 |
At a time when hopes are starting to fade again that a purely genetic approach in oncology will decisively and effectively control cancer alternative strategies based upon epigenetic mechanisms are being rediscovered.1,2 One of them is considering a link between the pineal gland3 , an unpaired diencephalic organ, and malignancy.4 Since the corpus pineale as it is also called is involved in the central control of temporal neuroimmunoendocrine processes5-7 this approach thus assumes a link between integrative neural processes and disintegrative tendencies manifesting in cancer.
During the first part of the twentieth century Engel and Bergmann8,9 as well as Hofstätter10 performed pioneering experimental and clinical studies in Vienna dealing with the antineoplastic activity of bovine pineal glands. After the discovery of melatonin by Lerner11 in 1958 these findings became almost forgotten. Vera Lapin working at the Vienna Cancer Research Center in the 1970s rediscovered them, reviewed the topic of pineal gland and cancer,12 and performed important experimental studies.13 A central finding was that surgical removal of the pineal gland (pinealectomy) stimulated both primary tumor growth and formation of metastases thus leading to reduced survival.14,15 This was in accordance to observations of other investigators16 including Bindoni17 who observed that pinealectomy even stimulates cell-division of normal tissues. The question arose how this general anti-proliferative effect of the pineal gland could be explained and whether it is mainly due to the pineal hormone melatonin or not. In order to find answers to this central question the current knowledge regarding the action of melatonin on experimental and on clinical cancers will be summarized in the following.
When surveying the available literature regarding the action of melatonin on experimental tumor growth in animals18 indications were found that the pineal hormone can exert divergent effects even within the same cancer model system. Based upon the fundamental findings of Reiter et al19 as well was Tamarkin et al20 with respect to the anti-reproductive effect of melatonin in rodents which can only be observed in case of late afternoon/evening injections to superimpose the secretory surge of endogenous melatonin (after the onset of darkness) it was tempting to assume that the same may also apply to the inhibitory action of melatonin on tumor growth. Using transplantable tumors in mice (the so-called Ehrlich tumor as well as a transplantable fibrosarcoma) as test systems it was demonstrated that late afternoon injections, preferably under long photoperiods (i.e., a light regimen of more than 12.5 hours per 24 hours lighting cycle), were able to inhibit whereas the same dose of melatonin given in the morning accelerated tumor growth or shortened survival.18 These findings thus demonstrated a circadian stage-dependent anti-tumor effect of melatonin which was subsequently confirmed by investigations of Wrba et al.21 On the basis of these results it became conventional that melatonin was administered in most subsequent experimental and clinical studies in the late afternoon or in the evening to achieve optimal tumor-inhibitory effects.
A considerable number of investigations have dealt with the effect of melatonin on experimental mammary cancers in female Sprague-Dawley rats induced by 7,12-dimethylbenz[a]anthracene (DMBA)22 being one of the most important model systems for human breast cancer.23 These hormone-dependent tumors are under the control of gonadal steroids24 and particularly prolactin.25 Melatonin is known to inhibit prolactin secretion as well as to affect the endocrine balance which regulates the ovulatory cycle.26 This renders a plausible explanation why the pineal hormone is able to inhibit tumors of this type27,28 affecting both their promotion and initiation.29 The anti-initiational effect is due to a competitive interaction of melatonin with hepatic phase I enzymes (P450-monoxygenases)30,31 which hydroxylate melatonin to yield 6-hydroxymelatonin32,33 and DMBA leading to the formation of 3,4-dihydrodiol-1,2-epoxy derivative, the ultimal carcinogen.29,34 It can be assumed furthermore that metabolically activated intermediary free radical forms of DMBA are scavenged directly by melatonin which has been reported to possess a pronounced anti-oxidative activity.35
Apart from the DMBA model system for breast cancer melatonin has been successfully tested to inhibit rat mammary cancer induced by N-nitroso-N-methylurea (NMU).36,37 These findings could have considerable clinical relevance since NMU-induced tumors are more alike human breast cancers showing a stronger estrogen-dependency as well as the formation of metastases.38 Since melatonin administration during the promotional phase inhibits tumor growth in this model system without substantially affecting the concentrations of circulating estradiol and prolactin it is assumed that the oncostatic action of the pineal hormone is predominantly exerted directly at the cellular level.
