Study Links Healthy Sleep Patterns to Lower Risk Across Several Cancers

Sleep may leave a measurable biological trace linked to cancer risk. Research presented at Digestive Disease Week 2026 connected healthier sleep patterns with lower risks of several cancers, and identified a plasma protein signature that may help explain the association. The findings point to sleep as a set of related habits and symptoms, rather than simply a nightly hour count. That distinction matters because a person can spend enough time in bed and still sleep poorly, or feel tired during the day despite a schedule that looks regular on paper. The study gives researchers a way to examine those differences alongside biological measurements, though it doesn’t show that changing sleep habits will change cancer risk.

Two ways to measure sleep

Researchers analyzed data from 472,105 UK Biobank participants, scoring five features: sleep duration, chronotype, insomnia, snoring, and daytime sleepiness. Adults who typically slept seven to nine hours, followed a schedule suited to their natural sleep-wake preference, and reported fewer sleep problems scored better. Participants were grouped into healthy sleep (four or more points), intermediate sleep (two to three), and poor sleep (one or fewer). The score combines reported behaviors and symptoms, so it offers a broader picture than duration alone, while still depending on how participants described their own sleep.

The team also examined plasma protein measurements from 52,920 participants. From those data, researchers built a score based on 303 proteins associated with sleep patterns. That second measure produced some of the study’s strongest cancer risk associations. It’s a different kind of evidence from the five-part sleep score: rather than relying only on reported habits, it looks for a biological pattern that tracks with them. But a protein signature can reflect many processes in the body, and its presence doesn’t by itself explain which processes matter or how they relate to cancer.

Using both measures lets the researchers ask related but distinct questions: whether reported sleep characteristics correlate with cancer outcomes, and whether a protein pattern linked to sleep shows a similar relationship. That design can help generate hypotheses about possible mechanisms. It can’t settle them. The protein analysis covered fewer participants than the broader sleep analysis, and the reported associations don’t establish that the signature is specific to sleep or to any one cancer.

The cancer associations

Compared with people in the poor-sleep group, those with healthy sleep had a 36% lower risk of liver cancer and a 14% lower risk of lung cancer. The difference was larger for liver cancer than for lung cancer, and the results don’t mean that every person with healthier sleep will have lower risk. They describe group-level associations in this study. That matters when interpreting percentages: a relative difference can sound decisive without telling a reader the underlying absolute risk.

The protein score showed larger differences between participants in its highest and lowest thirds. The high-score group had lower risks of:

• Liver cancer: 71%
• Stomach cancer: 54%
• Pancreatic cancer: 53%
• Gallbladder cancer: 54%
• Esophageal cancer: 35%
• Colorectal cancer: 28%

Those figures make the protein results striking, particularly for liver cancer. They also call for care: a comparison between the highest and lowest thirds of a score isn’t the same as showing that sleep itself reduced risk by those amounts. The score reflects proteins associated with sleep patterns, and other factors could contribute to both the protein profile and cancer risk. The study’s figures identify patterns for further investigation; they don’t tell an individual what their personal risk is.

These results are associations, not proof that better sleep prevents cancer. The protein findings offer clues to possible biological links, but the study doesn’t establish what causes the differences. Similar observational analyses can reveal that two measures vary together, while leaving open whether one contributes to the other or whether both are shaped by other factors. Here’s the thing: the size of an association can make a finding newsworthy, but it doesn’t remove that basic uncertainty.

Why the five-part score matters

A person’s sleep can look adequate by duration and still be disrupted by insomnia, snoring, or daytime fatigue. The study’s framework captures that distinction: sleep health reflects timing and how restorative sleep feels, as well as time spent in bed. Chronotype adds another dimension, since a schedule that fits one person’s natural preference may not fit another’s. The score doesn’t make these features interchangeable; it gives researchers a way to consider them together.

This is more significant than it looks for research design. If studies measure sleep only by hours, they may miss symptoms and timing that could help explain why people with similar durations have different sleep experiences. A combined score can make those patterns easier to compare, although it also compresses several different experiences into a single category. Someone classified as having poor sleep because of one feature may have a different profile from someone whose score reflects several problems.

For people looking to improve their sleep patterns, practical steps include getting morning light, keeping regular bed and wake times, limiting alcohol before bed, and reducing evening screen exposure. Persistent insomnia, loud snoring, or pronounced daytime sleepiness may warrant a discussion with a doctor, since these symptoms can point to sleep disorders. If you're working on your own sleep, the study supports paying attention to more than the clock, but it doesn’t show that these steps prevent cancer.

And yet, the symptoms themselves may deserve attention even apart from cancer research. A persistent pattern of poor sleep can affect daily functioning, and the study’s inclusion of snoring and daytime sleepiness reflects how sleep problems show up beyond bedtime. The useful signal is the combination of timing, duration, and symptoms (and this is the part most people overlook). Not a single number.

What the findings mean next

The clearest takeaway is that sleep quality has several dimensions worth tracking. The 303-protein signature gives researchers a starting point for studying how those dimensions may connect to cancer risk; further research will be needed to test the relationship. Researchers will need to determine whether the signature consistently tracks with sleep patterns and whether it helps explain the observed cancer associations. The current findings don’t answer those questions.

For now, the results are best read as a map of possible connections rather than a prevention claim. The sleep score points to features that can be measured in everyday life, while the protein score offers a biological avenue to investigate. If future work clarifies how those measures relate, it could sharpen the questions scientists ask about sleep and cancer. The evidence presented here is an early lead, not a clinical rule.