Genetically predicted sleep duration in relation to risk of breast tumor

Authors

Abstract

Aim The study aims to reveal the causal relationship between sleep duration and breast tumors.
Material and Methods In the “Material and Methods” section, we conducted a study using a two-sample bi-directional Mendelian randomization (MR) analysis. Our aim was to investigate the potential cause-and-effect relationship between genetically predicted sleep duration and the risk of developing breast tumors. To analyze the MR results, we employed the inverse variance weighting (IVW) method, which is considered the gold standard. We conducted supplementary analyses using the MR Egger regression, weighted median, weighted mode, and simple mode methods. Sensitivity analyses were carried out to assess the reliability and validity of the identified causal relationships. Furthermore, we performed a reverse-direction MR analysis to examine whether breast tumors have a causal effect on sleep duration.
Results We found suggestive evidence indicating a potential causal relationship between genetically predicted sleep duration and breast cancer. The odds ratio (OR) was 1.328, with a 95% confidence interval (CI) of 1.013 to 1.741, and a p-value of 0.04. When analyzing the different molecular subtypes of breast cancer, we observed a trend suggesting a causal association between sleep duration and both estrogen receptor-positive ER+ and ER- breast cancers. (ER+ breast cancer p = 0.051, ER- breast cancer p = 0.077). However, we did not find any evidence of a causal effect of sleep duration on benign breast tumors (OR = 1.002, 95%CI: 0.999 - 1.005, p = 0.242) or carcinoma in situ of the breast (OR = 1.090, 95%CI: 0.403 - 2.947, p = 0.08). Additionally, our reverse MR analysis did not indicate that having breast cancer could influence sleep duration (OR = 1.002, 95% CI: 0.995 - 1.010, p = 0.494).
Conclusion The findings of this study indicate that there is a causal relationship between genetically predicted sleep duration and breast cancer. These results suggest that maintaining an appropriate sleep duration and avoiding excessive sleep may be beneficial in reducing the risk of developing breast cancer.

Keywords

Introduction

Breast cancer is the most common cancer among women worldwide, accounting for a substantial number of new cancer cases and cancer-related deaths. Lifestyle factors play a critical role in the development of breast cancer. For instance, smoking has been identified as a potential risk factor, particularly in premenopausal women ¹. Conversely, regular physical activity has been linked to a decreased risk of breast cancer ². Understanding these modifiable factors empowers individuals to make informed choices and adopt healthy behaviors that can potentially reduce their risk of developing breast cancer.
Sleep-related characteristics, such as sleep duration, quality, habits, rhythms, and disorders, have been extensively studied in relation to health outcomes ³. Numerous studies have explored the association between sleep traits and the incidence of breast cancer, revealing that both insufficient sleep duration and excessive sleep duration are associated with an increased risk of breast cancer ⁴. For instance, a study involving 23,620 cases found that sleeping less than 6 hours per night was associated with a 43% higher risk of cancer, including breast cancer ⁵. Other studies have also reported a positive correlation between long sleep duration (more than 9 hours per night) and breast cancer ^6,7. Nevertheless, a definitive causal relationship between sleep duration and breast cancer has not been firmly established. Several epidemiological studies and meta-analyses have been conducted, and they have consistently reported no significant association between sleep duration and the risk of developing cancer ^8,9.
Mendelian randomization is a statistical technique that leverages genetic variants as instrumental variables to establish causal relationships between exposures and outcomes ¹⁰. In this approach, genetic variants are utilized as proxies or instrumental variables for the modifiable risk factors or exposures of interest in Mendelian randomization studies.
These genetic variants are randomly allocated during meiosis and inherited independently of confounding factors, mimicking the process of a randomized controlled trial. This random allocation provides a natural experiment-like setting, making Mendelian randomization a powerful tool for causal inference. By using genetic variants, which are determined at conception and not influenced by confounders, the risk of confounding bias is greatly reduced.
In this study, we employed a two-sample Mendelian randomization (MR) analysis to examine the potential causal relationship between sleep duration and various types of breast tumors, including breast cancer, ER+/ER- breast cancer, HER2+/HER2- breast cancer, benign breast tumors, and carcinoma in situ of the breast. Additionally, we conducted a reverse Mendelian randomization analysis to investigate whether breast tumors have a causal effect on sleep duration.