Melatonin has been tested on a considerable number of experimental in vivo tumors (for a review see ref. 38). It is evident that those tumors are inhibited most effectively which show hormone-dependency and/or a relatively high degree of cellular differentiation. Typical examples are the behavior of transplantable prostatic or mammary cancers. The growth of experimental prostatic cancers is effectively blocked by melatonin only if they are androgen-dependent or possess melatonin receptors,39,40 androgen-insensitive tumor sub-lines are refractory41,42 or may even be stimulated.43 Serial transplants of a DMBA-induced mammary tumor in inbred rats44,45 were found to be inhibited by melatonin only at an early and slow-growing passage of relatively high differentiation (carcinosarcoma) whereas a later and fast-growing passage of low differentiation (sarcoma) showed no response to chronic administration of the pineal hormone.46 Other undifferentiated hormone-independent in vivo tumors, such as the so-called Yoshida tumor, Walker 256 carcinosarcoma and others are not affected by melatonin.47 Since such tumors are on the other hand stimulated in their growth by pinealectomy16 it is obvious that the pineal gland contains other as yet unidentified anti-tumor substances48-54 with potentially important actions even on general development of multi-cellular organisms.55,56
Experiments with spontaneous endometrial carcinomas in BDII/Han rats57 illustrate complex neuroendocrine actions of melatonin during the development of sex-hormone-dependent tumors. The pineal hormone prolonged survival only if chronic night-time administration in drinking water was initiated by day 30 of life,58 i.e., shortly before beginning of puberty, whereas life-long treatment was ineffective if treatment was started on day 50, when animals had attained maturity.59 Since the pineal hormone delays pubertal development in both male and female rats60,61 it can be anticipated that the action of melatonin on endometrial cancer is exerted by delaying reproductive maturation. It is well conceivable that if melatonin treatment is started even earlier in postnatal life or if maternal melatonin secretion is modulated it could affect spontaneous endometrial cancer development even more profoundly. The validity of this assumption requires to be tested by further experiments in this as well as on other spontaneous hormone-dependent tumors such as breast cancer-prone C3H mice which are known to be inhibited by melatonin.62
The mechanism involved in the inhibition of well-differentiated tumors by melatonin not only consists of a neuroendocrine hormone-receptor-mediated component but is likely to include immune-mediated processes leading to tumor rejection. Melatonin possesses well-documented stimulatory effects on different parts of the haematopoietic and immune systems6,63 involving both membrane and nuclear melatonin receptors.64,65 This could also render an explanation why melatonin was found to stimulate the growth of virus-induced leukemia in mice whereas pinealectomy was inhibitory.66,67
According to recent experiments of D.E. Blask and his group evidence exists that melatonin controls tumor growth by inhibiting the metabolism of linoleic acid to 13-hydroxyoctadecadienoic acid (13-HODE),68,69 an important mitogenic signalling molecule which amplifies EGF-responsive mitogenesis and thus stimulates cancerous growth.70 The effect of melatonin on in vivo experimental tumor growth may thus be understood as a process involving both neuroimmunoendocrine as well as metabolic mechanisms.
