ARTICLE
Auteur(s) : B
Dréno1, S Euvrard2, C
Frances3, D Moyse4, A
Nandeuil5
1CHU Nantes, Clinique Dermatologique, Hôtel Dieu, BP
1005, 44305 Nantes Cedex, France
2Hopital Edouard Herriot, Service de Dermatologie, 5,
place d’Arsonval, 69437 Lyon Cedex 03, France
3Hopital Tenon, 4, rue de la Chine, 75970 Paris Cedex
20, France
4DM consultant, 8, rue Saint-Jean-Baptiste de la Salle,
75006 Paris, France
5Laboratoire LABCATAL, 7, rue Roger Salengro, BP 305,
92541 Montrouge Cedex France
accepté le 18 Octobre 2006
Skin cancer, especially squamous cell carcinoma (SCC), is the most
common cancer in organ graft recipients [1]. Sun exposure,
phototype, immunosuppression duration and papilloma virus
infections all seem to be strongly correlated with its occurrence
[2]. In France, all organ transplant recipients are advised to have
regular dermatological checkups, since actinic keratoses occur
significantly earlier (54.8 vs. 70.0 years) in organ transplant
recipients compared with non-transplanted controls [3]. The
incidences of pre-malignant and malignant epithelial lesions
increase with the survival time post transplantation, reaching 40%
to 60% after 20 years [4]; finally 1.5% of the carcinomas appear
within 5 years, 10% to 20% within 10 years and 40% within 20 years
[5]. The onset of these lesions differs at differing latitudes [6],
this phenomenon is related to the phototype [7].In vitro, selenium
has shown to be essential for effective immunity and protection
against oxidative damage induced by ultraviolet irradiation, and
for protecting human keratinocytes from apoptosis induced by
exposure to ultra-violet [8]. It has also been shown to enhance the
activation of glutathione peroxidase [9]. Selenium is also able to
modulate the immune response, by activating Langerhans cells [10].
The concentration of selenium directly influences the number of
Langerhans cells in the skin, and a smaller number of these cells
could compromise cutaneous immunity [11].In rats, dietary selenium
has been shown to inhibit both cancer of the colon induced by
nitrosamine [12], and skin cancer induced by UV light in hairless
mice [13]. In human beings, two studies have shown that selenium
could be an effective trace element in preventing various
carcinomas [14, 15], but not skin cancer.This multicentre study was
designed to explore the possible effect of selenium in preventing
the development of epithelial lesions linked to human papilloma
virus (HPV) in an organ transplant recipient population.At the
epidemiological level, this trial is also interesting in that a
dermatologist started monitoring the skin of organ transplant
recipients just after transplantation.
Material and methods
The study was designed as a 5-year, (3 years of treatment plus 2
years of follow up), multicentre, randomised, placebo-controlled,
parallel-group trial. Patients were seen by a dermatologist before
grafting; and any patients presenting with a non-malignant or
malignant skin keratosis or viral warts that had been present for
less than 3 months were not selected. Within 10 weeks
following the graft, a second visit was performed by a
dermatologist to check that no new cutaneous lesion had appeared.
Patients were then randomised to the selenium group or to the
placebo group.
The other inclusion criteria were as follows: male and female
outpatients, aged between 18 and 65 years, having undergone a
kidney, liver or heart transplant within the previous 10 weeks, and
showing no sign of acute rejection when included. Such patients
were eligible, provided they gave their written informed consent.
Exclusion criteria included any disease or condition requiring
treatment with selenium salts, acid vitamin A, retinoids, zinc
gluconate, vitamin A, E or C or interferon, and a severe condition
or an allergy to any of the components of the treatment being
tested.
At inclusion (M0), the patients were randomly allocated either
to 200 μg of selenium (as selenium-enriched yeast) or a
matching placebo. The treatment was to be taken with breakfast
every morning for three years (M0 to M36). The patients then
entered a 2-year, treatment-free, follow-up period (M36 to M60).
Visits were scheduled every 4 months during the first year and
every 6 months for the remaining 4 years, or as soon as possible in
case of unscheduled withdrawal. This schedule complied with that of
normal graft follow-up.