Materials and Methods

Study DesignWe conducted a bidirectional two-sample Mendelian randomization (MR) analysis to examine the causal effects between sleep duration and breast tumors. Both sleep duration and breast tumors were considered as exposure factors to assess their respective causal effects on each other. In the bidirectional analysis, we utilized specific genetic variants known as short nucleotide polymorphisms (SNPs) that are associated with sleep duration or breast tumors as instrumental variables (IVs) to infer causal relationships between the exposures and outcomes. To obtain the necessary GWAS datasets, we utilized the IEU OpenGWAS database (https://gwas.mrcieu.ac.uk/), which houses a vast collection of over 40,000 GWAS datasets containing SNPs associated with various traits.
Determination of IVsAll SNPs correlated with the exposure trait at a genome-wide significance level (p < 5 × 10−8) were extracted as potential IVs. SNPs in high linkage disequilibrium (r2 > 0.001 or clump windows < 10,000 kb) were excluded to eliminate bias caused by linkage disequilibrium (LD). Harmonization was carried out to eliminate ambiguous SNPs that exhibited non-concordant alleles. When a SNP was not presented in the outcome summary statistics, a proxy SNP highly correlated with the variant of interest (LD, r2 > 0.8) was selected for substitution. However, if a substitute could not be identified, the SNP was excluded. Palindromic SNPs were aligned when minor allelic frequencies were less than 0.3. IVs with an F-statistic of < 10 were excluded due to their weak correlations.
MR StatisticsWe used the inverse variance weighted (IVW) method to obtain the main results of the two-sample MR analysis ¹¹. The other four MR analysis methods, including MR Egger regression, weighted median, weighted mode, and simple mode methods, were used to obtain secondary results and validation. A clear causal relationship between exposure and outcome was considered for IVW results with P-values of less than 0.05, and the p-value of the other four methods was in the same direction as the IVW method p-value. Sensitivity analyses were performed, including the heterogeneity test, horizontal pleiotropy test, and leave-one-out analysis, to further validate the obtained causal relationships and assess the reliability of the results. All MR statistical analyses were conducted using the TwoSampleMR package (version 0.5.7) in R software (version 4.2.2).
Ethical ApprovalThis study was approved by the Ethics Committee of the Affiliated Hospital of Shaanxi University of Traditional Chinese Medicine (Date: 2024-05-09, No: SZFYIEC-KYBC-2024-03).

