We included case studies and cohort studies with the following characteristics: (i) reported on breast cancer incidence or breast cancer adverse progression or breast cancer mortality; (ii) measured at least one sleep trait (sleep quality or sleep duration). We excluded studies that were published in languages other than English, those for which full text was not available, or that consisted of narrative reviews, comments or letters.

We followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines [Transparent Report of Systematic Reviews and Meta-Analysis (Moher et al. 2009)] as a methodological template for this review (Figure 1). Four databases (PubMed, EMBASE, Web of Science and Cochrane library) were searched to identify papers examining the association between sleep traits and the risk of breast cancer disease incidence, adverse progression and mortality. The effects of low quality sleep, sleep duration <6 h and sleep duration >9 h on the breast cancer risk were extracted. Low quality sleep includes terms such as sleep difficulties, insomnia, and sleep disorders. And measurement tools used include such as the Pittsburgh Sleep Quality Index, polysomnography, and self-report measures. The detailed search strategies are included in the Supplementary Material (Supplementary Table S1–S4). PubMed searches were conducted using a combination of Medical Subject Headings (MeSHs) and Keywords (in title or abstract.) EMBASE was searched using EMTREE and title/abstract search. Web of Science and Cochrane library were searched using keywords. The evolving approach to breast cancer control is demonstrated by the high frequency of updates to clinical practice guidelines [34], where older studies may no longer reflect current best practices or technical standards [35]. The role of lifestyle factors, including sleep, on breast cancer-related health indicators can be influenced by changes in clinical treatment modalities [36], creating new treatment-lifestyle synergies or antagonisms [37], and there is a particular need to dynamically assess the amount of lifestyle factor effects in different treatment contexts [38]. At the same time, we consider that studies in the last decade have typically adopted more rigorous study designs, statistical methods, and reporting standards, with higher quality literature [39]. Ultimately, database searching was carried out from 2014 to 2024.

Two independent co-authors (JY and YB) screened titles and abstracts and then screened the full text of the identified articles against predefined eligibility criteria. Our search retrieved a total of 3,225 articles from PubMed, EMBASE, Web of Science and Cochrane library. After manually removing 1,688 duplicates, we were left with a total of 1,537 unique references. Of the 1,537 screened articles, 1,464 studies - that did not report sleep traits, reported measures of sleep traits other than those included in our definition, and did not report either incidence or progression were excluded. We obtained full text for all 73 remaining articles, and after review, we additionally excluded 39 articles that did not report breast cancer-specific outcomes or were themselves systematic reviews. The remaining 34 articles were included in this review. Author disagreements were resolved through discussion.

The risk of bias in the studies included in the review was rated independently by Y.B. and J.Y., following the same consensus procedure employed for study selection. The Newcastle-Ottawa Scale (NOS) quality assessment tool [40] was used for assessing risk of bias. A “star system” has been developed in which a study is judged on three broad perspectives: the selection of the study groups; the comparability of the groups; and the ascertainment of either the exposure or outcome of interest for case-control or cohort studies respectively. Studies with a score of 6 or higher were considered as high-quality research [21]. A summary figure of the assessed bias of the included studies was created. See Supplementary Material (Supplementary Table S5). We employed funnel plots to assess the potential bias of publication included. (Supplementary Figure S1)

Data were independently extracted by three co-authors (Y.B and J.Y), capturing Author/Year, Study Type and Period, Sample Amount, Outcomes, Sleep Traits, odds ratio or rate ratio or hazard ratio (OR/RR/HR) indicator, corresponding confidence intervals (CI), confounding factors adjusted and main conclusion. See Supplementary Material (Supplementary Table S6). OR, RR and HR are clinically and statistically similar in that they are all “relative effect sizes,” reflecting the risk of the exposure/intervention group compared with the control group, and, although the mathematical definitions differ, OR ≈ RR ≈ HR at low rates of outcome events (<10%) [41]. Also, it is possible to combine different effect sizes when study clinical homogeneity is high (e.g., similar interventions, populations, outcome definitions) [42]. Therefore, we harmonized and merged the three effect sizes. When multiple measures of association were reported, we reported results from the fully adjusted model. A fixed or randomized effect model was applied according to the heterogeneity. Forest plots [43] are used to display the results of the merge. Publication bias was assessed by funnel plots. All statistical analyses were carried out in Review Manager v.5.3 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration 2014).

