2.1 The Molecular Clock

The mammalian eye is a complex sensory organ that contains three classes of specialized photoreceptor cells involved in light detection. One of these classes, known as intrinsically photosensitive retinal ganglion cells (ipRGCs), is pivotal in circadian rhythm photoentrainment and unlike the other classes (rods and cones), ipRGCs are not involved in image formation [21]. Instead, ipRGCs express the G protein-coupled receptor (GPCR) photopigment melanopsin, making them exquisitely photosensitive [22]. Melanopsin is most strongly stimulated by blue light (~480 nm), the dominant entraining zeitgeber in mammals, and responds minimally to longer red-light wavelengths (> 600 nm) [23]. The wavelength sensitivity spectrum of melanopsin may represent an evolutionary adaptation to Earth's natural solar cycle (i.e., predictable light patterns resulting from atmospheric dispersion of solar rays leading to shorter wavelengths ~480 nm during the day with a gradual shift to longer wavelengths > 600 nm in the evening). Thus, the differential response of melanopsin to different wavelengths of solar light translates to a greater ability to discriminate between night and day, strengthening the photoentrainment of approximately 24-h circadian rhythms [24].

Activation of melanopsin-containing ipRGCs via blue light results in the propagation of neural impulses along ipRGC axons via the retinohypothalamic tract (RHT) to the central pacemaker (a bilaterally symmetric structure known as the suprachiasmatic nucleus [SCN] located in the anterior hypothalamus) [25, 26]. As neural impulses traversing the RHT terminate in the SCN, the excitatory neurotransmitters glutamate and pituitary adenylate cyclase-activating peptide (PACAP) are released. Glutamate and PACAP then diffuse across the synapse and bind to receptors within the SCN core, stimulating Ca²⁺ and cyclic adenosine monophosphate (cAMP) influx and activation of intracellular signaling cascades, respectively [17, 22, 23]. Ultimately, photic information conveyed via the monosynaptic RHT to the SCN leads to increased expression of Period family genes that set the molecular clock [1]. Furthermore, peripheral clocks located in nearly all other cells of the body are set through a complex hormonal and neural crosstalk with the SCN molecular clock, ensuring 24-h periods of systemic physiological synchronicity [27, 28].

The SCN molecular clock produces rhythmic, 24-h expression of core canonical clock components via the interaction of multiple transcriptional-translational feedback loops [17, 26] (Figure 2). The primary feedback loop begins with the formation of circadian locomotor output cycle kaput (CLOCK) and brain and muscle ARNT-like protein 1 (BMAL1) protein heterodimers in the cytosol of SCN cells [29]. The CLOCK-BMAL1 heterodimeric complex translocates to the nucleus where it binds to specific DNA elements (E-boxes) in the promoter region of period (PER1, PER2, PER3) and cryptochrome (CRY1 and CRY2) genes, resulting in their expression [19]. PER and CRY proteins then heterodimerize, translocate to the nucleus, and repress their own transcription by acting on CLOCK-BMAL1 complexes [30]. Investigation of this feedback loop in mice has demonstrated time-of-day differences in the activation of these heterodimer complexes, such that activation of CLOCK-BMAL1 occurs in the early morning (via activation of ipRGC with blue light), leading to transcription of Per and Cry genes in the afternoon, and formation of PER-CRY complexes that represses CLOCK-BMAL1 transcription in the evening/night [28]. BMAL1 is further regulated by an interacting feedback loop in which the CLOCK-BMAL1 complex activates the expression of nuclear receptors (nuclear receptor subfamily 1 group Ds [REV-ERBs] and retinoid-related orphan receptors [RORs]) whose protein products competitively bind ROR response elements in the BMAL1 promoter. REV-ERBs repress the transcription of BMAL1 whereas RORs activate it. Together, these two interacting feedback loops form the basis of the central SCN molecular clock [29].