Results

Data for MR AnalysisFigure 1 displays the workflow of the study. The sleep duration-related dataset (GWAS ID: ukb-b-4424 ) that was selected consisted of 460,099 samples and 9,851,867 SNPs from European populations. GWAS breast tumor datasets included breast cancer datasets (malignant, HER2 and ER status undefined), ER+/ER- breast cancer, HER2+/HER2- breast cancer, benign breast tumors, and carcinoma in situ of the breast (Table 1).
Causal Effect of Sleep Duration on Breast CancerFigure 2 and Table 2 display the results of the MR analysis of the causal effect estimate of sleep duration on breast cancer. A total of 66 SNPs were included in the MR analysis. We identified a significant causal relationship between sleep duration and the risk of breast cancer, with a p-value of 0.04 (odds ratio (OR) = 1.328, 95% confidence interval (CI): 1.013 – 1.741) in the IVW analysis (Figure 2A and 2B). A positive result was also detected using the weighted median method (OR = 1.320, 95%CI: 1.030 – 1.693, p = 0.028). While the secondary indicators (MR Egger, simple mode, and weighted mode methods) did not indicate statistically significant results, b values consistently pointed in the same direction. This consistency indicates a positive association between sleep duration and breast cancer, despite the lack of significant findings.
The MR-Egger regression results confirmed that our results were not significantly affected by horizontal pleiotropy (p = 0.995). The funnel plot and Cochran’s Q test indicated significant heterogeneity among the SNPs involved in the IVW analysis (p = 1.828e-19) (Figure 3A). A random-effects model was used in the analysis to avoid statistical bias due to heterogeneity. The leave-one-out analysis showed that single SNP removal did not significantly affect the overall estimates (Figure 3B), suggesting that the results were highly reliable.
Causal Effect of Sleep Duration on ER+/ER- or HER2+/HER2- Breast CancerAdditional MR analyses were performed in ER+/ER- and HER2+/HER2- breast cancers to further verify the causal effect between sleep duration and different molecular subtypes of breast cancer. A total of 65 SNPs were finally selected for the causal effect analysis of sleep duration on ER+ breast cancer. Although the IVW method detected no statistical significance, the p-value was close to 0.05 (p = 0.051). Analyses using the weighted median and weighted mode methods revealed a significant causal relationship between sleep duration and ER+ breast cancer (weighted median, p = 0.003; weighted mode, p = 0.036) (Figure S1, Table 2). Similarly, a positive causal effect was identified between sleep duration and ER- breast cancer. A p-value with borderline statistical significance (p = 0.077) was obtained using the IVW method. However, a significant statistical difference was obtained by the weighted median method (p = 0.028). The results obtained by the other three methods showed similar trends in the causal effect of sleep duration on ER- breast cancer, but without statistical significance (MR Egger p = 0.569, simple mode p = 0.090, weighted mode p = 0.097) (Figure S2, Table 2).
The IVW approach did not show a significant causal connection in the MR analysis of sleep duration on HER2+/HER2- breast cancer (HER2+ breast cancer: OR = 0.963, 95%CI: 0.564 – 1.642, p = 0.889; HER2- breast cancer: OR = 1.825, 95%CI: 0.958 – 3.477, p = 0.067) (Figure S3, Figure S4, Table2). Although the p-value for the causal effect of sleep on HER2- breast cancer was close to 0.05, we did not believe that there was a clear causal relationship between sleep duration and HER2- breast cancer because MR Egger’s method detected a causal effect in the opposite direction (b = -0.278).
Causal Effect of Sleep Duration on Benign Breast TumorA total of 44 SNPs were used for MR analysis of sleep duration on benign breast tumors. The IVW approach did not show a causal effect of sleep duration on benign breast tumors (OR = 1.002, 95%CI: 0.999 – 1.005, p = 0.242). Secondary analysis methods, including MR-Egger regression (OR = 0.996; 95%CI: 0.970 – 1.022, p = 0.757), weighted median (OR = 1.001; 95%CI: 0.997 – 1.005, p = 0.606), simple mode (OR = 0.999; 95%CI: 0.991 – 1.008, p = 0.826), and weighted mode (OR = 1.000; 95%CI: 0.991 – 1.008, p = 0.949) approaches, exhibited the same results (Figure S5, Table 2).
Causal Effect of Sleep Duration on Carcinoma in Situ of the BreastThe causal relationship between sleep duration and carcinoma in situ of the breast was also examined. Heterogeneity was not found for changes in SNPs (p = 0.811). The MR analysis found no statistical causal effect using the IVW method (OR = 1.090, 95%CI: 0.403 – 2.947, p = 0.086) and the other four methods. Finally, the MR Egger regression intercept provided no evidence for directional pleiotropy (p = 0.743) (Figure S6, Table 2).
Reverse MR ResultsWe performed a reverse MR analysis to verify the possibility of reverse causality, in which breast tumors were used as the exposure and sleep duration as the outcome. A total of 126 SNPs related to breast cancer were finally included in reverse MR analysis. The IVW analysis results did not indicate that having breast cancer could affect sleep duration (OR = 0.998, 95%CI: 0.989 – 1.006, p = 0.670) (Table 3). Further analysis based on breast cancer molecular typing generated similar results, that ER+/ER- and HER2+/HER2- breast cancer did not have a causal effect on sleep duration. The analysis investigating the causal relationship between sleep duration and benign breast tumors did not produce statistically significant results (odds ratio = 2.011, 95% confidence interval: 0.26 - 153.896, p = 0.752). It’s important to note that this analysis was based on only one SNP and had low confidence. As for the analysis examining the causal relationship between sleep duration and carcinoma in situ of the breast, no relevant SNP was identified, resulting in no results from this particular aspect of the reverse analysis.

