Chronodisruption and melatonin: the need for sensible exposure metrics in epidemiological studies

Chronodisruption (CD), a relevant disturbance of the circadian organization of physiology, endocrinology, metabolism and behaviour, links light, biological rhythms and the development of cancers [1, 2] with melatonin being a key biological intermediary. In numerous experiments, the pineal indolamine has been shown to have anti-cancer properties via many different mechanisms [3]; indeed, it can offer some protection against all six hallmarks of cancer [4], namely against self-sufficiency in growth signals, insensitivity to growth-inhibitory signals, evasion of programmed cell death (apoptosis), limitless replicative potential, sustained angiogenesis and tissue invasion and metastasis [5]. With circadian or CD added to the hypothesized causal chains of melatonin-associated carcinogenesis, experimental and epidemiological studies into possible relationships between light, shift-work, melatonin, circadian rhythms and cancer developments were reviewed by an authoritative panel of 24 scientists in October of 2007 [6]. The overall evaluation was that shift-work that involves circadian disruption is probably carcinogenic to humans (Group 2A; emphasis added). Intriguingly, IARC scientists commented that the available sufficient evidence in experimental animals for the carcinogenicity of light during the daily dark period (biological night) [6] could have justified a Group I classification of shift-work that involves circadian disruption , namely that it is carcinogenic to humans. And yet, for such judgement we would need sufficient evidence that what experimental data and biological plausibility suggest is really valid for human beings. At this stage, though, the Working Group saw limited evidence in humans for the carcinogenicity of shift-work that involves nightwork [6]. With specific regard to observational studies in man, one of the foremost epidemiologists of the 20th century, Sir Richard Doll, pointed out that epidemiology is certainly a poor tool for learning about the mechanism by which a disease is produced, but it has the tremendous advantage that it focuses on the diseases and the deaths that actually occur, and experience has shown that it continues to be second to none as a means of discovering links in the chain of causation that are capable of being broken [7]. In this vein, epidemiological studies are necessary when researching CD, and its effects on humans, for two reasons: first, to complement the experimental information with observational data which, at present, are too limited to further classify shift-work that involves circadian disruption – be it that future study results strengthen or, indeed, weaken the overall evidence of carcinogenicity in humans. Second, if associations between CD and cancer may be interpreted as causal, real life observational studies would be needed as a basis for assessing what level of CD may be tolerable and what not. Importantly, IARCs call for epidemiological studies investigating further cancer endpoints beyond breast cancer on which the panel focussed has been addressed subsequently in two ways. In November of 2007, Cancer Research published a first study into shift-work and endometrial cancer, with a twofold risk increase in obese women working rotating hours [8]. In January of 2008, meta-analyses of investigations in individuals exposed to CD evinced increased prostate cancer risks (40%) in nine studies in flight personnel and in two studies of shift-workers [1]. These results will provide further incentives for epidemiologists to look diligently into the disconcerting possibility that breast, prostate, endometrial or other cancers may be causally associated with CD which can be widespread in our 24-hour-7-day societies, at work and at homes. But rather than having tens of studies in the near future employing incomparable exposure assessments, it seems advisable to see the 2007 IARC meeting in Lyon, France, as a starting point for concerted efforts by experimental and epidemiological scientists. Clearly, some aspects of relevant exposure variables will become overt only in the course of further research but, equally clearly, some candidate information can and should be considered a priori. To exemplify, both occupational (length, type, regularity, direction and tolerance of shift-work) and non-occupational (individual s age, exercise, medication, light exposure history , etc.) facets may critically determine the severity of CD, and its effects on the continuum from health to disease. An additional challenge which experimental and epidemiologicial researchers should try to meet stems from further melatonin considerations which may have to be taken into account and possibly integrated into comprehensive exposure metrics, namely that there are individuals who have genetically weak machinery and produce low melatonin levels nightly, i.e. they suffer from hypomelatoninaemia [9]. Moreover, circadian changes in melatonin receptor density or sensitivity [10], or the lack of, may be extremely relevant but this is usually not appreciated. In fact, while some of melatonin s oncostatic actions may not involve receptors, some actions on cancer growth surely do require them. This issue is often ignored because it is more difficult to get to the receptors in humans. Also, besides the membrane receptors for melatonin, nuclear-binding sites for the indoleamine may also be important for its ability to restrain cancer cell proliferation [11]. But until we define the definitive mechanisms (there may be many different J. Pineal Res. 2008; 45:335–336