control procedures have been previously described [23, 24]. Briefly, DNA was analyzed by either of two genotyping platforms. Two thousands nine hundred and one breast cancer cases were genotyped using the Affymetrix Genome-wide Human SNP Array 6.0 (Abbr. GWAS). A custom-designed Illumina Human Exome-12v1-A BeaChip (Abbr. Exome-chip) was used to genotype the other 3445 breast cancer cases. All GWAS identified variants as reported in the US National Human Genome Resource Institute (NHGRI) list by August 16, 2011, were included in this chip. Samples and variants that failed the quality control were excluded.
Genetic variants associated with T2D were searched in the GWAS catalog [27] (having disease trait “Type 2 diabetes” or “Type 2 diabetes and other traits” and p value <5×10-8) in Asian-ancestry population, and a total of 44 SNPs were identified. Data on 39 SNPs representing 32 independent loci (linkage disequilibrium, LD r2<0.3) were available for GWAS data set; One SNP from each of these 32 independent loci was selected to construct the genetic risk score (GRS) as described below. For subjects who were genotyped by the Exome-chip, 16 diabetes related SNPs were directly genotyped. For another 6 diabetes related SNPs, we were able to identify proxy SNPs (with a minimal r2 of 0.8 with the index SNP) and then used the proxy-index haplotype (use hapmap r23 data for JPT and CHB) to determine the corresponding risk allele for the proxy SNPs. Therefore, among subjects with Exome-chip data, genetic risk score (GRS) was constructed using the 22 SNPs (16 index and 6 proxy SNPs); of which, 20 SNPs overlapped with the 32 SNPs included in estimation of GRS in the subjects with GWAS data.
Statistical analysis
To measure the cumulative effect of the multiple diabetes genetic variants, a weighted GRS was calculated using the odds ratio of the published T2D loci to weight the T2D-related risk alleles at each locus and then summing across all T2D loci. The weighted GRS was created as or
, where SNPi is the number of the risk alleles one subject carries in a SNP in data from GWAS or Exome-chip, respectively, and the Wi is the odds ratio with T2D published previously. The weighted GRSs were categorized into quartiles separately for subjects from GWAS data or Exome-chip data based on its own distribution.
Outcomes of the study were total mortality (for the overall survival analysis) and breast cancer recurrence or breast cancer-related death (for the disease-free survival analysis). Cox proportional hazards regression models were applied to derive hazard ratios (HRs) and 95% confidence intervals (95% CI) with adjustments of age at diagnosis (treated as continuous using a restricted cubic spline with 5 knots), and disease stage. Stratified analyses by ER status (ER+, ER-), PR status (PR+, PR-), BMI (2 levels: <25, >=25), and diabetes (yes, no, unknown) were conducted to evaluate the consistency of the effects. The homogeneity of the associations across the strata was evaluated using likelihood ratio test. The associations between the GRS quartiles and breast cancer outcomes were evaluated separately by each genotype platform (Table B and Table C in S1 File), and then combined using the inverse-variance meta-analysis with fixed-effect or random-effect models as appropriate. Genotype platform specific results and combined results were reported. A sensitivity analysis by excluding the participants with extreme follow-up time (shortest 1% and longest 1%) was also performed and reported (Table D in S1 File). The association of each SNP with breast cancer survival was also investigated using similar approaches described above. The results for individual SNP and breast cancer survival analysis were shown in Table A in S1 File.
All statistical tests were based on two-sided probabilities with significance determined by P-value <0.05. Statistical analyses were carried out using R (http://www.R-project.org/).
Results
After a median follow-up of 8.08 years (range, 0.01–16.95 years), 1208 total all-cause deaths and 1160 breast cancer recurrences or deaths were documented. Subjects from GWAS data and Exome-chip data had similar 5-year overall survival rates, 88.23% and 88.13% respectively. Older age at diagnosis, advanced disease stage, negative ER or PR status, high BMI, not receiving radical mastectomy, receiving radiotherapy, and not receiving tamoxifen treatment were associated with worse overall survival and disease free survival (Table 1).