Melatonin has been shown to inhibit the proliferation of a number of in-vitro cell lines at physiological concentrations including the human mammary cancer cell lines MCF-7,71,72 T47D and ZR75-1,47,73,74 the prostate cell-line LNCaP,75,76 biopsies derived from human melanomas77 as well as a murine adenohypophyseal prolactinoma.78 Studies on other cancer cell lines showed that melatonin inhibits only at pharmacological concentrations or has no effect. This included the human cell line HEp-2 originating from a laryngeal carcinoma, K562 being an erythroleukemia, EFO-27 of human ovarian origin as well as the mammary cell line EFM-1952,79 (for a detailed review see ref. 80). Also in case of MCF-7 cells the inhibitory action of melatonin is not always observable and appears to be confined to certain sub-clones.81 In some experiments even a tumor-stimulatory action of melatonin was detected such as in case of the breast cancer cell line MDA-MB-231 as well as on melanoma cells.47
In endometrial cancer cell-lines it was found that only estrogen receptor positive (SNG-II) but not estrogen receptor negative cells (Ishikawa) were inhibited by melatonin.82 Divergent effects of melatonin were also detected if the pineal hormone was given to primary cell cultures derived from different human mammary as well as ovarian cancer biopsies54 but no correlation with the presence of sex-steroid hormone receptors was detectable in this case.
A missing inhibitory effect of melatonin on tumor cells can be explained by a progressing loss of differentiation leading to not only absence of receptors for sex-steroids but also for melatonin. Recent studies showed that both membrane (MT1)83-86 and nuclear melatonin receptors (RORα)87-89 are essential determinants for an inhibitory effect of the pineal hormone on tumor cells. The detailed mechanisms involved are under investigation by ongoing studies.
In order to achieve a better predictability for an oncostatic effect of melatonin it would therefore be necessary to determine the levels of both sex-steroid hormone and melatonin receptors since the pineal hormone is apparently involved in the regulation of the estrogen-response system90,91 and thus codetermines functional sensitivity to circulating 17β-estradiol. According to findings of Gilad et al92 melatonin possesses only a transient inhibitory effect on androgen-dependent benign human prostatic epithelial cells since the pineal hormone inactivates its own receptors via a protein kinase C-mediated mechanism. This indicates that the growth-inhibitory action of melatonin on sex-steroid hormone dependent cells can be subtle and may be confined to certain phases of the cell cycle.
From these findings it is plausible why a progressing loss of differentiation of cancer cells (affecting both melatonin and other hormone receptors) of cancer cells is bound to lead to insensitivity to melatonin. A typical example for this was found in case of a human melanoma cell line where the pineal hormone inhibited only early passages which were well-differentiated and slow-growing whereas undifferentiated and fast-growing late passages were refractory and were even stimulated at millimolar concentrations93. Stimulation of cancer cells by melatonin does not appear to be an uncommon feature since the pineal hormone stimulated the growth of some primary cell cultures derived from human mammary as well as ovarian tumor biopsies.54 Such paradox reactions of cancer cells to hormone treatment have also been found in sub-clones of the mammary cell-line MCF-7 with respect to the antiestrogen tamoxifen.94 Such unexpected hormonal effects on cellular growth can be attributed to mutational processes in cancer cells leading to profound derangements within intracellular signalling cascades which in normal cells underlie a complicated and stringent fine tuning by both intra- and extra cellular signals.
Clinical trials have been confined to patients suffering from advanced or even terminal malignancies since melatonin is not an officially approved drug which underwent systematic studies to determine its efficacy as well as toxicity. Comprehensive reviews of the studies performed until now were published recently.95,96 The studies performed by Lissoni and colleagues deserve special mention since they are the most detailed and best documented. From the experience of this group having treated several hundreds of cancer patients it appears that the pineal hormone may indeed possess favourable effects if given in the late afternoon or evening to superimpose the endogenous surge of the hormone.