2.2 The Molecular Clock in Cancer

Several aspects of cell cycle progression and maintenance are regulated by core clock machinery. For example, PER1 binds Ataxia-Telangiectasia Mutated (ATM) and checkpoint kinase 2 (CHK2) to form a complex that shields p53 from ubiquitination, allowing for cell cycle arrest and DNA repair in response to radiation-induced double-strand DNA breaks (e.g., UV sun rays) [31, 32]. These data and others suggest that PER1 has important tumor-suppressing abilities [33, 34]. Indeed, attenuation of PER1 expression in murine mammary tumors leads to increased tumor cell proliferation and a shortened survival [33]. Data from human breast cancer patients also demonstrate a positive correlation between PER1 levels and overall survival [33]. Furthermore, a meta-analysis of PER1 expression in human breast cancer survivors indicates that low levels of intratumoral PER1 are associated with poorer prognoses [33]. Outside of breast cancer, PER1 has similar implications in ovarian cancer [35], gastric cancer [36], endometrial cancer [37], head and neck squamous cell carcinoma [38], oral squamous cell carcinoma [39], and non-small cell lung cancer (NSCLC) [40]. For example, NSCLC patient samples contained decreased levels of PER1 compared to normal lung tissue [40]. This has been attributed to hypermethylation in the PER1 promoter region [40], leading to attenuated expression.

PER2 also has strong implications in oncogenesis. It preserves p53 activity by preventing its ubiquitination via mouse double minute 2 (MDM2) [31, 41]. Mice homozygous for a mutation in Per2 (Per2^m/m) demonstrate increased sensitivity to tumor development compared to their WT peers following γ radiation [42]. Per2^m/m mice also lack an increase in clock gene transcription that is apparent in WT mice following irradiation [42]. Furthermore, PER2 suppresses metastasis by acting as a transcriptional corepressor of genes like twist-related protein 1 (TWIST1) and snail family transcriptional repressors (SNAIs) that are crucial for epithelial-to-mesenchymal transition (EMT) [43]. In breast cancer patient samples, hypoxic conditions lead to PER2 degradation [43], supporting the general positive correlation between intratumoral PER2 and overall survival [36, 44-48]. Together, the evidence clearly outlines critical anti-tumor properties of Period genes. These data provoke the postulation of Period genes as viable drug targets, specifically through stabilization of their expression. Indeed, RNA stability and abundance can be targeted through nucleic acid-based therapeutics [49]. One could postulate that a treatment targeting PER2 mRNA could be given alongside traditional therapies to patients with normal (i.e., non-mutated) PER2. If such a treatment were to be administered, the intrinsic rhythms (i.e., time of peak Period gene expression) of the target tissue would need to be considered. Protein modulation, however, would require extra consideration due to potential post-translational modifications.

Although there are limited data in cancer survivors, evidence suggests that clock gene disruption could play a role in disease severity (Table 1). For example, a 2021 study in elderly patients with thyroid nodules revealed aberrant expression of clock-related proteins in normal tissue surrounding malignant nodules (> 5 mm away from the nodule) [50]. Compared to tissue from healthy patients or patients with non-malignant nodules, CLOCK and BMAL1 levels were significantly increased in tissue surrounding malignant nodules. The opposite was true for CRYs (i.e., higher levels were discovered in non-malignant thyroid tissue). Of note, the authors established that poor sleep quality (assessed via the Pittsburgh Sleep Quality Index [PSQI]), sleep latency, and daytime dysfunctions (e.g., mood disturbances, lack of energy, increased errors at work, and worrying about sleep) were independent risk factors for developing malignant thyroid nodules [50]. CLOCK and BMAL1 overexpression also inhibits cell growth and stalls G1 to S phase transition in human colon cancer cells [51]. Furthermore, the same study demonstrated an increased resistance to paclitaxel in human colon cancer cells overexpressing CLOCK and BMAL1. This was concluded to be due to the prevention of cell cycle progression to the G2/M phases induced by the drug [51]. However, BMAL1 overexpression increased sensitivity to oxaliplatin in a mouse model using a human colorectal carcinoma cell line. BMAL1 expression has also been correlated with higher overall survival in patients with colorectal cancer [52], nasopharyngeal carcinoma [53], metastatic melanoma [54], and others. Taken together, these studies reveal potentially protective intrinsic clock mechanisms that have uncertain effects on drug efficacy. Further studies are needed to determine the effects of not just clock proteins, but clock dimers (i.e., PERs/CRYs and BMAL1/CLOCK) on specific anti-tumor treatments.