The diabetes GRS was not associated with ER/PR status, TNM stage, age at diagnosis but was associated with diabetes history at time of cancer diagnosis (P <0.01) (Table 2). As shown in Table 3, overall, the diabetes GRS was not associated with OS and DFS. Analyses stratified by ER status showed that the diabetes GRS was inversely associated with OS among women with ER- with a multivariable adjusted HR of 1.38 (95% CI: 1.05–1.82) when comparing the highest to the lowest GRS quartiles. As a contrast, in the ER+ subgroup, patients with higher diabetes GRS tended to have better survival, though this association was not statistically significant (P for trend = 0.45, P for interaction = 0.28). Additional adjustment for BMI did not change the results appreciably. No statistically significant association was observed between the diabetes GRS and DFS, regardless of ER status.
The associations of the diabetes GRS with OS and DFS also varied by diabetes status (P for interaction <0.01, and <0.01 for OS and DFS respectively). In women with self-reported history of diabetes, higher GRS was significantly associated with worse OS and DFS (P for trend < 0.01 and = 0.03 respectively). For OS, the HR was 2.22 (95% CI: 1.28–3.88) when comparing the highest to the lowest quartiles (Table 3). For the DFS, however, none of the 3 upper quartiles reached the statistical significant level, though the linear trend was statistically significant (P = 0.03).
Additional analyses further stratified by ER status and diabetes status were conducted and showed that the diabetes GRS was associated with increased overall survival for ER+ breast cancer patients without diabetes history (P for trend = 0.02) but a decreased OS for those had a history of diabetes (P for trend = 0.01, P for interaction<0.01). The diabetes GRS was associated with decreased OS among patients with ER- breast cancer, particularly those with a history of diabetes, with a HR of 4.59 (95% CI: 1.04–20.28) (Table 4). A sensitivity analysis by excluding 129 participants who had extreme short or extreme long follow up times did not change the results appreciably (Supplementary Table 4).
Analyses stratified by BMI and PR status did not find any heterogeneity in the diabetes GRS-OS or diabetes GRS-DFS associations (data not shown).
Discussion
To our best knowledge, no study, to date, has evaluated the effects of the common inherited genetic variants related to T2D on breast cancer survival. The reasonably high heritability of diabetes, the availability of GWAS-identified risk variants for diabetes among Asian [20], and the strong relationship between diabetes and breast cancer survival [9, 10, 11, 12, 13] motivated us to evaluate the association of T2D genetic variants with breast cancer prognosis. We applied the GRS to measure the combined genetic effects of T2D-associated SNPs on breast cancer survival. We found no overall association between T2D GRS and breast cancer survival among all breast cancer cases. However, among ER- breast cancer patients, higher GRS was associated with worse overall survival. In addition, we found that the association of genetic susceptibility to T2D with survival among breast cancer patients was modified by the history of diabetes. In women with a history of diabetes, higher diabetes GRS was significantly associated with worse OS. The association did not change after adjusting for other prognostic factors, including age at diagnosis, TNM stage, and BMI, indicating it was independently associated with overall survival among patients with ER- breast cancer.
While the impact of diabetes on breast cancer survival has been previously reported, its mechanism remains unclear. It has been suggested that diabetes may affect breast cancer prognosis, possibly due to increased diabetes-related comorbidity, or direct effects of insulin resistance and/or hyperinsulinemia [9, 10, 11, 12, 13, 14, 15]. Insulin is involved in the promotional stage of breast tumorigenesis and progression to metastatic phenotype[28]. In vitro studies suggested that insulin promoted growth of the ER+ human breast cancer cell lines and the effects of insulin on ER- human breast cancer cell lines are quite complex and inconsistent with limited number of cell lines[28]. The epidemiological results are not consistent for the modifying influence of ER status on the association of diabetes and breast cancer survival. Two studies, including a study conducting among early stage Asian breast cancer patients, showed that the prognostic influence of T2D on breast cancer outcome is modified by ER status [11, 13]. A large cohort study recruited 6342 early-stage breast cancer patients in the US, however, suggested that the diabetes was inversely associated with the prognosis of ER+, but not ER-, breast cancer patients[29]. Chlebowski et al. [30, 31] had suggested that insulin or IGF may be important in the ER- breast cancer subset, and the evidence has limited and more studies are needed to eliminate the influence of insulin on ER- breast cancer. A significant association of diabetes with a higher distant metastasis rate during the follow-up period of 5 years was only related to the ER- subgroup [10].