Melatonin was initially administered to patients with advanced malignancies that were refractory to other types of treatment.97 Among the 54 patients treated only one patient showed an objective tumor response, minor responses were found in two cases and disease stabilization occurred in 20 patients. Also in other related studies98,99 melatonin did not effectively stop or slow down the course of advanced malignant tumor processes underlining that the pineal hormone cannot not be viewed as a cytostatic agent. Lissoni, however, often had the impression that melatonin treatment led to an improved general condition of his patients. This observation encouraged him to perform further studies on more than 200 patients affected by terminal malignant disease. These studies were mostly performed under randomised conditions using melatonin at 10-30 mg per day and achieved the following results: in bronchial cancer patients with metastases daily melatonin treatment significantly elevated one-year survival (+20%) compared to controls;100 patients suffering from glioblastoma melatonin given with brain irradiation significantly elevated one-year survival (+37%) compared to irradiation alone;101 in patients with heavily pretreated metastatic breast cancer melatonin co-administered with tamoxifen highly significantly extended one-year survival (+39%) compared to antiestrogen alone102; patients with resected melanoma plus lymph node involvement had significantly less relapses after one year under melatonin (-44%) compared to best supportive care alone,103 and in patients with different solid tumors with brain metastases melatonin plus irradiation significantly increased one-year survival (+25%) compared to irradiation alone.104
The greatest number of patients treated by Lissoni with melatonin received the pineal hormone in combination with interleukin-2 (IL-2), a lymphokine which is known to possess considerable side-effects leading among others to high fever as well as hypotension. This treatment if combined with melatonin (40-50 mg per day) was found to be better tolerated and the therapeutic results were improved: among more than 500 patients with locally advanced or metastasised solid tumors of different origins one-year survival was significantly elevated.108-105 Even in case of patients with advanced solid tumors having a life-expectancy of less than six months and not responding to either chemotherapy or any other types of adjuvant treatment melatonin combined with IL-2 not only improved survival but also their quality of life.109
A favourable effect upon the quality of life appears to be a particularly important feature of chronic late afternoon/evening melatonin treatment of cancer patients since, from ethical point of view, a mere life extension in agony does not appear to be a desirable therapeutic aim. This effect of melatonin is most probably due to the well-documented sleep-inducing effect110 of the pineal hormone as well as a supportive effect on the endorphin system leading to reduction of pain.111 It can be assumed that an improvement of the general well-being of cancer patients helps to foster endogenous defence mechanisms against tumor growth and would thus indirectly contribute to a longer survival although, according to the preceding chapters, melatonin is unlikely to directly inhibit the growth of advanced malignancies. According to the critical review of Hrushesky95 the clinical results obtained so far with melatonin by Lissoni and colleagues are encouraging and justify a further systematic verification under double-blind placebo controlled conditions. If these results will indeed be confirmed they would justify a use of melatonin in cancer patients as an effective supportive measure to optimise existing oncotherapeutic strategies.
Initially, the circadian profiles of urinary melatonin excretion were analysed in untreated postmenopausal Indian patients suffering from breast cancer (mostly primary localized tumors) as well as in controls with uterovaginal prolapse. A 30% depression of the 24-hour excretion of melatonin among the cancer patients was accompanied by a phase delay of the circadian peak leading to higher levels in the early morning than at night.112 Almost parallel to this study, Tamarkin et al113 detected a significant depression of nocturnal serum melatonin in unoperated primary breast cancer patients of clinical stages I and II if estrogen receptor positive tumors were present. Bartsch et al114 subsequently analysed the circadian rhythm of serum melatonin in German breast cancer and found a 56 % depression of the amplitude compared to age-matched controls with benign breast disease. This depression showed a tumor-size dependency being more pronounced if big tumors were present (T3: -73%; T2: -53%, T1: -27%), an observation which was also reported by Hoffmann et al.115 Patients with recurrent breast tumors on the other hand which appeared after surgical removal of the primary tumor did not exhibit any depression of melatonin. Since the main metabolite of melatonin, 6-sulfatoxymelatonin (aMT6s), was found to show parallel circadian changes to melatonin in serum among patients with primary breast cancer116 it was concluded that the observed changes of melatonin in these individuals were not due to a modified hepatic metabolism of the pineal hormone. This finding paved the way for further studies using the noninvasive measurement of nocturnal urinary aMT6s to estimate the levels of circulating melatonin. In a subsequent study on untreated German breast cancer patients with localized primary tumors the nocturnal urinary excretion of aMT6s was found to be significantly depressed (-48%) showing an inverse correlation with tumor size as in the preceding study.117