Discussion

Our results suggested evidence for causal effects between sleep duration and the risk of breast cancer. Women with longer sleep duration might have a significantly increased risk of breast cancer. However, we did not detect a clear causal effect of sleep duration on HER2+/HER2- breast cancer, benign breast tumors, or carcinoma in situ of the breast, nor did we find that having breast tumors increased or decreased sleep duration.
There is a traditional belief that shorter sleep duration might be a risk factor for breast cancer, and long sleep duration might reduce the risk of cancer ^12,13. However, our study generated inconsistent conclusions. Sleep duration had a causal effect on breast cancer, and longer rather than shorter sleep duration might increase the risk of breast cancer. The positive association between long sleep duration and breast cancer has attracted considerable attention in recent years ². A study by Wang et al. reported a positive association between long sleep duration (> 9 hours a night) and breast cancer incidence compared to a reference sleep duration (6.1 – 8.9 hours a night) ¹⁴. Another case-control study revealed that the risk of breast cancer increased with increasing sleep duration for every additional sleeping hour (OR = 1.06, 95%CI: 1.01 – 1.11) ¹⁵. In addition, a meta-analysis including 10 studies showed that women with longer sleep durations had a significantly increased risk of breast cancer, and the effect was dose-dependent ¹⁶. Our study provided definitive evidence for the association between increased breast cancer risk and long sleep duration using the two-sample MR analysis with SNPs as IVs. Unlike prospective or retrospective cohort studies, this analysis eliminated biases that could arise from confounding factors, making the conclusions more reliable.
The detailed mechanism by which sleep duration affects the occurrence of breast cancer is not clear. Several studies suggested that reduced melatonin secretion due to sleep deprivation might be a major mechanism promoting cancer development, and melatonin is known to have an effect on protecting the body from cancer ^17,18. However, such an explanation does not fit our results. Long-duration sleepers have higher cortisol levels than short-duration sleepers ^19,20. Cortisol is involved in multiple processes involved in the genesis and development of breast cancer, such as 1) the regulation of mammary epithelium growth, 2) the impairment of immune activity, and 3) the suppression of natural killer cell activity ^21,22. These effects of cortisol may be the underlying mechanisms by which long-term sleep duration increases the risk of breast cancer ²³.
Patients with cancer were reported to be at a high risk of sleep disorders, the most prominent of which was insomnia. No causal effect of breast tumors on sleep duration was detected in our reverse MR analysis. However, we could not draw a strong conclusion that breast tumors do not affect patients’ sleep duration because there were too few SNPs as IVs in the evaluation of ER- breast cancer (SNPs = 7), HER2+/HER2- breast cancer (SNPs = 8/4), benign neoplasm of the breast (SNPs = 1), and carcinoma in situ of the breast (SNPs = 0). In fact, the causal effect of breast tumors on sleep duration is complex. Although some patients may experience sleep problems during cancer or treatment, not all patients are affected. The prognosis of patients with breast cancer is generally good, and endocrine therapy or targeted therapy can replace traditional chemotherapy and reduce toxic side effects ²⁴. Therefore, the psychological stress caused by the disease and treatment of breast cancer patients is less than that of other malignant tumors, such as liver cancer and lung cancer. Psychological stress-related anxiety, depression, etc., are causes of sleep problems. In contrast, sleep disorders in cancer patients are closely related to treatment factors, individual differences, economic factors, and the sleeping environment ²⁵.
Some limitations existed in this study. First, heterogeneity was detected in the evaluation of breast cancer, ER+ breast cancer, and ER- breast cancer. We used the random-effects IVW method to correct for heterogeneity to ensure the validity of the results. Second, we only used the IEU OpenGWAS database for analysis, which does not represent all GWAS data related to sleep and breast tumors. Finally, sleep traits include sleep duration, chronotype, sleep quality, and so on. However, we only analyzed sleep duration in this study. Thus, further studies are needed.