The positive association between higher diabetes GRS and the worse overall survival in women with ER- breast cancer in this study suggested that a genetic background favorable to development of T2D was positively associated with total mortality. Most of T2D genes identified by GWAS are likely responsible for pancreatic β-cell function, which likely mediate their influence on T2D susceptibility [20]. Insulin resistance is also widely accepted to be a major component of T2D susceptibility [32]. Compared with Europeans, East Asians may be more susceptible to insulin resistance [33]. Whether the increased mortality risk among breast cancer survivors with diabetes is driven by an increase in cancer recurrence or a result of competing diabetes-related comorbidities such as cardiovascular disease remains unclear. Previous studies [9, 10, 12] reported that T2D was not significantly associated with DFS, suggesting that the worse survival following BC may due to increased deaths attributed to diabetes-related comorbidities rather than effect of diabetes itself. Finding of a null association for GRS and DFS in the present study also support a role of the diabetes-related access of mortality and morbidity. The finding of an inverse association between GRS and total mortality among ER+ breast cancer patients who had no history of diabetes was unexpected and could possibly be due to chance. It also might be the influence of its effective treatment, especially for hormonal therapy, among ER+ breast cancer patients who had no history of diabetes, which weakened the association of T2D GRS and the risk of mortality. The true association can be estimated only in the ER- breast cancer,
Diabetes is treated with a variety of medications. These has been recently recognized that diabetes treatments, particularly the use of metformin, may mitigate negative influence of diabetes on mortality, and may even reduce the risk of recurrence [18, 19], making it difficult to interpret the diabetes and breast cancer outcomes associations. Unfortunately, this study did not collect information on diabetes medication before or after breast cancer diagnosis. Our findings could at least partly indicate that genetic markers of T2D may affect the ER- breast cancer outcome through some mechanisms related to diabetes alone or others, suggesting a shared genetic component between T2D and ER- breast cancer survival. Furthermore, our findings showed an interactive effect between T2D-related SNPs and diabetes on breast cancer survival. The precise mechanism for our findings is still uncertain and further studies will be needed to further explore this relationship.
Strengths of this study include the large sample size, genetically homogeneous population, prospective investigation of disease outcome, and the availability of detailed clinical characteristics of disease. However, some limitations also warrant mention. One limitation of our study is that it relies on the GWAS catalog for identification of SNPs for inclusion. The threshold for entry of SNPs into the GWAS catalog is higher than that used for most association studies, so some potentially interesting loci may have been missed. Another limitation is that not all T2D-related SNPs among Asian were captured across the entire genome and the percentage of T2D variation captured by the GWAS was low. Furthermore, not all GWAS diabetes SNPs have been included in our study; Two SNPs located on chromosome 23 were not available in our GWAS platform and no proxy SNPs were successfully identified for them, thus our GRS included limited information for these two SNPs. Similarly, in Exome-chip platform, our GRS only captured information from 22 loci. Missing information may compromise the statistical power of our study. Investigations on the molecular pathway specific GRS may provide additional valuable insights into the etiology of these two diseases. However, the hypotheses that lead to these investigations are beyond the scope of the current study. Among the 33 investigated SNPs in the current study, 15 had unclear functionality, 4 had function in insulin action and the rest 14 were related to β-cell function. Future studies with high-density genotyping on certain regions that related to these pathways may provide more convincing information.
In summary, we found that higher diabetes GRS was related with low overall survival among ER- breast cancer patients, particularly those with a history of T2D but was not significantly associated with disease free survival or outcomes of ER+ breast cancer. Further studies are warranted to replicate our findings and to investigate the underlying mechanism(s) of these diseases.
Supporting Information
S1 File. This file contains Tables A-D.