Assessments performed at these visits included a complete
clinical examination of skin to evaluate any fresh skin lesion(s)
and record any adverse events reported by the patient. The type,
location, number, duration and treatment options for each skin
lesion observed was reported by the investigator. Non-malignant
skin lesions could be treated with laser, electrocoagulation or
liquid nitrogen, and any malignant or suspect lesions were removed
for histological diagnosis. Five-ml blood samples were collected
into special dry Vacutainer for trace element analysis, before the
patient had taken the study treatment (M0) and after 4 (M4), 12
(M12), 24 (M24) and 36 (M36) months of Selenium or placebo
supplementation (figure
1). Samples were left for one hour at room temperature
before being centrifuged for 10 min at 3000 rpm. Plasma
samples were stored in polypropylene tubes at – 20 °C
until analysis. An atomic absorption spectrometer with a Zeeman
background correction equipped with a furnace autosampler was used
for all the Selenium analyses. No other laboratory test was
scheduled, as the normal graft follow-up already included a large
number of lab parameters, which were available for the
dermatologist at the graft centre. Compliance was assessed by
questioning the patient and counting the material returned. The
study was conducted according to the Helsinki Declaration, and to
European Good Clinical Practice. The protocol and consent form had
been approved by a National Ethics Committee prior to starting the
study. Written informed consent was mandatory. An independent
Scientific Committee was responsible for checking that the trial
had been conducted properly and for validating the data before
analysis.
The main efficacy criterion was based on the occurrence of warts
and various keratoses. Various keratoses included condyloma, plan
wart, common wart, papilloma, porokeratoses and actinic keratoses;
an occurrence of skin cancer (including basal cell carcinoma,
keratoacanthoma, squamous cell carcinoma, leucokeratoses and
Bowen’s disease) was a secondary efficacy criterion. All other skin
lesions were taken into consideration for assessing dermatological
safety. General safety was assessed by adverse event
monitoring.
The calculation of the sample size required for this study was
based on the annual expected occurrence of warts and various
keratoses in 17% of the placebo group and in 10% of the selenium
group, i.e. a difference of 7% between the two treatment groups.
This led to a total expected rate after 3 years of 51% in the
placebo group and 30% in the selenium group; with a two-way α risk
of 0.05, and a power of 0.80, this meant that at least 84 patients,
extended to 100, had to be enrolled per group. A Fisher’s exact
test was used for comparing all qualitative variables between the
two treatment groups, including the primary criterion.
Non-parametric tests were used for quantitative variables, the
Kaplan-Meyer method was used to compare the warts and various
keratoses and skin cancer first occurrence rates between treatment
groups versus the overall population and by sub-groups (graft, age,
phototype) and the different type of non-malignant lesions (warts
and various keratoses).
Results
Description of the population
Of 196 patients who were randomised in the study, 184 received
treatment (91 selenium and 93 placebo) (figure 2). They were
included in the full analysis set (FAS). During the treatment
phase, 38 in the selenium group and 37 in the placebo group
withdrew from the study. This distribution was similar in both
treatment groups. Reasons for discontinuation are described in
Table 1.
There were no significant differences between the 2 treatment
groups (Table 2). The overall mean trial
duration was 33.66 ± 15.31 months and treatment duration was 27.51
± 11.55 months.
Selenium levels were in the normal range at baseline in both
groups, and subsequently increased in the selenium group and
thereafter remained stable throughout the treatment period (figure 3). This
confirmed the bioavailability of selenium-enriched yeast, and the
compliance of the treated patients.
Comparison of the treatment groups
Table 3( Table 3 ) shows the incidence
of the first observed lesion per year in each treatment group,
globally and by type of lesion. The Kaplan Meyer analyses for the
main efficacy criteria did not reveal any significant difference
between the two groups (Odd ratio = 1.09; p = 0.72; (figure 4). With regard to
the incidence of skin cancer, there was no significant difference
between the two groups, with 6.6% of patients in the selenium group
and 2.2% in the placebo group presenting with a skin cancer (Odds
Ratio = 3.08; p = 0.15; (figure 5)). Details of the
different types of skin cancer are described in Table 4. The lack of difference for the main
efficacy criterion made it possible to pool both arms for a global
description of epithelial events.