Karasek et al118, 119 detected a significant depletion of the nocturnal surge of circulating melatonin by around 50% in endometrial cancer patients compared to age-matched controls. Grin and Grünberger120 observed an even greater depletion of circulating melatonin by about 90 % in such patients. In contrast to endometrial cancer it appears that the presence of cervical cancer hardly affects circulating melatonin.119 In a large-sized study on patients with ovarian cancer (n=119) a high variability of the nocturnal urinary excretion of aMT6s was found, some of them showing very low levels whereas others exhibited exceedingly high values.121 A similar observation was made in French ovarian cancer patients for the levels of circulating melatonin (Touitou and Bartsch, unpublished results). Karasek et al,119 however, did not detect changes of the circadian profiles of serum melatonin in patients suffering from ovarian cancer. It is conceivable that intra-ovarian melatonin production122 may have contributed to the high levels of circulating melatonin observed in some of these patients.
In two consecutive studies performed under comparable clinical conditions using the same RIA-methodology the circadian rhythm of serum melatonin was determined in untreated patients with benign prostatic hypertrophy (BPH) or with primary localized malignant prostate (PC) tumors. Patients with PC showed extremely low levels of nocturnal melatonin.123-125 When pooling the results of the studies a 71% depression of the melatonin amplitude resulted compared to patients with BPH.126 A sub-division of these patients according to the stage of their tumors showed an inverse correlation with tumor-size. At the T1-stage the amplitude of melatonin was depressed by 28% compared to patients with small BPH but patients with T2- and T3/T4-tumors exhibited a drastic depletion by almost 80% compared to patients with BPH of comparable size. Interestingly, patients with so-called incidental carcinomas (PCi) which are small foci of highly differentiated malignant cells detected during the histological examination of BPH showed nocturnal serum melatonin concentrations that were higher than in BPH patients.123,124 Since the melatonin metabolite aMT6s in both serum and urine showed parallel changes to melatonin in the different groups of patients studied125 it has to be concluded that the depression of circulating melatonin in primary prostate cancer patients is not due to a modified peripheral metabolism of the pineal hormone but may either be caused by a reduced pineal secretion or an enhanced binding/degradation by tumor tissue.
In female Russian patients with primary thyroid cancer before surgery a 56% lower nocturnal excretion of aMT6s was found than in controls of comparable age not affected by thyroid disease.127,128 A similar depression of aMT6s was also detected in controls suffering from different types of benign thyroid disease127,128 indicating that thyroid enlargement, irrespective whether it is of malignant or benign nature, negatively affects the levels of circulating melatonin. In contrast to Kvetnaia et al127 Karasek et al96 found highly significantly elevated levels of nocturnal serum melatonin in thyroid cancer patients compared to healthy age-matched controls. It could well be that this elevation was due to the presence of distant metastases since elevated levels of melatonin were also detected in breast and prostate cancer patients suffering from disseminated disease.129
Male patients with primary larynx cancer only showed a marginal increase of the average nocturnal urinary aMT6s-excretion (+12%) compared to age-matched healthy controls. Sub-division of these patients according the size of their primary tumor revealed that patients at the T2-stage exhibited a 125% increase compared to controls whereas patients with T3-tumors were at the level of controls and those with T4-tumors showed a 62% depression.127
In male patients with bronchial cancer of clinical stages T1-4N0-3M0-1, the majority being nonsmall cell cancers, a highly significantly depression of nocturnal aMT6s-excretion by 60% was observed.127 Viviani et al130 found obliterated day/night variations of circulating melatonin in patients with nonsmall cell bronchial cancer whereas Dogliotti et al131 reported very high early morning and night-time melatonin levels in blood of patients at clinical stages III and IV, however, very low concentrations if small cell lung cancers were present.