Conclusion

Our study suggested the potential causal effect of genetically predicted sleep duration on breast cancer, while a causal effect of sleep duration on benign breast tumors and carcinoma in situ of the breast was not identified. The reverse analysis did not prove the existence of a causal effect of breast tumors on sleep duration. We recommend that women maintain an appropriate sleep duration and avoid excessive sleep, which is helpful for preventing breast cancer.

Declarations

References

1. Park HA, Neumeyer S, Michailidou K, Bolla MK, Wang Q, Dennis J, et al. Mendelian randomisation study of smoking exposure in relation to breast cancer risk. Br J Cancer. 2021;125(8):1135-45.
   PubMed Google Scholar
2. Xu Y, Rogers CJ. Physical activity and breast cancer prevention: Possible role of immune mediators. Front Nutr. 2020;7:557997.
   PubMed Google Scholar
3. Richmond RC, Anderson EL, Dashti HS, Jones SE, Lane JM, Strand LB, et al. Investigating causal relations between sleep traits and risk of breast cancer in women: Mendelian randomisation study. BMJ. 2019;365:l2327.
   PubMed Google Scholar
4. Beverly HC, Hale L, Naughton MJ. Contributions of the Women’s Health Initiative to understanding associations between sleep duration, insomnia symptoms, and sleep-disordered breathing across a range of health outcomes in postmenopausal women. Sleep Health. 2020;6(1):48-59.
   PubMed Google Scholar
5. von Ruesten A, Weikert C, Fietze I, Boeing H. Association of sleep duration with chronic diseases in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam study. Plos One. 201;7(1):e30972.
   PubMed Google Scholar
6. Shen J, Chrisman M, Wu X, Chow WH, Zhao H. Sleep duration and risk of cancer in the Mexican American Mano-a-Mano Cohort. Sleep Health. 2019;5(1):78-83.
   PubMed Google Scholar
7. Lu C, Sun H, Huang J, Yin S, Hou W, Zhang J, et al. long-term sleep duration as a risk factor for breast cancer: Evidence from a systematic review and dose-response meta-analysis. Biomed Res Int. 2017; 2017:4845059.
   PubMed Google Scholar
8. Wong A, Heath AK, Tong T, Reeves GK, Floud S, Beral V, et al. Sleep duration and breast cancer incidence: Results from the Million Women Study and meta-analysis of published prospective studies. Sleep. 2021;44(2):zsaa166.
   PubMed Google Scholar
9. Chen Y, Tan F, Wei L, Li X, Lyu Z, Feng X, et al. Sleep duration and the risk of cancer: A systematic review and meta-analysis including dose-response relationship. Bmc Cancer. 2018;18(1):1149.
   PubMed Google Scholar
10. Emdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA. 2017;318(19):1925-6.
    PubMed Google Scholar
11. Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife. 2018;7:e34408.
    PubMed Google Scholar
12. Yin KJ, Huang JX, Wang P, Yang XK, Tao SS, Li HM, et al. No genetic causal association between periodontitis and arthritis: A bidirectional two-sample mendelian randomization analysis. Front Immunol. 2022;13:808832.
    PubMed Google Scholar
13. Cao J, Eshak ES, Liu K, Muraki I, Cui R, Iso H, et al. Sleep duration and risk of breast cancer: The JACC study. Breast Cancer Res Treat. 2019;174(1):219-25.