When clock genes are mutated, circadian control of the cell cycle can become dampened or lost. For example, a study conducted in Finnish patient samples (n = 101) concluded that CLOCK was mutated in 53% of colorectal cancer samples with microsatellite instabilities [55]. Single nucleotide polymorphisms (SNPs) may also play a role in an individual's risk of developing cancer. Certain SNPs in Period/Cryptochrome genes, BMAL1 and CLOCK have been reported to correlate with increased risk of cancer (reviewed extensively in [56]). However, the data on specific types of cancer and demographics of patients correlated to SNPs and increased risk of cancer is unclear. For instance, a meta-analysis examining publications with patient samples concluded that there is not a significant correlation between CLOCK SNPs and risk of developing breast cancer [57]. Further, a systematic review examining The Cancer Genome Atlas, Cancer Therapeutics Response Portal, and The Genomics of Drug Sensitivity in Cancer databases concluded that very few samples contain mutations in core clock genes themselves. Rather, mutations in clock-controlled genes in the cell cycle (i.e., EP300 and CREBBP) were more common [58]. Because clinical data come from patients only after they have developed cancer, it remains unclear exactly how circadian rhythm disruption alters the cell cycle on a molecular level due to the difficulty of separating the effects of core clock disruption on oncogenesis and tumor progression, and vice versa.

3.1 Circadian Disruption and Oncogenesis

One of the most common causes of circadian rhythm disruption in modern times is exposure to artificial light at night (ALAN). Since the Industrial Revolution, electric lighting has been widely implemented across the planet, causing severe light pollution in densely populated locations. Currently, around 80% of the world's population is consistently exposed to ALAN. This number increases to > 99% when only considering populations of Europe and the United States [59]. Recently, the popularity of light-emitting diodes (LEDs) has only compounded this problem. LEDs emit primarily blue light (440–495 nm) with a wavelength similar to solar blue light that cues entrainment of the central clock (i.e., the suprachiasmatic nucleus) [60]. This becomes largely problematic after sunset, when mammals have historically stopped receiving this type of photic input. Constant blue light stimuli (e.g., phones, television, and other electronic devices) after sunset dampens the rhythm of melatonin production, ablates rhythms in Period family gene expression, and suppresses rhythmic expression of other core clock genes within the hypothalamus, liver, and other peripheral tissues [18, 61]. A recent meta-analysis examining 19 studies internationally from 2001 to 2023 denoted a positive correlation between exposure to both indoor and outdoor sources of ALAN and the risk of developing breast cancer in women [62]. The authors also noted a lack of existing data sufficient enough to draw similar conclusions in other cancer types. Interestingly, one 2024 study analyzed satellite images obtained by the Visible Infrared Imaging Radiometer Suite (VIIRS), the Defense Meteorological Program Operational Linescan System (DMSP-OLS), and the International Space Station (ISS). Pixel analysis of data collected by the ISS demonstrated a 1.8× increase in the instances of colorectal cancer in participants living within the highest tertile of blue light exposure [63]. However, these results varied with the data source (i.e., VIIRS vs. DMSP-OLS vs. ISS) [63]. Similar studies using geographical ALAN exposure data and models demonstrate positive correlations between higher ALAN exposure and development of pediatric papillary cancer [64] and lung cancer [65]. However, other similar studies conclude that there is not a significant correlation between ALAN exposure and prostate cancer [66], endometrial cancer (in postmenopausal women) [67], acute lymphoblastic leukemia (in Hispanic juveniles) [68], or breast cancer [69, 70]. It is evident that there is no concrete consensus within the field, however, it appears as if the majority of studies report a significant correlation between ALAN exposure and risk of developing breast cancer in women [71]. For other cancer types, more data is needed to draw significant conclusions.

