Melatonin in the treatment of cancer: A systematic review of randomized controlled trials and meta-analysis

Abstract

Most observational studies show an association between melatonin and cancer in humans. We conducted a systematic review of randomized controlled trials (RCTs) of melatonin in solid tumor cancer patients and its effect on survival at 1 yr. With the aid of an information specialist, we searched 10 electronic databases from inception to October 2004. We included trials using melatonin as either sole treatment or as adjunct treatment. Prespecified criteria guided our assessment of trial quality. We conducted a meta-analysis using a random effects model. We included 10 RCTs published between 1992 and 2003 and included 643 patients. All trials included solid tumor cancers. All trials were conducted at the same hospital network, and were unblinded. Melatonin reduced the risk of death at 1 yr (relative risk: 0.66, 95% confidence interval: 0.59-0.73, I2=0%, heterogeneity P<or=0.56). Effects were consistent across melatonin dose, and type of cancer. No severe adverse events were reported. The substantial reduction in risk of death, low adverse events reported and low costs related to this intervention suggest great potential for melatonin in treating cancer. Confirming the efficacy and safety of melatonin in cancer treatment will require completion of blinded, independently conducted RCTs.

Full text

Melatonin in the treatment of cancer: a systematic review of
randomized controlled trials and meta-analysis
Introduction
Melatonin, an indolamine secreted from the pineal gland,
follows a circadian rhythm determined both by its produc-
tion and secretion [1]. Melatonin is associated with effects
on sleep, mood, sexual maturation and reproduction,
immune function, aging, and the antioxidative defense
system [1, 2].
The association between melatonin levels and cancer
progression has suggested to some that melatonin may be
a modifier of cancer progression. The mechanisms by
which melatonin may act in this way have not been fully
elucidated. One of the potential mechanisms is the
possibility that the hormone has antimitotic activity as a
result of intranuclear downregulation of gene expression
or through the inhibition of growth factor release and
activity [3, 4]. There is also evidence to support the
inhibition of solid cancer growth in vivo by suppressing
tumor linoleic acid uptake and metabolism via a melato-
nin receptor-mediated mechanism [5, 6]. Other possible
anti-cancer mechanisms include protection from oxidative
damage [7], anti-angiogenic activity [8], anti-inflammatory
activity [9], anticachectic properties [10, 11], and immu-
nostimulation [1, 12].
Data on the relationship between melatonin and cancer in
humans is somewhat conflicting, however the majority of
reports show a positive action. Associations of low levels of
melatonin with human cancer include breast cancer [13];
prostate cancer [14], and endometrial, lung, gastric, and
colorectal cancers [4, 15]. In addition, there is evidence for
the beneficial use of melatonin during chemotherapy
[12, 16–19]. Claims include the potential for melatonin to
attenuate damage to blood cells from both radiation therapy
and chemotherapy [20, 21]. Moreover, melatonin may
induce a decline in the frequency of chemotherapy-induced
asthenia, stomatitis, cardiotoxicity, and neurotoxicity [12].
A number of clinical trials have addressed the impact of
melatonin on solid tumors; as yet, however, there is no
satisfactory synthesis of the data. We performed a system-
atic review and meta-analysis of the literature for all
randomized controlled trials (RCTs) examining survival at
1 yr that involve the use of melatonin in the treatment of a
variety of cancers.
Methods
Data selection
With the aid of an information specialist, we (EM, PW)
searched the following databases independently, in dupli-
cate (from inception to October 2004): AltHealthWatch,
AMED, CancerLit, CinAhl, Cochrane Controlled Trials
Abstract: Most observational studies show an association between melatonin
and cancer in humans. We conducted a systematic review of randomized
controlled trials (RCTs) of melatonininsolidtumorcancerpatientsandits
effect on survival at 1 yr. With the aid of an information specialist, we
searched 10 electronic databases from inception to October 2004. We
included trials using melatonin as either sole treatment or as adjunct
treatment. Prespecified criteria guided our assessment of trial quality. We
conducted a meta-analysis using a random effects model. We included
10 RCTs published between 1992 and 2003 and included 643 patients. All
trials included solid tumor cancers. All trials were conducted at the same
hospital network, and were unblinded. Melatonin reduced the risk of death
at 1 yr (relative risk: 0.66, 95% confidence interval: 0.59–0.73, I
2
¼0%,
heterogeneity P£0.56). Effects were consistent across melatonin dose, and
type of cancer. No severe adverse events were reported. The substantial
reduction in risk of death, low adverse events reported and low costs related
to this intervention suggest great potential for melatonin in treating cancer.
