TL;DR
This study investigates the link between pre-diagnosis sleep patterns and overall survival (OS) in ovarian cancer (OC) patients. It found that longer sleep durations and good sleep quality are associated with better survival, while later bedtimes, poor sleep quality, and disrupted sleep patterns are linked to worse outcomes, suggesting sleep could be a modifiable factor in OC prognosis.
What Problem Does This Paper Address?
Ovarian cancer (OC) remains a significant cause of mortality in women, often diagnosed at advanced stages due to non-specific symptoms. While survival rates have improved, they still lag behind other cancers. Existing research points to various modifiable lifestyle factors influencing cancer outcomes, but the specific role of pre-diagnosis sleep status on OC survival has been largely unexplored. This paper aims to fill this knowledge gap by examining whether sleep habits before an OC diagnosis can predict survival.
Methods
The researchers conducted a prospective cohort study involving 853 newly diagnosed OC patients in China between 2015 and 2020. Sleep status was assessed using the Pittsburgh Sleep Quality Index (PSQI), a validated questionnaire covering subjective sleep quality, latency, duration, efficiency, disturbance, medication use, and daytime dysfunction. Patients' vital status was tracked through active follow-up and linkage to medical records and cancer registries.
The core analytical approach was the Cox proportional hazards regression model. This statistical method allowed the researchers to estimate hazard ratios (HRs) and their 95% confidence intervals (CIs) to quantify the association between various pre-diagnosis sleep parameters (e.g., bedtime, wake-up time, sleep duration, sleep quality, sleep patterns) and overall survival (OS). They performed crude analyses and then adjusted for a comprehensive list of potential confounders, including age, BMI, physical activity, electronic product use, smoking, alcohol drinking, education, family income, menopausal status, parity, tumor characteristics (histological type, grade, FIGO stage, residual lesions), and comorbidities. Interaction effects between sleep parameters and certain clinical factors were also explored, as was the non-linear association between PSQI score and OS using restricted cubic splines.
Datasets
The study utilized data from a prospective cohort of 853 newly diagnosed ovarian cancer patients recruited from Shengjing Hospital of China Medical University. This dataset was collected between January 2015 and December 2020. The data included demographic information, clinical characteristics, lifestyle factors (collected via self-administered questionnaires), and sleep status measured by the PSQI. Vital status was obtained through active follow-up and linkage to medical records and cancer registries. No new datasets were introduced.
Key Insights and Ideas
- Sleep as a Prognostic Indicator: The study strongly suggests that pre-diagnosis sleep habits are not just a symptom of illness but can be an independent predictor of survival in ovarian cancer patients. This opens up possibilities for interventions.
- The "Sweet Spot" for Sleep: It's not just about sleeping more or less; there appears to be an optimal range. Longer night sleep (≥7.5 hours) and total sleep (≥8 hours) were associated with better survival, implying that insufficient sleep is detrimental.
- Timing Matters: Sleeping later (after 10 PM) was linked to worse outcomes, suggesting that adherence to a more "normal" circadian rhythm might be beneficial.
- Quality Over Quantity (Sometimes): Poor sleep quality, as measured by the PSQI, was a significant predictor of worse survival, even more so than some specific timing or duration metrics.
- Complex Interactions: The study revealed that the impact of sleep on survival isn't uniform. It can be modified by factors like residual tumor size and menopausal status, indicating a complex interplay between sleep, cancer biology, and patient characteristics.
- Curvilinear Relationship with PSQI: The PSQI score showed a non-linear association with OS, with the worst survival occurring at a specific PSQI score range (around 7.7-7.8), suggesting that both very poor and moderately poor sleep might have different implications.
Key Results
- Later Bedtime: Sleeping after 22:00 was associated with a 2.13-fold increased hazard of death (HR: 2.13, 95% CI: 1.42–3.18).
- Poor Sleep Quality: Patients with poor sleep quality (PSQI score > 5) had a 2.43-fold increased hazard of death compared to those with good sleep quality (HR: 2.43, 95% CI: 1.64–3.62).
- Bad Sleep Patterns: Late bed-early rise and late bed-late rise patterns were associated with significantly worse OS (HRs of 2.26 and 1.93, respectively) compared to early bed-early rise.
- Longer Sleep Duration: Longer night sleep duration (≥7.5 h/day) was associated with a reduced hazard of death (HR: 0.40, 95% CI: 0.24–0.67), and longer total sleep duration (≥8 h/day) also showed a protective effect (HR: 0.53, 95% CI: 0.29–0.98).
- Interaction Effects: Significant interactions were found between residual lesions and wake-up time, night bedtime, and sleep pattern. Total sleep duration also interacted significantly with menopausal status and parity.
- Curvilinear PSQI: A significant curvilinear association was observed between PSQI score and OS (p < 0.05), with the worst OS at a PSQI score of 7.7 to 7.8.
Limitations and Future Work
The authors acknowledge several limitations:
- Self-Reported Data: Sleep status was collected via self-report, which can introduce recall bias and social desirability bias.
- Short-Term Sleep Assessment: The study focused on sleep status in the month before diagnosis, which may not fully represent long-term sleep habits.
- Pre-Diagnosis Focus: The study only examined pre-diagnosis sleep. Future research should investigate post-diagnosis sleep and its impact, as well as the differences between pre- and post-diagnosis sleep.
- Progression-Free Survival (PFS): PFS was not explored, though the authors suggest it might be similar to OS given the high mortality of OC.
- Residual Confounding: Despite extensive adjustments, the possibility of unknown or unmeasured confounding factors (e.g., pre-diagnosis pain levels) remains.
- Generalizability: Sleep habits can vary geographically, so findings should be cautiously extrapolated to other regions.
Future work should include investigating post-diagnosis sleep, exploring the impact of pain on sleep and survival, and conducting larger studies with longer follow-up periods to confirm these findings and explore interventions.
Summary updated