Although the most common disruptor of circadian rhythms is exposure to blue wavelength light after dark, these disturbances can also be provoked by shift work. Night shift work, defined as working outside of daylight hours (including both rotating and split shifts) [5], disrupts rhythms in core clock genes and alters melatonin secretion. This in turn dampens estrogen and aromatase-inhibiting properties of melatonin [5], leading to an increased susceptibility to reproductive cancers. In fact, suppression of rhythms in melatonin secretion is so crucial that The International Agency for Research on Cancer and the National Toxicology Program (United States) have classified night shift work as a potential carcinogen [5]. Epidemiological studies have also revealed an increased risk of breast cancer in women who work night shifts [72-74]. Indeed, a study spanning 10 years analyzed 562 female night shift nurses and reported a 23% increase in risk of breast cancer in premenopausal women who worked at least 3 days of night shift per month for 1–14 years compared to non-night shift workers. This risk increased to 30% in postmenopausal women working for ≥ 30 years [75]. Other studies have also reported an increased risk of developing breast cancer in female night shift workers [73, 74, 76], however, data from these studies overwhelmingly report only a modest increase in risk. These results have been attributed to increased exposure to light at night, which disrupts secretion of pineal melatonin [77]. Another potential contributor to developing breast cancer could be SNPs. A 2014 French study conducted in night shift nurses revealed that SNPs rs1482057 and rs12914272 in RORA were correlated with breast cancer in these women [78]. Within the group of women with breast cancer who had one of these SNPs, there was also a significant correlation with the rs11932595 mutation in CLOCK [78]. Shift work has also been correlated with an increased risk of developing endometrial [79], colorectal [72], and prostate [80, 81] cancer. Additional studies fail to present a significant correlation between extended night shift work and increased risk of developing breast [77, 82, 83] or ovarian [84-86] cancer. Likely, many other factors such as duration of night shift work [73, 75, 76, 87], disrupted sleep cycles [73], inappropriate timing of food intake [18], and genetics [88] contribute to the risk of developing breast and other cancers in night shift workers.

Another type of circadian disruption, jetlag, can be split into two categories: chronic jetlag (CJL) and social jetlag (SJL). Although limited to a very small population of people, CJL is a common model of general circadian disruption in animal research. This is typically accomplished by a phase shift (advancement or delay of intrinsic rhythms) in animals via the light cycle (i.e., lights come on 6–8 h earlier every few days) [89-91]. In mice, circadian dysfunction via CJL drives oncogenesis and metastasis of hepatocellular carcinoma [90], Lewis lung carcinoma [92], pancreatic ductal carcinoma [93], radiation-induced lymphoma [94], and breast cancer [95]. Murine models allow for testing mechanistic hypotheses like disruption of rhythms in pineal melatonin [96], elevated stemness/ability to metastasize [95, 97], and increased inflammation [98]. However, the ways in which CJL affects cancer development in humans is less understood. This is potentially due to the very specific population of people affected by CJL (i.e., pilots and flight attendants who regularly fly internationally). In flight personnel, a 2008 meta-analysis spanning 13 years and 21 studies reports a 70% increased risk of developing breast cancer and a 40% increased risk of developing prostate cancer [99]. Additional retrospective studies spanning 60 [100], 50 [101], and 42 [102] years also demonstrate that airline pilots (most of whom routinely flew international routes) have an increased risk of developing prostate cancer [100, 101] and skin cancers (melanoma, non-melanoma, and basal cell carcinoma) [100-102]. This risk increases with exposure to cosmic radiation and hours flown [100-102]. The findings of a self-reported survey of participants of the Harvard Flight Attendant Health Study suggest that female flight attendants also have an increased risk of developing both melanoma and non-melanoma skin cancers as well as breast cancer [103]. However, it is unclear if the increased risks presented in these studies are due exclusively to circadian dysfunction. It is likely that other environmental factors (i.e., radiation exposure) also provoke oncogenesis of airline pilots and flight attendants. Due to these confounds and the very limited pool of subjects (i.e., airline pilots and flight attendants) that are affected by CJL, this form of circadian disruption is not fully understood in humans. However, foundational studies in animal models clearly illustrate a connection between CJL and cancer risk [90, 94, 104] and progression [92, 93, 104-106] that is independent of cancer type.