    PubMed Google Scholar
14. Liu L, Bao H, Wang F, Yu L, Cong S, Zhou F, et al. Depressive symptoms and sleep duration as risk factors for breast cancer - China, 2020. China CDC Wkly. 2023;5(15):333-9.
    PubMed Google Scholar
15. Wang P, Ren FM, Lin Y, Su FX, Jia WH, Su XF, et al. Night-shift work, sleep duration, daytime napping, and breast cancer risk. Sleep Med. 2015;16(4):462-8
    PubMed Google Scholar
16. Wu AH, Stanczyk FZ, Wang R, Koh WP, Yuan JM, Yu MC. Sleep duration, spot urinary 6-sulfatoxymelatonin levels and risk of breast cancer among Chinese women in Singapore. Int J Cancer. 013;132(4):891-6.
    PubMed Google Scholar
17. Lu C, Sun H, Huang J, Yin S, Hou W, Zhang J, et al. Long-term sleep duration as a risk factor for breast cancer: Evidence from a systematic review and dose-response meta-analysis. Biomed Res Int. 2017;2017:4845059.
    PubMed Google Scholar
18. Sadoughi F, Dana PM, Asemi Z, Shafabakhash R, Mohammadi S, Heidar Z, et al. Molecular and cellular mechanisms of melatonin in breast cancer. Biochimie. 2022;202:26-33.
    PubMed Google Scholar
19. Kubatka P, Zubor P, Busselberg D, Kwon TK, Adamek M, Petrovic D, et al. Melatonin and breast cancer: Evidences from preclinical and human studies. Crit Rev Oncol Hematol. 2018;122:133-43.
    PubMed Google Scholar
20. Dettenborn L, James GD, van Berge-Landry H, Valdimarsdottir HB, Montgomery GH, Bovbjerg DH. Heightened cortisol responses to daily stress in working women at familial risk for breast cancer. Biol Psychol. 2005;69(2):167-79.
    PubMed Google Scholar
21. Ramirez-Exposito MJ, Duenas-Rodriguez B, Carrera-Gonzalez MP, Navarro-Cecilia J, Martinez-Martos JM. Circulating levels of beta-endorphin and cortisol in breast cancer. Compr Psychoneuroendocrinol. 2021;5:100028.
    PubMed Google Scholar
22. Meszaros CE, Lopez-Gigosos R, Mariscal-Lopez E, Agredano-Sanchez M, Garcia-Casares N, Mariscal A, et al. Psychosocial interventions reduce cortisol in breast cancer patients: Systematic review and meta-analysis. Front Psychol. 2023;14:1148805.
    PubMed Google Scholar
23. Zheng Y, Zhang J, Huang W, Zhong L, Wang N, Wang S, et al. Sini San Inhibits Chronic Psychological Stress-Induced Breast Cancer Stemness by Suppressing Cortisol-Mediated GRP78 Activation. Front Pharmacol. 2021;12:714163.
    PubMed Google Scholar
24. Harbeck N, Penault-Llorca F, Cortes J, Gnant M, Houssami N, Poortmans P, et al. Breast cancer. Nat Rev Dis Primers. 2019;5(1):66.
    PubMed Google Scholar
25. Imanian M, Imanian M, Karimyar M. Sleep quality and fatigue among breast cancer patients undergoing chemotherapy. Int J Hematol Oncol Stem Cell Res. 2019;13(4):196-200.
    PubMed Google Scholar

Additional Information

Publisher’s Note
Bayrakol MP remains neutral with regard to jurisdictional and institutional claims.

Rights and Permissions

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). To view a copy of the license, visit https://creativecommons.org/licenses/by-nc/4.0/

View full license details →

About This Article

Published Online:

March 11, 2026