June 26, 2019

expert reaction to study on sleep traits and breast cancer

Research published in The BMJ shows there is consistent evidence for a protective effect of morning preference and suggestive evidence for an adverse effect of increased sleep duration on breast cancer risk.

Prof Chris Bunce, Professor of Translational Cancer Biology, University of Birmingham, said:

“This study is impressive in its scale. However, the associations observed are very small correlative. It is dangerous to suggest, even unintentionally, to women that changing their sleep patterns will significantly alter their risk of breast cancer.

“The authors acknowledge that a limitation of their study is that the groups in the study bias the analysis to women of European ancestry. However, they do not consider that women included in the UK Biobank and BCAC are themselves a biased sample of women from the UK; first by virtue have having been eligible and second having volunteered to these studies.

“The study reports that of the women with an exclusive diagnosis of breast cancer in the UK biobank, approx. 60% were prevalent cases, which I take to mean had breast cancer at the time they entered the study and approx. 40% developed breast cancer after entering the study, half of whom developed cancer within just three years of entering the study. If this is correct, then most of the women either had cancer at the time of the study or probably had undetected early disease. In this case self-reported sleep patterns and other non-genetic measures taken at time of study entry may not be the same as prior to developing cancer which is when the primary risk occurs.”

Prof Justin Stebbing, NIHR Research Professor of Cancer Medicine and Medical Oncology, Imperial College London, said:

“We all often need or always need a good night’s sleep and a new study has related our sleep preferences to developing breast cancer in some people. In 2007 the WHO classified shift work that involves disruption to our natural body clocks or rhythms as being probably carcinogenic to humans and there are lots of biologic reasons this may be the case as a bad night’s sleep for example can affect our immune system and in fact disturbed sleep can affect us in all sorts of ways.

“This study looking at over 100,000 women with breast cancer (and over 100,000 women without cancer) used very interesting statistical methods to work out that there was consistent evidence for a protective effect of morning preference, meaning people who considered themselves as a ‘morning person’, and it perhaps also suggested that the longer we slept for, the greater the risk of breast cancer. It also showed that the evidence for insomnia symptoms and breast cancer risk was inconclusive. Like all studies limitations included the self-reported nature of the results. These findings have potential implications for influencing sleep habits of the general population to improve health, but the study highlighted that we’re all different genetically speaking and this needs to be kept in mind too.”

Dr Luca Magnani, Senior Research Fellow in the Department of Surgery & Cancer at Imperial College London, said: 

 “This is an interesting study where the researchers have identified a link between a genetic variant (genetic variants associated with three particular sleep traits) and breast cancer risk – which is significant, but a very small effect size.

“It is important to note that these data do not suggest in any way that modifying sleep habits could eventually lead to a decrease in the risk of breast cancer (there is no association with sleep habits or insomnia).

“What they suggest is that it appears that the risk of breast cancer is associated with a genetic (thus not modifiable) trait that is in itself associated with a “morning” or “night” preference – what we call ‘larks’ and ‘owls’. This being a correlation, it is likely that these genetic traits are linked to breast cancer risk though a different (and more precise) mechanism.”

Dr John O’Neill, Research Group Leader, MRC Laboratory of Molecular Biology (LMB), said:

“The press release says: ‘an association that is observed using Mendelian randomisation strengthens the inference of a causal relationship’. It is still just a correlation, it does not matter how sophisticated the analysis method, causation cannot be proven by post-hoc analysis.

“Also <1% difference is a tiny effect size. I would tend to make the opposite interpretation to their press release, that having an evening chronotype has very, very little bearing on the risk of breast cancer.”

‘Investigating causal relations between sleep traits and risk of breast cancer in women: mendelian randomisation study’ by Richmond et al. was published in The BMJ at 23:30 UK time on Wednesday 26 June.

Declared interests

None received.