In male patients with primary unoperated stomach cancer without metastases (T3-4N2-XM0) Kvetnaia et al127 found a 59% depletion of the nocturnal urinary excretion of aMT6s. In an earlier study Kvetnoi and Levin132 detected a phase delay of urinary melatonin excretion in such patients leading to depressed levels at night and elevated levels in the morning.
In patients with primary unoperated colorectal carcinoma, with or without metastases, Khoory and Stemme133 observed a very pronounced depletion of nocturnal plasma melatonin. In contrast to this, Kvetnaia et al127 found a 44% higher nocturnal urinary aMT6s-excretion in operated and untreated male patients who were mostly affected by large tumors of stages T3 and T4 and which in some cases showed a pronounced lymph node involvement and distant metastases.
Incongruent changes were found for the 24h urinary excretion of aMT6s in patients with osteosarcoma:134 80% of them showed lower whereas the others had highly elevated levels compared to controls. In Hodgkin's sarcoma Lissoni et al135observed clearly elevated concentrations of nocturnal circulating melatonin.
From the above summarized results in patients with cancer of the reproductive tract or outside of the same it is obvious that melatonin can show considerable variations with respect to the levels of circulating melatonin even among patients affected by the same tumor type. Further studies are therefore required to better understand the mechanisms involved in such changes. For this purpose, studies analysing melatonin secretion and production in tumor-bearing animals are relevant.
Initially, Vera Lapin observed a negative correlation between pineal melatonin content and tumor-size in rats bearing Yoshida tumors136 indicating an inhibition of pineal melatonin biosynthesis by cancer growth. This view was supported by studies of Leone and Skene137 as well as Schmidt et al138 who found that supernatants of cancer cells inhibit the production of pineal melatonin under in vitro conditions. As opposed to that, it was observed in F344 Fischer rats with DMBA-induced mammary tumors that their circannual rhythm of nocturnal urinary aMT6s-excretion was obliterated due to an elevated melatonin production.139 A similar observation was made in female BDII/Han rats during the development of spontaneous endometrial adenocarcinomas.139 F344 Fischer rats with an early tumor passage (derived from the above-mentioned DMBA-induced mammary tumors) showed an elevated melatonin production which was accompanied by an enhanced activity of arylalkylamine-N-acetyltransferase140 (AA-NAT, the rate-limiting step of pineal melatonin biosynthesis) which is under adrenergic control.141 These tumor-bearing animals also showed an activation of the sympathetic nervous system (elevated urinary excretion of norepinephrine but not of epinephrine) which in turn was due to a stimulation of cellular immunity (elevated urinary excretion of macrophage-derived biopterin and of γ-interferon in plasma).142 In contrast to that, rats with tumors of similar size but of a later passage, being fast growing and showing distant metastases, nocturnal peak plasma melatonin was depressed by almost 75%.142 This depression was not accompanied by either a reduced activity of AA-NAT or other steps of melatonin biosynthesis but circulating tryptophan, the precursor amino acid of melatonin, was drastically reduced. This reduction was not caused by cachexia since other amino acids remained unchanged [Bartsch C. et al unpublished results]. Maestroni and Conti143 found that human mammary cancer tissue binds considerable amounts of melatonin. This renders an additional explanation for the observed depression of circulating melatonin in patients as well as animals with advanced tumors. It is conceivable that not only melatonin but also its precursor tryptophan may be trapped by cancer cells thus contributing to a deficiency of this amino acid in blood. Slominski et al144 recently reported that tryptophan is converted to melatonin within melanoma cells whereas Bartsch et al142 are assuming that melatonin may be catabolically cleaved to kynurenine derivatives within tumor tissue. This further adds to the complexity of the metabolism of melatonin in a cancer-affected organism. Further investigations are urgently needed to clarify details of these pathophysiological phenomena.
