Sleep apnea and other types of sleep disorders appear to elevate the risk for some types of cancer, specifically prostate cancer, more so than others. But the overall risk can be highly variable, and some sleep problems were found to be associated with a lower risk for cancer and cancer-related death, an analysis of a large observational cohort study of cardiovascular patients found.
Results of the analysis were published online in the journal Cancer Epidemiology. Investigators analyzed the presence of sleep apnea and insomnia and cancer risk in more than 8,500 patients in the Cardiovascular Health Study (CHS). “The fact that we observed certain sleep problems, like apneas, to be associated with elevated risk of some cancers but not others reflects the fact that cancer is a heterogeneous disease,” senior author Amanda Phipps, PhD, said in an interview. Dr. Phipps is an associate professor of epidemiology at the University of Washington and the Fred Hutchinson Cancer Research Center, both in Seattle.
Variable cancer links
The researchers assessed sleep problems in two groups in the CHS: an incident cancer group of 3,930 patients and a cancer mortality group of 4,580 patients. Within those respective groups, the investigators identified 885 first-incident cancers and 804 cancer deaths with a median follow-up of 12 and 14 years. The average age of the study population was 73 years, and 57% were women.
Sleep apnea symptoms (SAS) were associated with a lower risk for incident cancers – a 16% lower baseline risk and a 24% lower time-dependent risk. The study showed no association between cancer incidence and daytime sleepiness and apneas.
However, there was a significantly elevated risk relationship between sleep problems and prostate cancer. A time-dependent analysis of apnea showed more than double the risk (hazard ratio, 2.34), and baseline snoring carried a 69% greater risk. There was also a dose-response relationship for baseline cumulative SAS, compared with not having symptoms: an HR of 1.30 for one symptom, and 2.22 for two or more symptoms.
Risks for lymphatic or hematopoietic cancers were also associated with baseline daytime sleepiness (HR, 1.81), but not with insomnia (HR, 0.54).
With regard to cancer mortality, the study found no relationship between sleep problems and cancer death. In fact, it found an overall inverse relationship with snoring (time-dependent HR, 0.73; cumulative average HR, 0.67) and baseline apnea (HR, 0.69). Likewise, patients reporting SAS had lower risks than those having no SAS: an HR of 0.90 for one symptom and 0.75 for multiple symptoms. No relationships were found between any insomnia symptom and cancer death.
“We know the pathways that lead to prostate cancer can be very different than the pathways that lead to colorectal cancer,” Dr. Phipps said. “What we don’t yet understand is why these associations differ or what mechanisms are responsible for these cancer site-specific associations.”
Need for sleep assessment
The findings don’t change much for how clinicians should evaluate cancer risks in patients with sleep problems, Dr. Phipps said. “Other studies have clearly demonstrated the implications that sleep apnea has for a variety of other important health conditions – such as cardiovascular disease – so there are already plenty of good reasons for clinicians to ask their patients about their sleep and to connect patients with resources for the diagnosis and treatment of sleep apnea,” she added. “This study provides another possible reason.”
These findings provide context for future studies of the relationship between sleep problems and cancer. “But, given that sleep is something we all do and given that sleep problems are so pervasive, it’s important that we keep trying to better understand this relationship,” Dr. Phipps said.
“My hope is that future cancer studies will build in more detailed, longitudinal information on sleep patterns to help us fill current gaps in knowledge.”
Dr. Phipps has disclosed no relevant financial relationships.
A version of this article first appeared on Medscape.com.