Women who don't eat for at least 13 hours overnight after surviving breast cancer are at less risk of the disease returning, according to new research.
The study shows that those who didn't fast for 13 hours were at 36 per cent higher risk of breast cancer recurring and at 21 per cent higher probability of dying from the disease.
And there was a 22 per cent higher risk of dying from any cause among patients with breast cancer who fasted for shorter periods compared to those who fasted for 13 hours or more overnight.
Fewer fasting hours per night was associated with significantly less sleep and higher levels of glycated hemoglobin (HbA1c) - a measure of average blood sugar levels over a period of months.
Researchers say their findings are relevant to cancer prevention and control because elevated HbA1c and poor sleeping habits have been linked to an increased risk of breast cancer.
A previous study found shorter overnight fasts were associated with worse blood sugar control.
And mice experiments found prolonged fasting during sleep can protect those fed a high-fat diet against abnormal glucose metabolism, inflammation and weight gain, all of which are associated with poor cancer outcomes.
The researchers suggested not having a late dinner or early breakfast could help stop other cancers returning too.
Catherine Marinac, a doctoral candidate at University of California, San Diego School of Medicine, said: "Prolonging the overnight fasting interval may be a simple, non-pharmacological strategy for reducing a person's risk of breast cancer recurrence and even other cancers.
"Previous research has focused on what to eat for cancer prevention, but when we eat may also matter because it appears to affect metabolic health."
The study involved 2,413 women with early-stage breast cancer and without diabetes who were 27 to 70 at diagnosis and part of the Women's Healthy Eating and Living study between 1995 and 2007.
The research, published by JAMA Oncology, looked at invasive breast cancer recurrence and new primary breast tumours during an average of 7.3 years of follow-up, as well as death from breast cancer or any cause during an average 11.4 years of surveillance.
Women in the study were an average age of 52.4 and had an average fasting duration of 12.5 hours per night.
Other analysis indicates that each two-hour increase in nightly fasting was associated with lower hemoglobin A1c levels and a longer duration of nighttime sleeping.
Professor Dr Ruth Patterson added: "If future trials confirm that habitual prolonged nightly fasting improves metabolic health, this would be an important discovery in prevention that could reduce the risk of cancers, type 2 diabetes, and cardiovascular disease."