Table 1 Numbers of patients withdrawn from the study
and the main reasons for withdrawal
|
Main reason for discontinuation
|
Selenium
|
Placebo
|
Total
|
|
N = 91
|
N = 94
|
N = 185*
|
|
Adverse event
|
12
|
10
|
22
|
|
Bad compliance
|
2
|
4
|
6
|
|
Lost to follow up
|
2
|
3
|
5
|
|
Administrative reason
|
1
|
2
|
3
|
|
Consent withdrawn
|
21
|
18
|
39
|
|
Total
|
38
|
37
|
75
|
Table 2 Demography and baseline characteristics (D0) -
FAS population
|
Criterion
|
|
Selenium
|
Placebo
|
Comparison test*
|
|
N = 91
|
N = 93
|
|
Sex n (%)
|
Male:
|
60 (65.9%)
|
67 (72.0%)
|
p = 0.43
|
|
Female:
|
31 (34.1%)
|
26 (28.0%)
|
|
Age (years)
|
Mean ± SD
|
44.3 ± 13.0
|
44.4 ± 10.7
|
p = 0.84
|
|
Weight
|
Mean ± SD
|
66.8 ± 14.1
|
64.5 ± 12.1
|
P = 0.44
|
|
Ethnic group (%)
|
Caucasian:
|
89%
|
88%
|
P = 1.00
|
|
Other:
|
11%
|
12%
|
|
Eye colour
|
Blue:
|
32%
|
19%
|
P = 0.15
|
|
Green:
|
13%
|
17%
|
|
Brown/Black:
|
55%
|
64%
|
|
U.V. exposure (%)
|
Yes:
|
54%
|
57%
|
P = 0.77
|
|
No:
|
46%
|
43%
|
|
Time since graft (weeks)
|
Mean ± SD
|
5.06 ± 2.54
|
6.03 ± 4.47
|
p = 0.19
|
|
Grafted organ n (%)
|
Heart
|
9 (9.9%)
|
10 (10.8%)
|
p = 1.0000
|
|
Liver
|
9 (9.9%)
|
10 (10.8%)
|
|
Kidney
|
69 (75.8%)
|
69 (74.2%)
|
|
Kidney + pancreas
|
4 (4.4%)
|
4 (4.3%)
|
Table 3 Occurrence of the first lesion, cumulative per
year - FAS population
|
Type of lesion
|
Selenium
|
Placebo
|
Fisher’s
|
|
N = 91
|
N = 93
|
|
[Year]
|
Lesion
|
Lesion
|
test p
|
|
n (%)
|
n (%)
|
|
Non-malignant and malignant epithelial lesion
|
|
[0 , 1]
|
22 (24.2)
|
17 (18.3)
|
0.37
|
|
[0 , 2]
|
32 (35.2)
|
27 (29.0)
|
0.43
|
|
[0 , 3]
|
35 (38.5)
|
31 (33.3)
|
0.54
|
|
[0 , 4]
|
35 (38.5)
|
32 (34.4)
|
0.65
|
|
Non-malignant epithelial lesion only
|
|
[0 , 1]
|
22 (24.2)
|
17 (18.3)
|
0.37
|
|
[0 , 2]
|
31 (34.1)
|
26 (28.0)
|
0.43
|
|
[0 , 3]
|
33 (36.3)
|
30 (32.3)
|
0.64
|
|
[0 , 4]
|
33 (36.3)
|
31 (33.3)
|
0.76
|
|
Verruca plana or Verruca vulgaris
|
|
[0 , 1]
|
16 (17.6)
|
15 (16.1)
|
0.85
|
|
[0 , 2]
|
23 (25.3)
|
19 (20.4)
|
0.48
|
|
[0 , 3]
|
26 (28.6)
|
24 (25.8)
|
0.74
|
|
[0 , 4]
|
26 (28.6)
|
25 (26.9)
|
0.87
|
|
Actinic keratoses and porokeratoses
|
|
[0 , 1]
|
9 (9.9)
|
3 (3.2)
|
0.08
|
|
[0 , 2]
|
12 (13.2)
|
8 (8.6)
|
0.35
|
|
[0 , 3]
|
12 (13.2)
|
9 (9.7)
|
0.49
|
|
[0 , 4]
|
12 (13.2)
|
9 (9.7)
|
0.49
|
|
Malignant Epithelial Lesion
|
|
[0 , 1]
|
1 (1.1)
|
0 (0.0)
|
0.49
|
|
[0 , 2]
|
4 (4.4)
|
1 (1.1)
|
0.21
|
|
[0 , 3]
|
6 (6.6)
|
2 (2.2)
|
0.17
|
|
[0 , 4]
|
6 (6.6)
|
2 (2.2)
|
0.17
|
Table 4 Number and type of malignant lesions by
patient
|
Patient Number
|
Basal cell carcinoma
|
Bowen’s disease
|
Keratoacanthoma
|
Leucokeratoses
|
Spinocellular carcinoma
|
|
Placebo
|
5
|
|
1
|
|
|
|
|
91
|
1
|
|
1
|
|
|
|
Selenium
|
35
|
2
|
|
|
|
1
|
|
55
|
|
|
1
|
|
|
|
107
|
1
|
|
|
|
|
|
125
|
|
|
|
1
|
|
|
191
|
|
|
|
|
3
|
|
218
|
2
|
|
|
|
|
Global analyses
As shown in table 2, 146 patients underwent a kidney graft and 19 a
heart graft. Within the treatment period, 54 (37.0%) patients in
the kidney group and 3 (15.8%) in the heart group presented with an
epithelial lesion (non-malignant or malignant) with median
occurrence times of 260 days and 567 days respectively. This
difference was statistically significant (p = 0.05), revealing a
33% greater risk of presenting an epithelial lesion in the kidney
group versus the heart group (Odds Ratio = 0.33 [0.10, 1.06]).
