Abstract
Purpose:
Both long and short sleep duration have been linked with risk of some cancers, but evidence for glioma is lacking.
Methods:
Using prospective data from the UK Biobank (UKB), the Nurses’ Health Study (NHS), and the Health Professionals Follow-Up Study (HPFS), we examined the association between self-reported hours of sleep and incident glioma in multivariable-adjusted Cox proportional hazards models.
Results:
In the UKB, compared to 7 hours, sleep durations of < 7 hours (HR=0.90; 95% CI, 0.70-1.16) or > 7 hours (HR=1.05; 95% CI, 0.85-1.30) were not significantly associated with glioma risk. Likewise, no significant associations were found between sleep duration and glioma risk in the NHS/HPFS for either < 7 hours (HR=0.93; 95% CI, 0.69-1.26), or > 7 hours (HR=1.22; 95% CI, 0.94-1.57), compared to 7 hours. Results were similar for low-grade and high-grade glioma, did not materially change after lagging 2 years, or after accounting for factors known to disrupt sleep.
Conclusion:
Sleep duration was not associated with incident glioma in either the UKB or the NHS/HPFS cohorts.
Keywords: glioma, glioblastoma, incidence, etiology, sleep
Introduction
Malignant gliomas comprise nearly 80% of primary malignant brain tumors in adults (1). Glioblastoma multiforme (GBM), the most aggressive grade IV tumor, accounts for ~60% of gliomas, and has a poor prognosis regardless of treatment, with median survival of 15 months (1). The etiology of glioma is poorly understood; to date, no modifiable risk factors for glioma have been identified. The only environmental risk factor associated with risk of developing glioma is moderate to high doses of ionizing radiation (2-3).
Results from studies of sleep duration and cancer have been mixed (4-6), and none have evaluated the relationship between sleep and glioma risk. Insufficient sleep induces innate pro-inflammatory mediators in the brain, and glial cells contribute to sleep and immune system interactions in the brain (7). The ‘glymphatic system,’ a waste clearance system in the brain active only during sleep, is formed by astroglial cells that eliminate potentially neurotoxic waste products and is suppressed in various neurodegenerative diseases (8), raising the possibility that short sleep may contribute to glioma risk. Using prospective data from the UK Biobank (UKB), the Nurses’ Health Study (NHS), and the Health Professionals Follow-Up Study (HPFS), we examined the association between self-reported hours of sleep in every 24-hour period and incident glioma overall, and for GBM and lower-grade gliomas (non-GBM).
Methods
The UK Biobank (UKB) cohort follows approximately 500,000 participants enrolled from 2006 to 2010, who were aged 40-69 years at the time of recruitment. Potential participants were identified from National Health Service patient registries and invited to complete a baseline questionnaire at one of 22 recruitment centers. Details of the study have been published elsewhere (9). Sleep duration was assessed at baseline with the question “About how many hours sleep do you get in every 24 h? (Please include naps.)”. Participants were followed through on-going record linkage with the National Health Service central registers for information on cancer diagnoses, coded according to the WHO International Classification of Diseases.
After excluding participants with prevalent cancers, genetically related participants, and those missing sleep data, the analysis included 436,007 participants and 432 incident gliomas. Follow-up time included time from date of enrollment to diagnosis of glioma or other cancer, death, or last linkage. All UKB participants provided informed consent, and the study is approved by the North West Multi-Centre Research Ethics Committee, the Patient Information Advisory Group, and the Community Health Index Advisory Group (9).
The methods of the NHS and HPFS have been previously reported (10). Primary brain malignancy cases were self-reported on questionnaires or ascertained through mortality follow-up and confirmed by medical record review. Sleep was assessed initially in 1987 for HPFS, and in 1986 for NHS, and updated for both in 2000. The proportion of participants classified in the same sleep category in both assessments was 58% (p<0.001) for both the HPFS and NHS, suggesting reasonable consistency in sleep duration over time. Follow-up time began the date of return of the baseline questionnaire and continued to the date of glioma diagnosis, death, or the end of follow-up (June 30, 2014 for NHS; December 31, 2017 for HPFS). The analysis included 101,757 participants and 334 incident gliomas. The study protocol was approved by the institutional review boards of the Brigham and Women’s Hospital and Harvard T.H. Chan School of Public Health, and those of participating registries as required.
Only confirmed glioma cases were included in this analysis. Hours of sleep were categorized as < 7, 7, and >7 hours per 24-hour period, with 7 hours as the referent. We used multivariable Cox proportional hazards regression models to calculate hazard ratios (HR) and 95% confidence intervals (CI) for the association of sleep duration and incidence of glioma overall, and by tumor subtype. NHS and HPFS analyses were pooled by meta-analysis using fixed effects models. To mitigate possible effects of protopathic bias, we repeated the analyses by beginning follow-up two years after sleep assessment (lagged analysis). Current cigarette smoking, education, alcohol consumption, body mass index (BMI, kg/m2), history of diabetes, menopausal status in women, and caffeine intake were also considered as potential confounders or sleep modifiers. Covariates were collected at baseline assessment for UKB participants, and were time-varying for NHS and HPFS participants. All statistical analyses were performed using SAS 9.4 (SAS Institute, Cary, NC).