While CJL is limited to a small population, SJL is experienced much more prevalently. Social jetlag refers to the modern tendency to wake up earlier and sleep for shorter durations on weekdays while “catching up” on sleep during weekends [107] and has increasingly been the subject of studies by circadian biologists. Approximately 50% of the current work force/students experience ≥ 2 h of SJL, and around 70% experience ≥ 1 h [108]. Particularly of interest in the modern medical landscape is how SJL impacts public health (reviewed in [109]). At least 1 h of habitual SJL has been identified as a potential contributing factor in prostate [110] and lung [111] cancer. An interesting confound of these findings comes from a 2006 study demonstrating that people who lived lifestyles involving chronic SJL were more likely to engage in behaviors that increase the risk of cancer like drinking alcohol and smoking tobacco [107]. Studies on the mechanisms of how SJL potentially contributes to or otherwise affects oncogenesis are incredibly sparse and therefore represent a largely untapped area of research potential. While it is often difficult to separate the effects of circadian disruption on oncogenesis from resulting circadian disruption following oncogenesis, the data discussed thus far suggest that initial circadian disruption plays at least a moderate role in tumor development. This should therefore be considered by physicians when evaluating an individual's risk of developing cancer.

3.2 Circadian Disruption in Cancer Survivors

So far, we have considered cancer as a consequence of circadian dysregulation, however, the reverse relationship (i.e., circadian dysregulation as a consequence of cancer) is of equal clinical relevance. In cancer survivors, aberrant rhythms in hormones like melatonin and cortisol indicate circadian disruption. Indeed, breast cancer survivors have lower levels of nighttime melatonin [112] as well as lower amplitudes of rhythms in melatonin [112]. This aberrant melatonin expression is likely responsible for the poor sleep quality observed in these patients [112, 113]. In fact, supplementation of treatment with melatonin has been linked to better sleep quality, improved drug efficacy, and even a longer overall survival [114, 115]. Not only does this help to normalize rhythms and lead to better sleep quality, but melatonin has anti-tumor and immunomodulatory functions [116]. Interestingly, one study demonstrated that supplementary treatment with melatonin helped to normalize aberrant rhythms in cortisol in patients with advanced-stage solid tumors [117]. Rhythms in cortisol release, which typically peak in the morning, can be dampened in lung [118], ovarian [119], and breast [120] cancer. In patients with metastatic breast cancer [121], renal cell carcinoma [122], lung cancer [123], and epithelial ovarian cancer [124], cortisol rhythms have been used to predict survival, with flatter rhythms indicating poorer prognoses. Flattened cortisol rhythms may be indicative of increased stress, which is also correlated with decreased quality of life [125]. Indeed, nighttime salivary cortisol levels have been negatively correlated with quality of life in patients with head and neck cancers [125]. A study including 30 women who were 5+ year survivors of ovarian cancer concluded that dampened cortisol rhythms were associated with higher fatigue and lower quality of life [126]. Similar results have been demonstrated in breast cancer survivors [127].

Other than dampened expression of important rhythmic hormones, consequences of general circadian disruption are prevalent in cancer survivors. Even in remission, issues like chronic fatigue and decreased sleep quality lower the quality of life in survivors [128, 129]. One study used wearable actigraphs to measure circadian parameters in early-stage breast cancer survivors (156 adult women who underwent surgical treatment followed by chemotherapy) 1 year after receiving their last chemotherapy treatment. The mesor (24 h rhythm-adjusted activity counts), amplitude (distance between peak and trough activity), peak activity, and circadian quotient (strength of the rhythm) of survivors were consistently 5%–15% lower than those of healthy controls [130]. This indicates that breast cancer survivors may have dampened/disrupted activity rhythms compared to their peers. Studies in metastatic colorectal cancer [131], and various late-stage cancers (including colorectal, pancreatic, lung, and liver) demonstrate similar results [132]. Furthermore, fatigue and weight loss, which are typically indicators of chemotherapy-induced circadian disruption [133], are correlated with a decreased median survival rate in metastatic colorectal cancer [133, 134] and have significant associations with circadian disruption in breast [135], pancreatic [132], lung [132], liver [132], and esophageal [136] cancer. As discussed above, disruptions in these rhythms aggravate oncogenic mechanisms, meaning that continued disruptions could lead to prolonged disease. Broadly, it is possible that tumor-driven circadian disruptions instigate further tumor progression, in turn potentially decreasing the overall rate of remission in survivors.