The included studies consisted of 12 cohort studies and 11 case-control studies investigating breast cancer incidence risk, and 10 cohort studies and 1 case-control study examining prognostic outcomes. All cohort investigations (n = 22) established follow-up periods extending beyond 1 year, providing sufficient timeframes for observing relevant breast cancer outcomes (Table 1). The confounding factors primarily controlled for in these studies included demographic factors (race, age, income, education, reproductive age); health status (BMI, depression, family history of breast cancer, history of other cancers); and risk behaviors (smoking, alcohol consumption, diet, medication use, physical activity). All included studies were assessed to be of high quality using Newcastle-Ottawa Scale (NOS) quality assessment tools (Supplementary Table S5). Inspection of the funnel plot demonstrated no substantial evidence of publication bias in our meta-analysis. (Supplementary Figure S1) These contemporary studies, with their scientific designs and diverse populations, provide a robust foundation for evaluating the impact of sleep quality on breast cancer incidence and prognostic progression. Among the literature included, fifteen studies (65.22%) indicated that sleep quality and duration are associated with increased risk of breast cancer incidence, whereas eight studies concluded evidence did not support the relationship. In terms of adverse progression, four studies (66.67%) suggested that sleep quality and duration are correlated with breast cancer progression; four studies (50%) demonstrated that compromised sleep quality and duration are associated with elevated breast cancer mortality rates (Supplementary Table S6).

Figure 2 shows different subgroup analyses of the association between sleep traits and breast cancer incidence risk. Fourteen studies were pooled for the subgroup where the sleep trait was estimation of low quality sleep (N = 1,507,688). The result indicated that low quality sleep increased breast cancer incidence risk (OR = 1.09, 95% CI 1.05–1.13, p < 0.00001, I ² = 40%). Thirteen studies (N = 1,022,423) found no significant effect of sleepiness <6 h on breast cancer risk (OR = 1.01, 95% CI 0.97–1.05, p = 0.60, I ² = 2%). Likewise, 13 studies (N = 1,084,028) found that drowsiness, quantified as sleep duration >9 h, did not have a significant effect on breast cancer risk (OR = 1.03, 95% CI 0.99–1.07, p = 0.13, I ² = 0%) (Figure 2).

Figure 3 shows different subgroup analyses of the association between sleep traits and breast cancer progression (adverse progression and mortality). In the analysis of progression risk, seven studies (N = 115,428) found that low quality sleep significantly increased the risk of breast cancer progression (OR = 1.54, 95%CI 1.50–1.58, p < 0.00001, I ² = 90%). Six studies examining sleep duration <6 h (N = 13,394) showed no significant effect on the risk of breast cancer progression (OR = 0.99, 95%CI 0.97–1.00, p = 0.13, I ² = 21%). Similarly, five studies focusing on sleep duration >9 h (N = 13,297) found that hypersomnia increased the risk of breast cancer progression (OR = 1.20, 95%CI 0.97–1.49, p = 0.09, I ² = 0%) (Figure 3).

In the analysis of subgroup of adverse progression, five studiesresearches (N = 105,806) pooled indicated that low quality sleep increased the risk of breast cancer adverse progression (OR = 1.55, 95%CI 1.51–1.59, p < 0.00001, I ² = 93%). However, sleep duration showed nonsignificant effect on the risk of breast cancer adverse progression, with three studies (N = 8,798) estimating sleep duration <6 h (OR = 0.97, 95%CI 0.83–1.13, p = 0.66, I ² = 0%) and three studies (N = 8,798) estimating sleep duration >9 h (OR = 1.07, 95%CI 0.82–1.00, p = 0.60, I ² = 32%) (Figure 3).

In the analysis of subgroup of breast cancer-specific mortality, three studies (N = 13,841) pooled indicated that low quality sleep increased breast cancer-specific mortality (OR = 1.21, 95%CI 1.01–1.45, p = 0.04, I ² = 40%). Furthermore, the combined results of 4 studies (N = 7,643) showed nonsignificant outcome for sleep duration <6 h (OR = 1.21, 95%CI 1.01–1.45, p = 0.04, I ² = 40%). On the contrary, pooled results from three studies found that sleep duration >9 h (N = 7,546) could increase breast cancer-specific mortality (OR = 1.45, 95%CI 1.02–2.04, p = 0.04, I ² = 0%) (Figure 3).

Low quality sleep significantly increases the incidence and mortality risk of breast cancer, further validating those circadian rhythms, melatonin significantly influences tumor susceptibility and progression. For the occurrence of breast cancer, since the quality of sleep is closely related to the body’s immune system function, inflammatory response, hormone levels, lifestyle, etc., and low quality of sleep itself has been proven to be positively correlated with the long-term risk of various types of cancers, the result that low quality of sleep increases the incidence of breast cancer is to be expected [73, 74]. However, it has to be emphasized that although the effect of sleep quality on breast cancer incidence is significant, the estimated effect sizes are relatively small and do not elucidate the strong evidence for sleep quality on breast cancer risk.