If circulating melatonin is found to be depressed in patients with localized primary cancers it would appear logical to consider a substitutional therapy with the aim to control the malignant process. Such hopes, however, do not appear to be justified on the basis of the above-described experimental findings since tumor-bearing animals (e.g., with advanced serial transplants of DMBA-induced mammary cancers) showing a depression of circulating melatonin are totally refractory to a potential tumor-inhibitory effect of exogenous melatonin.145 Despite this, the clinical studies of Lissoni indicate that melatonin administration is apparently able to delay the course of advanced or even final malignant disease (see review ref. 95) leading to an extended survival. As mentioned before, it may be anticipated that this life-prolonging effect is probably due to favourable effects on the sub-systems of the body resulting in an improved neuroimmunological surveillance as well as endocrine balance which may help to control metastatic spread being a central determinant for the patients' prognosis. An integral part of this so-called neuroimmunoendocrine effect of melatonin on malignancy could be to re-establish and -synchronize circadian disturbances affecting the autonomic nervous including the sleep-wake cycle146 as well as the neuroendocrine system,126,147,148 and perhaps even the central circadian clock in the suprachiasmatic nucleus.149 In addition, melatonin administration could positively affect the production and secretion of endogenous anti-tumor substances present in the pineal gland52-54,150,151 as well as in other organs48,152-154 helping to resist the formation of metastases. The observed elevated production of melatonin in patients with metastases and local recidives could therefore be viewed as an effort of the organism to resist the detrimental and destructive malignant process. The same could apply to the phase of early tumor development when endogenous melatonin secretion is apparently up-regulated.142 In this case it would be worth testing whether an additional administration of melatonin could inhibit tumor development and growth since experimental findings obtained both under in vivo and in vitro conditions indicate that melatonin is able to effectively control well-differentiated tumors.
Finally, a word of caution regarding a potential administration of melatonin to patients suffering from haematopoietic neoplasias such as leukemia: experimental findings exist that the pineal hormone shortens the survival of mice with leukemia.66,67 It was also found that melatonin stimulated the growth of some primary cell lines derived from human mammary or ovarian biopsies.54 Therefore further systematic clinical studies on cancer patients are needed and it will not be advisable to advocate an uncontrolled self-treatment with melatonin by oncological patients.
A central question is whether changes of circulating melatonin in cancer patients could be used for diagnostic purposes. Reductions of melatonin are found in certain types of malignancy, such as breast and prostate cancer, but mainly if medium-sized or large primary tumors are present,121 i.e., at a time when the malignant process is clinically clearly evident. Therefore such changes of circulating melatonin will only have a limited diagnostic value compared to conventional tumor markers used in clinical chemistry. Endogenous melatonin on the other hand is up-regulated if local recidives or distant metastases develop.121 Relative to the depression of melatonin during the growth of the unoperated localized primary tumor this alteration is quite dramatic. If melatonin secretion was measured at regular intervals during the course of the malignant disease it would thus be possible to obtain indications for the growth of new cancer cells after surgical removal of the primary tumor. For this purpose noninvasive determinations of nocturnal urinary 6-sulfatoxymelatonin (being a reliable estimate of nocturnal melatonin production in cancer patients121) may be included in the monitoring program of oncological patients parallel to conventional tumor markers. A similar approach may also be used for the early diagnosis of cancer since the production of the pineal hormone is elevated by the growth of early stages of cancer, such as in case of patients with so-called incidental carcinoma124 as well as in animals bearing well-differentiated tumors.140,142 Since melatonin shows a considerable inter-individual variation but a high individual stability with properties of a personal marker rhythm155 it would be necessary to establish individual norms rather than normal ranges among healthy populations.