Sleep disturbance represents a particularly harmful morbidity affecting the quality of life of cancer survivors. In fact, a meta-analysis of 160 clinical studies assessing symptom prevalence in 46,279 cancer survivors found that 60.7% experience significant sleep disturbance [137]. Sleep quality is a multidimensional concept, however, and management strategies to restore the quality of sleep require an understanding of the complex underlying etiology. A small study of 20 women diagnosed with stage I-III breast cancer who had not yet started chemotherapy determined that 71% of them had difficulty sleeping (via PSQI), and more than half had sleep efficiencies lower than 85% [138]. Similar findings were reported in a meta-analysis of 27 studies involving cancer survivors which revealed an average pre-treatment PSQI of 7.11 (where > 5 indicates “poor sleep”) [139]. Studies directly assessing the treatment-independent effects of cancer on sleep should be expanded, however, to account for the psychological confound of coping with a recent cancer diagnosis. For this, animal models are extremely useful, as psychological stress is a much less significant variable. Research also demonstrates the chronic effects of the administration of cytotoxic chemotherapeutics on circadian rhythms and sleep, complicating the search for an underlying etiology. For example, a study examining 180 women undergoing various treatment regimens for breast cancer demonstrated a 10% decrease in sleep efficiency after their first round of chemotherapy compared to their baselines [140]. After the last round of chemotherapy, sleep efficiency was only 5% lower than their baselines [140]. Altogether, these results suggest that both a tumor-driven mechanism (i.e., independent from treatment) and a treatment-dependent mechanism contribute to circadian disruption and sleep disturbance in cancer survivors.

For survivors of childhood cancer, the effects of administering high doses of toxic treatments during crucial developmental periods are largely unknown. One study, however, reported that exposure to high amounts of anthracycline before the age of 21 was correlated with decreased sleep quality in adulthood [141]. It also examined metrics of sleep outcome including PSQI scores, insomnia, and nightly sleep duration [141]. Survivors of pediatric cancer (diagnosed before the age of 21 and have since completed treatment at least 5 years prior to the study) were compared to a random sample of closest-aged siblings of the survivors. Cancer survivors were more likely to report an overall sleep duration of less than 6 h as well as a longer latency to fall asleep. PSQI scores of survivors were higher (indicating poorer sleep quality) than those of siblings, and survivors also reported higher frequencies of awakenings and sleep medication use [141]. These data indicate that survivors of pediatric cancers suffer long-term effects on sleep, however, it is unclear if these effects are caused by circadian rhythm disruption from the cancers themselves, or if an underlying disruption contributing to oncogenesis still exists into adulthood. It is also worth noting that the two-process model of sleep, which states that two independent mechanisms drive sleep regulation (the sleep/wake homeostatic process and the circadian pacemaker process) [142], poses the possibility that mechanisms driving sleep disruption in cancer survivors are independent of circadian disruption. It is more likely, however, that circadian disruption before, during, or after tumorigenesis and treatment in large part contributes to prolonged sleep disturbances. Like adults, children report fatigue (not to be confused with sleepiness) as one of the most common treatment-related side effects [143]. The National Comprehensive Cancer Network (NCCN) defines cancer-related fatigue (CRF) as a distressing lack of energy that cannot be improved with normal amounts of rest or sleep. This definition encompasses physical, mental, and emotional exhaustion. CRF is intermingled with sleep disorders in cancer survivors, causing issues such as insomnia, excessive daytime napping, and increased sleep latency (i.e., the time it takes to fall asleep) that decrease quality of life [143]. Indeed, a study conducted in adults who had acute lymphoblastic leukemia during childhood determined that the survivors performed lower on an executive function test than a control population. This was correlated with increased fatigue and decreased processing speed and attention in female survivors [144]. Given the data discussed thus far, it is plausible to attribute these long-term symptoms in survivors of childhood cancer in part to circadian disruption experienced before, during, and after cancer treatment. Overall, evidence indicates an effect of circadian rhythm disruption on cancer outcomes, ranging from mild alterations in sleep patterns and alterations in melatonin, cortisol, and activity rhythms to predictors of overall/median survival.