Confirming the efficacy and safety of melatonin in cancer treatment will
require completion of blinded, independently conducted RCTs.
Edward Mills
1
, Ping Wu
2,3
, Dugald
Seely
3,4
and Gordon Guyatt
1
1
Departments of Clinical Epidemiology &
Biostatistics and Medicine, McMaster
University, Hamilton, ON, Canada;
2
London
School of Hygiene and Tropical Medicine,
University of London, London, UK;
3
Division of
Clinical Epidemiology, Canadian College of
Naturopathic Medicine, Toronto, ON;
4
Division
of Hematology/Oncology, Sick Children’s
Hospital, University of Toronto, Toronto, ON,
Canada
Key words: CAM, complementary and
alternative medicine, cancer, melatonin,
survival, systematic review
Address reprint requests to Edward Mills,
McMaster Health Sciences Centre, Depart-
ment of Clinical Epidemiology and Biostatis-
tics, Room 2C12, 1200 Main St. West,
Hamilton, Ontario, Canada, L8N 3Z5.
E-mail: millsej@mcmaster.ca
Received Febraury 1, 2005;
accepted May 23, 2005.
J. Pineal Res. 2005
Doi:10.1111/j.1600-079X.2005.00258.x
Copyright Blackwell Munksgaard, 2005
Journal of Pineal Research
1

Register (CENTRAL), MedLine, and EMBASE. To
identify unpublished research, we searched http://www.
clinicaltrials.gov, National Research Register (UK) and the
Meta-Register. Searches were not limited by language. We
additionally searched bibliographies of identified reviews
and contacted experts in the field. The following search
terms were used, but not limited to: Ômelatonin,ÕÔpineal
hormone,ÕÔcancer,Õand Ôrandom*Õ.
Table 1. Study characteristics
Reference
Description of
randomization
Allocation
concealment
Blinding
status Placebo
Ethics/
informed
consent
Source of
funding
Intention
to treat
26 Yes
a
Yes
a
Open No Yes
a
Unfunded
a
Yes
16 Yes
a
Yes
a
Open No Yes Unfunded
a
No
27 Yes
a
Yes
a
Open No Yes
a
Unfunded
a
Yes
32 Yes
a
Yes
a
Open No Yes Unfunded
a
No
30 Yes
a
Yes
a
Open No Yes Unfunded
a
Yes
28 Yes
a
Yes
a
Open No Yes Unfunded
a
Yes
31 Yes
a
Yes
a
Open No Yes Unfunded
a
Yes
19 Yes Yes
a
Open No Yes Unfunded
a
Yes
36 Yes
a
Yes
a
Open No Yes Unfunded
a
Yes
29 Yes
a
Yes
a
Open No Yes Unfunded
a
Yes
a
Information obtained from communication with author.
54 medline
16 Cochrane controlled trials register
9 AMED, CINAHL, Alt Health Watch
3 ongoing studies: clinical trials.gov
24 RCTs on melatonin or pineal
hormone and cancer were found
17 RCTs of melatonin as an
intervention treatment compared
with other treatments
9 RCTs were excluded:
1 IL-2 plus melatonin versus chemotherapy
1 melatonin plus chemotherapy versus melatonin
only
2 IL2 plus melatonin versus supportive care
1 melatonin versus melatonin plus 5-MTT
1 IL2 plus melatonin versus IL2 plus melatonin
and NTX
1 high dose IL2 versus low dose IL2 plus
melatonin
1 melatonin versus melatonin plus aloe
1 melatonin versus melatonin plus 5-MTT
10 RCTs of melatonin assessing
one-year survival were included
7 RCTs were excluded
2 observed the duration of response, tumor
regression and side effects
1 observed weight loss, tumor regression and
TNF serum level
1 observed AUC of the chemotherapy
medicine and hematological change.
1 observed hypotension depressive symptoms
and side effects
55 abstracts excluded as
unrelated.
3 ongoing studies excluded as
unfinished enrollment.