Results
In the UKB, compared to 7 hours, sleep durations of < 7 hours (HR=0.90; 95% CI, 0.70-1.16) or > 7 hours (HR=1.05; 95% CI, 0.85-1.30) were not significantly associated with glioma risk (Table 1). Likewise, no significant associations were found between sleep duration and glioma risk in the NHS/HPFS for either < 7 hours (HR=0.93; 95% CI, 0.69-1.26), or > 7 hours (HR=1.22; 95% CI, 0.94-1.57), compared to 7 hours (Table 1). In all three cohorts, there was no evidence for confounding or effect modification by factors known to disrupt sleep, including current smoking, alcohol consumption, caffeine intake, menopausal status among women, or BMI. Results did not materially change after excluding cases diagnosed within the first two years of follow-up.
Table 1.
Hazard ratios (HR) and 95% confidence intervals (95% CI) for hours of sleep and risk of glioma in the UK Biobank and Nurses’ Health Study/Health Professionals Follow-up Study overall, and lagged 2 years.
| Glioma subtype | UK Biobanka | NHS/HPFSb | |||||
|---|---|---|---|---|---|---|---|
| Overall | |||||||
| No. hours sleep in 24 hour period |
Cases | HR (95% CI) | p-value | Casesc | HR (95% CI) | p-value | |
| All glioma | 7 hours | 167 | Ref | 122 | Ref | ||
| >7 hours | 170 | 1.05 (0.85-1.30) | 0.66 | 121 | 1.22 (0.94-1.57) | 0.13 | |
| < 7 hours | 95 | 0.90 (0.70-1.16) | 0.42 | 71 | 0.93 (0.69-1.26) | 0.65 | |
| Ptrendd | 0.25 | e | |||||
| GBM | 7 hours | 120 | Ref | 86 | Ref | ||
| >7 hours | 127 | 1.09 (0.84-1.40) | 0.52 | 81 | 1.15 (0.84-1.56) | 0.38 | |
| < 7 hours | 73 | 0.96 (0.72-1.29) | 0.79 | 48 | 0.90 (0.63-1.29) | 0.58 | |
| Ptrendd | 0.39 | e | |||||
| Non-GBM | 7 hours | 47 | Ref | 36 | Ref | ||
| >7 hours | 43 | 0.96 (0.68-1.45) | 0.83 | 40 | 1.36 (0.86-2.16) | 0.19 | |
| < 7 hours | 22 | 0.75 (0.45-1.25) | 0.27 | 23 | 1.00 (0.59-1.70) | 0.99 | |
| Ptrendd | 0.43 | e | |||||
| Lagged 2 years | |||||||
| All glioma | 7 hours | 123 | Ref | 105 | Ref | ||
| >7 hours | 133 | 1.12 (0.88-1.43) | 0.37 | 76 | 1.31 (1.01-1.69) | 0.04 | |
| < 7 hours | 71 | 0.92 (0.69-1.24) | 0.59 | 114 | 1.06 (0.79-1.42) | 0.69 | |
| Ptrendd | 0.17 | e | |||||
| GBM | 7 hours | 87 | Ref | 75 | Ref | ||
| >7 hours | 98 | 1.17 (0.89-1.56) | 0.30 | 50 | 1.32 (0.97-1.79) | 0.08 | |
| < 7 hours | 56 | 1.02 (0.73-1.43) | 0.89 | 82 | 0.98 (0.68-1.40) | 0.89 | |
| Ptrendd | 0.38 | e | |||||
| Non-GBM | 7 hours | 36 | Ref | 30 | Ref | ||
| >7 hours | 35 | 1.00 (0.63-1.60) | 0.99 | 26 | 1.26 (0.78-2.03) | 0.34 | |
| < 7 hours | 15 | 0.68 (0.37-1.24) | 0.20 | 32 | 1.25 (0.75-2.09) | 0.39 | |
| Ptrendd | 0.25 | e | |||||
HR and 95%CIs adjusted for age (5-year age groups), gender, and race (White, non-White, missing).
HR and 95%CIs adjusted for age (months), calendar year, cohort (which adjusts for gender), and race (White, non-White, missing, in HPFS).
Does not sum to total case count due to missing values for sleep after baseline.
Ptrend was calculated using a continuous variable created from the medians within each category of consumption.
Unable to calculate due to categorical nature of sleep data. In HPFS, participants were asked to select from the following options for typical sleep in a 24 hour period: <=5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10, 10.5, or >=11 hours; in NHS, the options were <=5, 6, 7, 8, 9, 10, or >=11 hours.