For the progression of breast cancer, insufficient levels of melatonin secreted from the pineal gland in the dark may play a role in the association between sleep and breast cancer aggressiveness [68]. Low sleep quality, while not entirely representative of nighttime light and melatonin levels, may be directly related to melatonin levels, so melatonin can be listed as a possible key cause. There’s also the claim that sleep fragmentation leads to increased inflammatory cytokine production and natural killer cell dysfunction, potentially compromising immune surveillance against cancer progression [75]. This immune dysregulation creates a microenvironment conducive to tumor growth and metastatic spread. The clinical significance of these mechanisms is evident in prospective studies, where poor sleep efficiency (below 85%) has been associated with dramatically shorter survival times in women with advanced breast cancer (33.2 months versus 68.9 months) [31]. At the same time, studies have also pointed out that low quality of sleep can trigger negative attitudes towards death [76] and emotions such as depression and anxiety [77], which are detrimental to the health of breast cancer patients. The effect value of sleep quality on the progression impact of breast cancer is relatively large compared to the risk of morbidity. This suggests that we should pay more attention to sleep quality in breast cancer patients and emphasize the importance of sleep quality in the health management of this population.

From a clinical perspective, these findings support incorporating routine sleep quality assessments into oncology practice and providing targeted interventions such as cognitive behavioral therapy for insomnia to potentially improve patient prognosis [78].

The available evidence suggests that sleep duration has no significant effect on breast cancer incidence and only suggests that sleep duration >9 h has a detrimental effect on the risk of breast cancer-specific mortality. Studies have shown that the metastatic spread of breast cancer is accelerated during sleep [79], that this metastatic spread occurs via circulating tumor cells (CTCs), and that resting periods are highly susceptible to metastasis, which may provide a rationale for the high rate of breast cancer-specific mortality associated with prolonged sleep. At the same time, the melatonin hypothesis, which suggests that shorter sleep is associated with lower melatonin levels, and the fact that melatonin is known to modulate susceptibility to cancer and has antiproliferative activity, suggests that the link between longer sleep and breast cancer is biologically plausible [80]. Some scholars have also suggested that excessive sleep may lead to elevated levels of systemic inflammation and an increase in some inflammatory biomarkers, such as CRP and IL-6, which may predispose individuals to breast cancer [81]. In fact, breast cancer patients, the subjects in whom breast cancer-specific deaths occur, due to the side effects of cancer treatment [82], are often accompanied by an increase in cortical activity related to arousal and a decrease in active cortex related to sleep homeostasis [16], so that sleep duration is more likely to be shorter, and prolonged sleep is an anomalous event in itself in this group. At the same time, it is important to note that the effect of prolonged sleep on breast cancer-specific mortality, while significant, only included three studies and did not provide strong evidence.

Overall, there is still controversy about the effect of sleep duration on breast cancer risk. Findings from a large multi-ethnic cohort study [60] suggest that both short and long sleep are associated with a higher risk of breast cancer incidence compared to normal sleep. According to the results of the Million Women Study [63], the overall prospective evidence does not support an association between sleep duration and breast cancer incidence risk. We consider that there is still a paucity of high-quality studies on sleep duration and the disease process in breast cancer, making it difficult to draw uniform conclusions. Therefore, more research is needed to validate the effects of short and long sleep duration on breast cancer development and progression, pinpointing the exact biological processes also remains elusive, necessitating additional investigation. Future studies should prioritize prospective designs with objective sleep measurements (e.g., actigraphy or polysomnography) to minimize recall bias [83], while incorporating molecular biomarkers such as circulating inflammatory cytokines, melatonin metabolites, and circadian gene expression profiles to elucidate whether the U-shaped association reflects direct causal mechanisms or underlying health conditions [84, 85]. Additionally, intervention studies examining whether sleep optimization through behavioral or pharmacological approaches can modify cancer trajectories would provide critical evidence for establishing sleep duration as a targetable risk factor in breast cancer prevention and management strategies [75].

Nevertheless, clinicians should be aware that excessive sleep duration (>9 h) may warrant further evaluation for underlying conditions such as depression, excessive fatigue, or disease progression that could contribute to poor outcomes [86].

Our study has several limitations. First, this study only analyzed the effects of sleep quality and sleep duration on breast cancer, and did not consider the effects of other sleep traits, such as daytime naps, nighttime lights, sleep preference type, nighttime awakenings, and sleep medication use, due to data limitations. Second, most of the studies focused on two regions, China and the United States, which may affect the representativeness of the findings on a global scale. Third, we included only English-language publications, which may have biased our findings. By including only studies published in English, we may have missed important local studies that are more likely to be published in journals other than English. Finally, a limitation that we cannot ignore is that due to differences in sleep quality measurement tools, heterogeneity in assessment tools (e.g., Pittsburgh Sleep Quality Index, polysomnography, self-reported measures) may affect the comparability of results and may affect the strength of observed associations.

The available evidence points to sleep traits as primarily influencing progression in breast cancer patients and having a relatively small effect on breast cancer incidence. It can be confirmed that low quality sleep significantly increases adverse progression in breast cancer patients, suggesting that we should be concerned aboutsleep quality in breast cancer patients. Prolonged sleep may lead to breast cancer-specific mortality, but more research is needed in the future to continue to explore the impact of sleep duration and breast cancer risk.