Fig. 1. Flow diagram of studies examined
in this systematic review. 5-MTT, 5-meth-
oxytryptamine; AUC, area under the
curve; IL2, interleukin 2; NTX, naltrex-
one; RCT, randomized controlled trial;
TNF, tumor necrosis factor.
Mills et al.
2

Table 2. Study findings
Reference Population n Age range Interventions Outcomes measured Dosage ARR NNT RRR (95% CI)
26 Metastatic nonsmall cell
lung cancer resistant
to cisplatin
63 39–78 Melatonin versus
supportive care
Progression
Survival at 1 yr
10 mg/day orally at 19:00 hr 0.2 5 20% (3–40)
16 Brain metastases due
to solid tumors
50 38–72 Supportive care +
melatonin versus
supportive care alone
Survival at 1 yr 20 mg/day at 20:00 hr 0.25 4 30% (3–53)
27 Advanced solid tumors
other than renal cancer
and melanoma
80 36–74 Interleukin 2 (IL2) + melatonin
versus IL2 alone
Tumor regression
Survival at 1 yr
40 mg/day orally at 20:00 hr 0.31 4 36% (15–55)
32 Breast cancer (ER)) 40 42–80 Tamoxifen + melatonin
versus tamoxifen alone
Clinical response
Survival at 1 yr
Toxicity
20 mg/day at noon and
20 mg/day in evening
0.39 3 52% (14–76)
30 Brain glioblastoma 30 32–76 Radiotherapy + melatonin
versus radiotherapy alone
Progression free survival
Survival at 1 yr
20 mg/day 0.37 3 40% (9–46)
28 Malignant melanoma 30 38–81 No treatment versus
melatonin
Lymph node relapse
Disease free survival
Survival at 1 yr
20 mg/day orally in the evening 0.4 3 38.5% (9–84)
31 Advanced nonsmall cell
lung cancer
70 39–80 Cisplatin + etoposide +
melatonin versus cisplatin +
etoposide
Tumor regression
Survival at 1 yr
20 mg/day orally in the evening 0.25 4 30.5% (4–52)
19 Advanced metastatic
solid tumors
250 39–81 Chemotherapy + melatonin
versus chemotherapy alone
Disease progression
Survival at 1 yr
20 mg/day orally 0.28 4 36% (22–49)
36 Renal cell cancer 30 28–63 Morphine + melatonin
versus morphine alone
Tumor regression
Survival to 3 yr
20 mg/day orally in evening 0.29 4 57% ()17–86)
29 Metastatic nonsmall cell
lung cancer
100 38–81 Chemotherapy + melatonin
versus chemotherapy alone
Disease progression
Survival to 5 yr
Toxicity
20 mg/day orally in evening 0.37 3 47% (16–63)
ARR, absolute risk reduction; NNT, number needed to treat; RRR, relative risk reduction.
Melatonin in the treatment of cancer
3

Inclusion and exclusion criteria
We include RCTs enrolling participants with diagnosed
cancers and providing details of survival at 1 yr. We
included trials involving patients of any age, sex, or cancer
stage. We included trials using melatonin as either sole
treatment and as adjunct treatment. Trials had to treat
randomized patients equally with the exception that the
active group receive melatonin.
We excluded animal studies, pharmacokinetic trials, and
trials comparing melatonin when combined with other
anti-cancer agents aside from standard chemotherapy
regimens.
Data abstraction
EM, PW and DS developed and piloted data abstraction
forms. EM and PW extracted data independently and in
duplicate [22].
Quality assessment
Table 1 presents our assessment of trial quality. We
determined methods of randomization, allocation conceal-
ment, blinding status of patients and assessors, use of
placebo, ethics review and informed consent, sources of
funding and adherence to the intention-to-treat principle.
We contacted the study authors to determine items that
were inappropriately reported.
Quality assessment and trial inclusion was performed
independently, in duplicate (EM, PW) with third party
arbitration when uncertainty existed (DS). We did not rely
exclusively on the published reports of the trials as, in this
case, the authors did perform important methodological
criteria in the conduct of the trial, but did not report it in
the original manuscript.