Recent studies revealed that the central circadian pacemaker located in the suprachiasmatic nuclei (SCN) of the hypothalamus is involved in the control of cancer: destruction of the SCN149 as well as a deficit of the circadian Period2 gene156 accelerates tumor development. The circadian production and secretion of melatonin is driven by the SCN and it serves as an important output signal of the central clock conveying information regarding time of day to practically all parts of the body including the SCN itself.141 The endogenous circadian oscillation of the SCN and thereby the secretion of pineal melatonin is synchronized to environmental photoperiods by retinally perceived light.141 The SCN and the pineal gland together with the eyes are a functional unit serving as the central circadian time-keeping system of the body. This system is apparently negatively affected by cancer growth and the above-described increasing depression of circulating melatonin in cancer patients and tumor-bearing animals in the presence of localized primary tumors129 together with temporal neuroendocrine disturbances147,148 can be viewed as a weakening of central control mechanisms over malignancy to facilitate metastatic spread.
The circadian time-keeping system including melatonin is profoundly influenced by shift-work and East-West travels. Recent publications indicate that nurses on rotating night shifts over prolonged periods of time as well as female flight attendants working on long-distance flights seem to possess an increased risk to develop breast cancer.157,158 Light at night is known to inhibit melatonin production3,141 which when given uninterruptedly leads to total suppression of pineal secretion and is therefore called “physiological pinealectomy”. An obliteration of the nocturnal surge of melatonin, fully or even partially, stimulates experimental cancers.38,159 Erren160 discusses that demographic differences of breast cancer could follow systematic geographic patterns connected with seasonally modulated circadian rhythms of melatonin at different latitudes. Due to the high state of industrialization in Western nations connected with an unlimited access to artificial light it, however, appears likely that such geographic differences no longer exist and that a chronic self-chosen overexposure to light at night leads to a suppression of nocturnal melatonin secretion. This so-called “light pollution” may not only have extinguished previous seasonal endocrine and behavioural patterns including reproduction in humans161 but could also contribute to a higher risk for the development of hormone-dependent cancers. This view is shared by Stevens162 who initially hypothesized that electric power via an inhibition of melatonin may stimulate breast cancer163 but he extended this theory to different components of the electromagnetic spectrum including light. The effects of extremely low frequency electric and magnetic fields due to alternating currents as well as of pulsed high frequency electromagnetic fields connected with mobile telecommunication on both melatonin164-166 and experimental tumor growth167,168 are still quite controversial but according to our present knowledge appear to have no grave general health hazards. It has also been hypothesized that drugs which inhibit melatonin secretion may lead to an enhanced risk for breast cancer.162 Epidemiological studies, however, have provided no sound evidence that β-blockers and certain benzodiazepines which suppress melatonin secretion169-171 lead to an elevated cancer risk.172,173 It is possible that absence of pineal melatonin circadian secretion could be of adifferent physiological significance in humans than in experimental animals. There is no evidence that patients after pinealectomy due to the presence of a pineal tumor exhibit an increased cancer risk. The same applies to those individuals who naturally show very low or even absent circadian amplitudes of the pineal hormone. In order to understand the real pathophysiological role of melatonin for the aetiology of cancer it may perhaps be necessary to correlate life-long individual patterns of melatonin secretion with the trend to develop neoplasias. This, however, is beyond the current scope of medical research. Findings in experimental animals with a tendency to develop spontaneous tumors indicate that there may be a decisive temporal biological window before the onset of puberty during which manipulations of circulating melatonin could decisively modulate the development of endometrial cancer in adulthood.59 Nocturnal serum melatonin is known to physiologically decline during human growth and puberty.174 Could there perhaps be a connection between the earlier onset of human puberty as well as growth acceleration in our days, “light pollution” (to further reduce melatonin secretion) and the elevated incidence of hormone-dependent cancers such as of breast and prostate?
Free Full text in PMC]Free Full text in PMC].Free Full text in PMC]