Table A, Associations between 39 T2D-related SNPs in GWAS data and 22 SNPs in Exome-chip data with breast cancer outcomes. Table B, Association of the diabetes Genetic Risk Score (GRS) with breast cancer outcomes in Exome-chip data. Table C, Association of the diabetes Genetic Risk Score (GRS) with breast cancer outcomes in GWAS data. Table D, Association of the diabetes Genetic Risk Score (GRS) with breast cancer outcomes from a sensitivity analysis in which 129 patients with extreme follow-up time (shortest 1% and longest 1%) were excluded.
https://doi.org/10.1371/journal.pone.0117419.s001
(RAR)
Acknowledgments
The authors wish to thank the study participants and research staff for their contributions and commitment to this project,
Author Contributions
Conceived and designed the experiments: WL WZ XOS. Performed the experiments: YTG BZ PPB HC. Analyzed the data: PPB ZGZ. Wrote the paper: PPB ZGZ. Reviewed and approved the final version of the manuscript: PPB WL ZZG YTG YZ BZ HC WZ XOS.
References
- 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, et al. (2011) Global cancer statistics. CA Cancer J Clin 61: 69–90. pmid:21296855
- 2. Nechuta S, Lu W, Zheng Y, Cai H, Bao PP, et al. (2013) Comorbidities and breast cancer survival: a report from the Shanghai Breast Cancer Survival Study. Breast Cancer Res Treat 139: 227–235. pmid:23605082
- 3. Patnaik JL, Byers T, Diguiseppi C, Denberg TD, Dabelea D (2011) The influence of comorbidities on overall survival among older women diagnosed with breast cancer. J Natl Cancer Inst 103: 1101–1111. pmid:21719777
- 4. Sheppard AJ, Chiarelli AM, Marrett LD, Nishri ED, Trudeau ME (2011) Stage at diagnosis and comorbidity influence breast cancer survival in First Nations women in Ontario, Canada. Cancer Epidemiol Biomarkers Prev 20: 2160–2167. pmid:21803843
- 5. Land LH, Dalton SO, Jorgensen TL, Ewertz M (2012) Comorbidity and survival after early breast cancer. A review. Crit Rev Oncol Hematol 81: 196–205. pmid:21536452
- 6. Land LH, Dalton SO, Jensen MB, Ewertz M (2012) Impact of comorbidity on mortality: a cohort study of 62,591 Danish women diagnosed with early breast cancer, 1990–2008. Breast Cancer Res Treat 131: 1013–1020. pmid:22002567
- 7. Srokowski TP, Fang S, Hortobagyi GN, Giordano SH (2009) Impact of diabetes mellitus on complications and outcomes of adjuvant chemotherapy in older patients with breast cancer. J Clin Oncol 27: 2170–2176. pmid:19307509
- 8. Tammemagi CM, Nerenz D, Neslund-Dudas C, Feldkamp C, Nathanson D (2005) Comorbidity and survival disparities among black and white patients with breast cancer. Jama 294: 1765–1772. pmid:16219879
- 9. Erickson K, Patterson RE, Flatt SW, Natarajan L, Parker BA, et al. (2011) Clinically defined type 2 diabetes mellitus and prognosis in early-stage breast cancer. J Clin Oncol 29: 54–60. pmid:21115861
- 10. Schrauder MG, Fasching PA, Haberle L, Lux MP, Rauh C, et al. (2011) Diabetes and prognosis in a breast cancer cohort. J Cancer Res Clin Oncol 137: 975–983. pmid:21132511
- 11. Kaplan MA, Pekkolay Z, Kucukoner M, Inal A, Urakci Z, et al. (2012) Type 2 diabetes mellitus and prognosis in early stage breast cancer women. Med Oncol 29: 1576–1580. pmid:22083554
- 12. Lipscombe LL, Goodwin PJ, Zinman B, McLaughlin JR, Hux JE (2008) The impact of diabetes on survival following breast cancer. Breast Cancer Res Treat 109: 389–395. pmid:17659440
- 13. Chen WW, Shao YY, Shau WY, Lin ZZ, Lu YS, et al. (2012) The impact of diabetes mellitus on prognosis of early breast cancer in Asia. Oncologist 17: 485–491. pmid:22467665
- 14. Kalla Singh S, Tan QW, Brito C, De Leon M, De Leon D (2010) Insulin-like growth factors I and II receptors in the breast cancer survival disparity among African-American women. Growth Horm IGF Res 20: 245–254. pmid:20347606