Statistical analysis
The kappa (j) value provided a measure of chance-
corrected agreement between assessors of eligibility and
study quality. We determined the proportion of patients in
treatment and control groups alive at 1 yr [23], the relative
risks (RR) and applicable 95% confidence intervals (95%
CI), the absolute risk reductions and numbers needed to
treat (NNT) were determined. Pooled analysis of RR was
conducted using a random effects model. We pooled the
results of different trials for different cancers because the
similar putative mechanism of action in each cancer
suggests the possibility of similarity of response. We tested
for homogeneity using the Zalen test and the I
2
test [24].
A priori explanations of heterogeneity included cancer type,
dosage of melatonin and adjunct chemotherapy used.
Publication bias was tested using both the Egger test with
funnel plot and Kendall’s test on standardized effect versus
variance. In order to examine the temporal relationship of
the accumulated data, we conducted a cumulative meta-
analysis [25]. StatsDirect was used for all meta-analytic
procedures (StatsDirect, Copyright 1993–2004, Manches-
ter). We conducted both standard and cumulative meta-
analyses.
Results
Fig. 1 details the yield of the sources and the study
selection. jfor initial decisions on the inclusion of studies
was 0.9 (95% CI: 0.6–1) suggesting excellent agreement.
The 10 studies included (Table 2) were published between
1992 and 2003 and included 643 patients [16, 19, 26–32].
The included studies were all reported in English and were
all from Italy and Poland. We additionally located two
trials currently enrolling participants in the US (one trial
for nonsmall cell lung cancer conducted by the Cancer
Treatment Centers of America [33], and one for brain
metastases by the National Cancer Institute [34]).
Determination of study quality (Table 1) indicates that
the studies were of moderate quality, but lacked important
methodological techniques shown to potentially prevent
bias such as blinding and use of placebo. General reporting
of the studies was poor, but contact with the studiesÕlead
author clarified the missing information. All trials were
hospital funded.
There is a suggestion of publication bias evident in the
funnel plot [Eggers test: )1.260231 (approximate 95% CI:
)2.508723 to )0.011738), P¼0.0483, Fig. 2). Kendall’s
tau had too small a sample size to conduct a robust
evaluation, yet our extensive searches and contact with
investigators suggest that no further trials have been
conducted. The pooled RR, using a random effects model
for conservative application is 0.66 (95% CI: 0.59–0.73,
P£0.0001) (Fig. 3). We did not detect significant statis-
tical heterogeneity (P¼0.568, I
2
¼0%). Effects were
consistent across tumor type and dose of melatonin.
Authors reported no severe adverse events and reported
that melatonin was well tolerated in all trials. Fig. 4
displays the cumulative meta-analysis of the trials.
Discussion
Our meta-analysis indicates a consistent effect on 1-yr
survival of adjunct melatonin in a variety of advanced stage
cancers. In many cases the cancers that were being treated
Bias assessment plot
-2 -1 0 1
0.6
0.4
0.2
0.0
Lo
g
(relative risk)
Standard error
Fig. 2. Funnel plot: Eggers test: )1.260231 (approximate 95%
CI ¼)2.508723 to )0.011738) P¼0.0483), Kendall’s tau had too
small a sample size to conduct a robust evaluation.
Mills et al.
4

were refractory to standard therapy and as such more
amenable to the adjunct use of an untested and unproven
therapy like melatonin. The pooled RR was 0.66 (95% CI:
0.59–0.73). The large effect size and low number of serious
adverse events should be of interest to clinicians and
patients.
There are several strengths to our meta-analysis: we
conducted systematic searches of databases and contacted
experts in the field to identify all RCTs available; we
searched, abstracted and analyzed all data independently
and in duplicate; we evaluated important methodologic
criteria shown to influence trial outcomes; we contacted the
authors of the trials to clarify trial conduct; and we
conducted both standard and cumulative meta-analyses to
determine at what point investigators might reasonably
begin external trials to verify these findings.
There are several limitations to be considered in the
interpretation of our meta-analysis. Perhaps the most
significant is that the same network of investigators in
Italy and Poland conducted all 10 trials. While this will not
necessarily bias the results, the lack of independent verifi-
cation, particularly in the presence of an effect that is
perhaps surprisingly large, warrants skepticism. The funded
Cancer treatment centers of America (CTCA) and National
Cancer Institute (NCI) trials that are currently underway
will begin to address this concern. Nevertheless, the
cumulative meta-analysis suggests that evidence indicating
the need for externally conducted trials has been available
since 1992 (Fig. 4).
