MINNEAPOLIS – People who have the type of insomnia characterized by a sharply shortened duration of sleep are at increased risk for cancer, a longitudinal cohort study showed.
In the study of more than 1,600 adults from the general population, those who reported insomnia and slept 5 hours or less per night as determined by polysomnography had more than double the adjusted cancer risk of their insomnia-free counterparts who slept longer. But the association was no longer significant after depression was controlled for.
Dr. Julio Fernandez-Mendoza
"Insomnia with severe short sleep duration is associated with increased risk of cancer, particularly in those with comorbid depression," commented first author Julio Fernandez-Mendoza, Ph.D., of the sleep research and treatment center, department of psychiatry, Penn State College of Medicine, Hershey.
Previous research has established a dose-response relationship between objectively measured sleep duration and other adverse health outcomes, he noted. "For us, basically, objective sleep duration is a biomarker, is an assay, is the best we have right now. ... These findings expand on our previous studies, and it appears that we can continue using this assay to explore the medical morbidity associated with this insomnia phenotype."
In an interview, session cochair Dr. Ruth M. Benca, director of the center for sleep medicine and sleep research at the University of Wisconsin–Madison, commented, "The whole connection between sleep and cancer has now come to the fore with some of the recent studies showing, for example, that sleep apnea seems to be a risk factor for the ultimate development of cancer. And these new data suggest that insomnia, or insomnia and depression, may also play a role. We need more mechanistic studies to understand how those links may work."
The picture is complicated by overlaps between apnea and insomnia, she noted. "People with apnea can have high rates of insomnia, and both insomnia and apnea can be associated with fragmented sleep or insufficient sleep. So is it the insufficient sleep that’s a problem? Do hypoxemia and apnea also contribute? There are some animal studies that suggest that hypoxemia is related to cancer progression."
In the study, the investigators analyzed data from 1,620 individuals in the Penn State cohort who had no history of cancer at baseline. Insomnia was defined as self-reported insomnia present for at least 1 year, and very short sleep duration was defined as 5 hours or less as determined by polysomnography.
After a follow-up of about 15 years, 12.3% of the individuals experienced incident cancer, defined as a cancer diagnosis or death from the disease.
In an analysis adjusted for traditional confounders (sex, age, race, apnea-hypopnea index, body mass index, diabetes, and hypertension), relative to noninsomniacs who slept more than 5 hours nightly, insomniacs who slept 5 hours or less had significant 2.73-fold higher odds of incident cancer.
However, the association was no longer significant after additional adjustment for depression. "This makes sense because we do know very well two things: the strong association of depression with cancer, and second, the strong association of insomnia with depression. They have a lot in common, particularly inflammation. They have in common fatigue also," Dr. Fernandez-Mendoza said at the annual meeting of the Associated Professional Sleep Societies.
Similarly, the association was not significant after additional adjustment for smoking and alcohol use. "That was primarily driven by something that we learned from our natural history papers: Because these are basically behavioral factors, many insomniacs stop smoking or stop using so much alcohol, just related to the sleep hygiene thing," he commented.
The investigators have not yet assessed whether insomnia with very short sleep duration is associated with specific types of cancer, according to Dr. Fernandez-Mendoza.
Of note, insomniacs who slept more than 5 hours did not have elevated odds of cancer. Nor did noninsomniacs who slept 5 hours or less.
Dr. Fernandez-Mendoza disclosed no relevant conflicts of interest.