Discussion
Results from previous studies of sleep and cancer have been mixed (4-6), and a recent meta-analysis showed no overall increased risk for either short or long sleep duration (5). In a prospective analysis of sleep duration and multisite cancer risk, Gu and colleagues (6) observed no significant associations between short or long sleep durations (with 7-8 hours as the referent) and brain cancer overall for men or women. To date, there have been no studies of sleep duration and glioma.
Strengths of this analysis include the assessment of sleep prior to glioma diagnosis, and effort to replicate findings in three large cohort studies. Although we evaluated more granular categories of sleep duration, case numbers were sparse in categories beyond < 7 and < 7 hours. We were unable to evaluate the influence of changes in sleep duration over time. In addition, follow-up time was limited in the UKB cohort, and glioma case numbers were modest in all three cohorts, particularly for non-GBM tumors. In summary, our findings provide no support for an association between sleep duration and glioma risk.
Acknowledgments
We would like to thank the participants and staff of the Nurses’ Health Study and Health Professionals Follow-Up Study for their valuable contributions as well as the following state cancer registries for their help: AL, AZ, AR, CA, CO, CT, DE, FL, GA, ID, IL, IN, IA, KY, LA, ME, MD, MA, MI, NE, NH, NJ, NY, NC, ND, OH, OK, OR, PA, RI, SC, TN, TX, VA, WA, WY. The authors assume full responsibility for analyses and interpretation of these data. The work is based on the UK Biobank Resource under application number 16944.
Funding:
This project is supported by the H. Lee Moffitt Cancer Center & Research Institute, an NCI-designated Comprehensive Cancer Center (P30-CA076292), and funded by NIH grants PO1 CA87969, U01 CA167552, UM1 CA186107, UM1 CA167552, F30 CA235791 (DJC).
Footnotes
Conflicts of interest/Competing interests: The authors declare no potential conflicts of interest.
Availability of data and material (data transparency): The Health Professionals Follow-up Study and Nurses Health Studies datasets analyzed during the current study are available from the corresponding author on reasonable request. The UK Biobank data that support the findings of this study are available from the UK Biobank.
Ethics approval: This study was performed in line with the principles of the Declaration of Helsinki. Approval for the UK Biobank was granted by the North West Multi-Centre Research Ethics Committee, the Patient Information Advisory Group, and the Community Health Index Advisory Group. The study protocols for the Health Professionals Follow-up Study and Nurses Health Studies were approved by the institutional review boards of the Brigham and Women’s Hospital and Harvard T.H. Chan School of Public Health, and those of participating registries as required.
Consent to participate: Informed consent was obtained from all individual participants included in the study.
References
- 1.Ostrom QT, Bauchet L, Davis FG, Deltour I, Fisher JL, Langer CE,, et al. The epidemiology of glioma in adults: a "state of the science" review. Neuro Oncol 2014;16:896–913. 10.1093/neuonc/nou087 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Preston DL, Ron E, Yonehara S, et al. Tumors of the nervous system and pituitary gland associated with atomic bomb radiation exposure. J Natl Cancer Inst. 2002;94(20):1555–63. doi: 10.1093/jnci/94.20.1555 [DOI] [PubMed] [Google Scholar]
- 3.Ron E, Modan B, Boice JD Jr., et al. Tumors of the brain and nervous system after radiotherapy in childhood. N Engl J Med. 1988;319(16):1033–9. doi: 10.1056/NEJM198810203191601 [DOI] [PubMed] [Google Scholar]
- 4.Zhao H, Yin JY, Yang WS, Qin Q, Qin Q, Li TT, Shi Y, et al. Sleep duration and cancer risk: a systematic review and meta-analysis of prospective studies. Asian Pac J Cancer Prev 2013;14:7509–15. 10.7314/APJCP.2013.14.12.7509 [DOI] [PubMed] [Google Scholar]
- 5.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;18:1149. 10.1186/s12885-018-5025-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gu F, Xiao Q, Chu LW, Yu K, Matthews CE, Hsing AW, et al. Sleep duration and cancer in the NIH-AARP Diet and Health Study Cohort. PLoS One 2016;11:e0161561. 10.1371/journal.pone.0161561 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Ingiosi AM, Opp MR, Krueger JM. Sleep and immune function: glial contributions and consequences of aging. Curr Opin Neurobiol 2013;23:806–11. doi: 10.1016/j.conb.2013.02.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Jessen NA, Munk AS, Lundgaard I, Nedergaard M. The glymphatic system: a beginner's guide. Neurochem Res 2015;40:2583–99. DOI 10.1007/s11064-015-1581-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sudlow C, Gallacher J, Allen N, Bera V, Burton P, Danesh J, Downey P. et al. UK Biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med 2015;12(3): e1001779. doi: 10.1371/journal.pmed.1001779 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cote DJ, Downer MK, Smith TR, Smith-Warner SA, Egan KM, Stampfer MJ. Height, waist circumference, body mass index, and body somatotype across the life course and risk of glioma. Cancer Causes Control 2018;29:707–19. doi: 10.1007/s10552-018-1052-x. [DOI] [PMC free article] [PubMed] [Google Scholar]