An additional concern is the methodological limitations
of the study, particularly the lack of blinding. While all
trials were limited in their original publication reporting,
n
63
113
193
233
263
293
363
613
643
743
0.5 1
combined 0.69 (0.66, 0.72)
2003 0.66 (0.59, 0.73)
2000 0.67 (0.60, 0.75)
1999 0.67 (0.60, 0.75)
1997 0.69 (0.60, 0.79)
1996 0.69 (0.59, 0.80)
1996 0.70 (0.60, 0.81)
1995 0.71 (0.60, 0.83)
1994 0.73 (0.62, 0.86)
1994 0.76 (0.63, 0.92)
1992 0.79 (0.60, 0.98)
Relative risk (95% confidence inter val)
Fig. 4. Cumulative meta-analysis of 10
trials from 1992–2003.
0.1 0.2 0.5 1 2
Renal cancer, 2000, n = 30 0.43 (0.14, 1.17)
Melanoma, 1996, n = 30 0.42 (0.16, 0.92)
Breastc ancer, 1995, n = 40 0.48 (0.24, 0.86)
Solid tumors, 1999, n = 150 0.64 (0.52, 0.78)
Solid tumors, 1994, n = 80 0.63 (0.45, 0.85)
Glioblastoma, 1996, n = 30 0.61 (0.34, 0.91)
Brainm etastases, 1994, n = 50 0.71 (0.47, 0.97)
Lung cancer, 2003, n = 100 0.53 (0.37, 0.74)
Lung cancer, 1997, n =70 0.69 (0.48, 0.96)
Lung cancer, 1992, n = 63 0.79 (0.60, 0.98)
Combined (random) 0.66 (0.59, 0.73)
Relative risk (95% confidence interval)
Fig. 3. Relative risk meta-analysis of
10 RCTs in various cancers using the
random effects model.
Melatonin in the treatment of cancer
5

contact with the lead author revealed that these trials were
conducted using standard methods of enrollment, sequence
generation and analysis. This finding is consistent with
systematic evaluations of what is reported in trials com-
pared with what was actually done [35].
The 20–40 mg dosage of melatonin shown to be effective
in reducing the risk of cancer is much higher than the
1.5–5 mg used for the treatment of insomnia and jet lag.
This raises the question of toxicity and whether or not there
are significant side effects at these higher levels of intake.
Generally, melatonin is considered relatively innocuous
even at high doses, and the trials from Italy and Poland
reported no significant side effects [18, 30, 31, 36–38]. One
of the likeliest side effects of melatonin is the tendency to
produce sedation or sleepiness in some people. While
melatonin’s antioxidant activity is not related to the time of
day, to reduce the effect of sedation, melatonin is generally
administered in the evening.
An article reviewing the safety of melatonin explored
307 articles of which nine were related to melatonin’s
adverse effects. The range of melatonin dosage involved in
the adverse reactions spanned between 1 and 36 mg. The
adverse reactions were not necessarily related to melatonin
usage and were relatively rare; they included one patient
with autoimmune hepatitis, one case of confusion caused by
melatonin overdose, one case of optic neuropathy, four
patients with fragmented sleep, one psychotic episode, one
case of nystagmus, four cases of seizures, one case of
headache and two cases of skin eruptions [39]. In addition,
there is no long-term data on the safety of ingesting high
levels of melatonin and it is possible that some adverse
effects may not be realized in the short term [40]. It should
be noted however that there has been widespread usage of
over-the-counter melatonin with little indication of post-
marketing toxicity.
In summary, this is the first meta-analysis examining the
impact of melatonin on various cancers. This shows a
strong association. The small NNT (range 3–5), low
adverse events reported and low costs related to this
intervention should be of substantial interest to patients,
physicians and policy makers. Completion of independently
conducted studies is required to confirm the efficacy and
safety of melatonin in cancer treatment.
References
1. Brzezinski A. Melatonin in humans. N Engl J Med 1997;
336:186–195.
2. Allegra M, Reiter RJ, Tan DX et al. The chemistry of
melatonin’s interaction with reactive species. J Pineal Res 2003;
34:1–10.