- 15. Papa V, Belfiore A (1996) Insulin receptors in breast cancer: biological and clinical role. J Endocrinol Invest 19: 324–333. pmid:8796343
- 16. Gregg EW, Cadwell BL, Cheng YJ, Cowie CC, Williams DE, et al. (2004) Trends in the prevalence and ratio of diagnosed to undiagnosed diabetes according to obesity levels in the U.S. Diabetes Care 27: 2806–2812. pmid:15562189
- 17. Yang SH, Dou KF, Song WJ (2010) Prevalence of diabetes among men and women in China. N Engl J Med 362: 2425–2426; author reply 2426 doi: https://doi.org/10.1056/NEJMc1004671. pmid:20578276
- 18. Zhang ZJ, Li S (2014) The prognostic value of metformin for cancer patients with concurrent diabetes: a systematic review and meta-analysis. Diabetes Obes Metab 16: 707–710. pmid:24460896
- 19. Zhang P, Li H, Tan X, Chen L, Wang S (2013) Association of metformin use with cancer incidence and mortality: a meta-analysis. Cancer Epidemiol 37: 207–218. pmid:23352629
- 20. Cho YS, Lee JY, Park KS, Nho CW (2012) Genetics of type 2 diabetes in East Asian populations. Curr Diab Rep 12: 686–696. pmid:22993124
- 21. Hale PJ, Lopez-Yunez AM, Chen JY (2012) Genome-wide meta-analysis of genetic susceptible genes for Type 2 Diabetes. BMC Syst Biol 6 Suppl 3: S16. pmid:23281828
- 22. Zheng W, Chow WH, Yang G, Jin F, Rothman N, et al. (2005) The Shanghai Women′s Health Study: rationale, study design, and baseline characteristics. Am J Epidemiol 162: 1123–1131. pmid:16236996
- 23. Long J, Cai Q, Sung H, Shi J, Zhang B, et al. (2012) Genome-wide association study in east Asians identifies novel susceptibility loci for breast cancer. PLoS Genet 8: e1002532. pmid:22383897
- 24. Zheng W, Long J, Gao YT, Li C, Zheng Y, et al. (2009) Genome-wide association study identifies a new breast cancer susceptibility locus at 6q25.1. Nat Genet 41: 324–328. pmid:19219042
- 25. Shu XO, Zheng Y, Cai H, Gu K, Chen Z, et al. (2009) Soy food intake and breast cancer survival. Jama 302: 2437–2443. pmid:19996398
- 26. Matthews CE, Shu XO, Yang G, Jin F, Ainsworth BE, et al. (2003) Reproducibility and validity of the Shanghai Women’s Health Study physical activity questionnaire. Am J Epidemiol 158: 1114–1122. pmid:14630608
- 27.
Hindorff LA, MacArthur J, Wise A, Junkins HA, Hall PN, et al. A Catalog of Published Genome-Wide Association Studies. Available: http://www.genome.gov/gwastudies. Accessed 2012 Nov 19.
- 28. Rose DP, Vona-Davis L (2012) The cellular and molecular mechanisms by which insulin influences breast cancer risk and progression. Endocr Relat Cancer 19: R225–241. pmid:22936542
- 29. Jiralerspong S, Kim ES, Dong W, Feng L, Hortobagyi GN, et al. (2013) Obesity, diabetes, and survival outcomes in a large cohort of early-stage breast cancer patients. Ann Oncol 24: 2506–2514. pmid:23793035
- 30. Chlebowski RT, Blackburn GL, Thomson CA, Nixon DW, Shapiro A, et al. (2006) Dietary fat reduction and breast cancer outcome: interim efficacy results from the Women’s Intervention Nutrition Study. J Natl Cancer Inst 98: 1767–1776. pmid:17179478
- 31. Chlebowski RT, Anderson GL, Gass M, Lane DS, Aragaki AK, et al. (2010) Estrogen plus progestin and breast cancer incidence and mortality in postmenopausal women. Jama 304: 1684–1692. pmid:20959578
- 32. Muoio DM, Newgard CB (2008) Mechanisms of disease: molecular and metabolic mechanisms of insulin resistance and beta-cell failure in type 2 diabetes. Nat Rev Mol Cell Biol 9: 193–205. pmid:18200017
- 33. Chan JC, Malik V, Jia W, Kadowaki T, Yajnik CS, et al. (2009) Diabetes in Asia: epidemiology, risk factors, and pathophysiology. Jama 301: 2129–2140. pmid:19470990