3. Eck KM, Yuan L, Duffy L et al. A sequential treatment
regimen with melatonin and all-trans retinoic acid induces
apoptosis in MCF-7 tumour cells. Br J Cancer 1998; 77:2129–
2137.
4. Bartsch C, Bartsch H. Melatonin in cancer patients and in
tumor-bearing animals. Adv Exp Med Biol 1999; 467:247–264.
5. Blask DE, Sauer LA, Dauchy RT. Melatonin as a chrono-
biotic/anticancer agent: cellular, biochemical, and molecular
mechanisms of action and their implications for circadian-
based cancer therapy. Curr Top Med Chem 2002; 2:113–132.
6. Blask DE, Sauer LA, Dauchy RT et al. Melatonin inhibi-
tion of cancer growth in vivo involves suppression of tumor
fatty acid metabolism via melatonin receptor-mediated signal
transduction events. Cancer Res 1999; 59:4693–4701.
7. Reiter RJ. Mechanisms of cancer inhibition by melatonin.
J Pineal Res 2004; 37:213–214.
8. Lissoni P, Rovelli F, Malugani F et al. Anti-angiogenic
activity of melatonin in advanced cancer patients. Neuroend-
ocrinol Lett 2001; 22:45–47.
9. Lissoni P, Rovelli F, Meregalli S et al. Melatonin as a new
possible anti-inflammatory agent. J Biol Regul Homeost
Agents 1997; 11:157–159.
10. Lissoni P, Paolorossi F, Tancini G et al. Is there a role for
melatonin in the treatment of neoplastic cachexia? Eur J
Cancer 1996; 32A:1340–1343.
11. Sacco S, Aquilini L, Ghezzi P et al. Mechanism of the
inhibitory effect of melatonin on tumor necrosis factor pro-
duction in vivo and in vitro. Eur J Pharmacol 1998; 343:249–
255.
12. Lissoni P. Is there a role for melatonin in supportive care?
Support Care Cancer 2002; 10:110–116.
13. Davis S, Mirick DK, Stevens RG. Night shift work, light at
night, and risk of breast cancer. J Natl Cancer Inst 2001;
93:1557–1562.
14. Bartsch C, Bartsch H, Schmidt A et al. Melatonin and
6-sulfatoxymelatonin circadian rhythms in serum and urine of
primary prostate cancer patients: evidence for reduced pineal
activity and relevance of urinary determinations. Clin Chim
Acta 1992; 209:153–167.
15. Schernhammer ES, Laden F, Speizer FE et al. Night-shift
work and risk of colorectal cancer in the nursesÕhealth study.
J Natl Cancer Inst 2003; 95:825–828.
16. Lissoni P, Barni S, Ardizzoia A et al. A randomized study
with the pineal hormone melatonin versus supportive care
alone in patients with brain metastases due to solid neoplasms.
Cancer 1994; 73:699–701.
17. Cerea G, Vaghi M, Ardizzoia A et al. Biomodulation of
cancer chemotherapy for metastatic colorectal cancer: a
randomized study of weekly low-dose irinotecan alone versus
irinotecan plus the oncostatic pineal hormone melatonin in
metastatic colorectal cancer patients progressing on 5-fluor-
ouracil-containing combinations. Anticancer Res 2003;
23:1951–1954.
18. Lissoni P, Malugani F, Bukovec R et al. Reduction of
cisplatin-induced anemia by the pineal indole 5-methoxytryp-
tamine in metastatic lung cancer patients. Neuroendocrinol
Lett 2003; 24:83–85.
19. Lissoni P, Barni S, Mandala M et al. Decreased toxicity and
increased efficacy of cancer chemotherapy using the pineal
hormone melatonin in metastatic solid tumour patients with
poor clinical status. Eur J Cancer 1999; 35:1688–1692.
20. Vijayalaxmi,Reiter RJ, Tan DX et al. Melatonin as a
radioprotective agent: a review. Int J Radiat Oncol Biol Phys
2004; 59:639–653.
21. Vijayalaxmi,Thomas CR Jr, Reiter RJ et al. Melatonin:
from basic research to cancer treatment clinics. J Clin Oncol
2002; 20:2575–2601.
22. Meade MO, Richardson WS. Selecting and appraising
studies for a systematic review. Ann Intern Med 1997;
127:531–537.
23. Altman DG. Systematic reviews of evaluations of prognostic
variables. In: Systematic Reviews in Healthcare, Meta-analysis
in Context. Br Med J 2001; 323:224–228.
Mills et al.
6

24. Higgins JP, Thompson SG. Quantifying heterogeneity in a
meta-analysis. Stat Med 2002; 21:1539–1558.
25. Lau J, Schmid CH, Chalmers TC. Cumulative meta-analysis
of clinical trials builds evidence for exemplary medical care.
J Clin Epidemiol 1995; 48:45–57; discussion 59–60.
26. Lissoni P, Barni S, Ardizzoia A et al. Randomized study with
the pineal hormone melatonin versus supportive care alone in
advanced nonsmall cell lung cancer resistant to a first-line
chemotherapy containing cisplatin. Oncology 1992; 49:336–339.
27. Lissoni P, Barni S, Tancini G et al. A randomised study with
subcutaneous low-dose interleukin 2 alone vs interleukin 2 plus
the pineal neurohormone melatonin in advanced solid neo-
plasms other than renal cancer and melanoma. Br J Cancer
1994; 69:196–199.
28. Lissoni P, Brivio O, Brivio F et al. Adjuvant therapy with
the pineal hormone melatonin in patients with lymph node
relapse due to malignant melanoma. J Pineal Res 1996;
21:239–242.
29. Lissoni P, Chilelli M, Villa S et al. Five years survival in
metastatic non-small cell lung cancer patients treated with
chemotherapy alone or chemotherapy and melatonin: a rand-
omized trial. J Pineal Res 2003; 35:12–15.
30. Lissoni P, Meregalli S, Nosetto L et al. Increased survival
time in brain glioblastomas by a radioneuroendocrine strategy
with radiotherapy plus melatonin compared to radiotherapy
alone. Oncology 1996; 53:43–46.
31. Lissoni P, Paolorossi F, Ardizzoia A et al. A randomized
study of chemotherapy with cisplatin plus etoposide versus
chemoendocrine therapy with cisplatin, etoposide and the
pineal hormone melatonin as a first-line treatment of advanced
non-small cell lung cancer patients in a poor clinical state.
J Pineal Res 1997; 23:15–19.
32. Lissoni P, Ardizzoia A, Barni S et al. A randomized study of
tamoxifen alone versus tamoxifen plus melatonin in estrogen
receptor-negative heavily pretreated metastatic breast cancer
patients. Oncol Rep 1995; 2:871–873.
33. CTCA. A Multi-Centered Randomized Trial Evaluating the
Chronotherpaeutic Role of Melatonin in the Treatment of
Stage III and IV Non-small Cell Lung Carcinoma. Expected
n¼90. www.cancercenter.com/clinicaltrials/19.cfm.
34. NCI. Melatonin and Radiation Therapy in Treating Patients
with Brain Metastases. NCT00031967. www.clinicaltrials.gov/
ct/show/NCT00031967?order=1.
35. Soares HP, Daniels S, Kumar A et al. Bad reporting does
not mean bad methods for randomised trials: observational
study of randomised controlled trials performed by the
Radiation Therapy Oncology Group. Br Med J 2004;
328:22–24.
36. Lissoni P, Mandala M, Brivio F. Abrogation of the negative
influence of opioids on IL-2 immunotherapy of renal cell
cancer by melatonin. Eur Urol 2000; 38:115–118.
37. Lissoni P, Meregalli S, Fossati V et al. A randomized study
of immunotherapy with low-dose subcutaneous interleukin-2
plus melatonin vs chemotherapy with cisplatin and etoposide
as first-line therapy for advanced non-small cell lung cancer.
Tumori 1994; 80:464–467.
38. Lissoni P, Paolorossi F, Tancini G et al. A phase II study of
tamoxifen plus melatonin in metastatic solid tumour patients.
Br J Cancer 1996; 74:1466–1468.
39. Morera AL, Henry M, de La Varga M. Safety in melatonin
use. Actas Esp Psiquiatr 2001; 29:334–337.
40. Guardiola-Lemaitre B. Toxicology of melatonin. J Biol
Rhythms 1997; 12:697–706.
Melatonin in the treatment of cancer
7