The same type of analysis was performed for the phototype. The
100 patients who had phototype I, II or III formed one group, and
the 84 patients who had a phototype IV, V or VI formed the other.
There was no significant difference between the two groups. The
same comparison was performed between patients who were more or
less than 45 years old. Eighty-nine patients were ≤ 45 years old
and 95 patients > 45 years old. There was no significant
difference between the two groups.
The number and the types of malignant lesions by patients are
described in table 4. Two (2.2%) patients in the placebo group and
6 (6.6%) in the selenium group presented with malignant lesions.
There was no significant difference between the treatment
groups.
Regardless of the treatment group, 6 of the 8 patients who
presented with a malignant lesion had had an actinic keratosis and
2 had not. This difference was statistically significant (p <
0.001), showing that the Relative Risk (RR) of having a malignant
lesion was 17.5 (RR: 17.5 [3.7-82.0]) when this lesion was preceded
by an actinic keratosis.
Safety analyses
Safety was good. In the selenium group, 18 events reported by 12
(13.9%) patients were classified as being related to treatment by
the investigator; versus 18 events reported by 9 (9.57%) patients
in the placebo group.
Discussion
Although mechanisms through which selenium exerts a preventive role
against cancer are not totally elucidated and, beyond the reduction
of oxidative stress, it has been shown that selenium compounds are
able to inhibit cell growth and to induce apoptosis in vitro, this
could explain how they reduce the outgrowth of tumor cells in vivo
[16]. Tumor cells could be more sensitive to the induction of
apoptosis by some selenium compounds compared to normal cells [17].
A recent review [18] indicates that many clinical studies with
selenium have shown benefits in reducing the risk of cancer
incidence and mortality in all cancers combined, and specifically
in liver, prostate, colorectal and lung cancers.
Our study using selenium-enriched yeast containing 200 μg
selenium/day did not show any effect on the prevention of the
occurrence of warts or various keratoses induced by HPV in grafted
immuno-suppressed patients. Nor did our study demonstrate any
preventive effect on the development of skin cancer. Two major
recent preventative studies, the NPC study [14] and the SUVIMAX
study [15], used selenium. In the former study, 1312 subjects
presenting with previous skin carcinoma were investigated. This
study demonstrated that selenium did not have any preventative
effect on the occurrence of new basal cell carcinomas, but an
increased risk of squamous cell carcinoma was observed. However, as
secondary criteria, patients presented with fewer prostate, colon,
and breast cancers in the selenium group, (p = 0.001) and overall
cancer mortality was reduced in this group (p = 0.002).
A similar phenomenon was observed in our study, where only two
(2.2%) patients presented with malignant lesions in the placebo
group versus 6 in the selenium group 6 (6.6%). However, there was
no significant difference between the treatment groups.
More recently in the SUVIMAX study, women were not protected by
the supplementation, and actually displayed an increased risk of
presenting skin carcinomas (except basal cell carcinoma).
In the NPC study, selenium supplementation did not demonstrate
any protective effect in patients with a high selenium
concentration, and indeed the risk of presenting a non-melanoma
skin cancer was increased. In spite of results showing the role
played by selenium acting on normal or tumoral skin cells [16, 19,
20] no increased protection against skin cancer has been observed
as a result of selenium supplementation. These results confirm in
immuno-depressed patients the results observed in studies with
selenium in non immuno-suppressed patients. These two studies were
not published at the onset of our study. This could be related to
the selenium dosage used, but at this time we are unable to confirm
this hypothesis.
These findings raise the question of the value of selenium
supplementation in patients who do not have a low selenium level,
and even the possible risk of inducing cutaneous cell carcinomas.
This implies that another prospective study is required to answer
to this question.
A very important finding to emerge from our study should be
pointed out: epithelial lesions appeared at the much higher rate of
35% within the three years following the graft, versus the average
rate of 15% usually described in the literature. This fact has been
observed both for warts and for various keratoses. This finding,
which has been confirmed in a large number of patients, probably
highlights the impact of a specific methodological approach, which
involves dermatologists in the follow up of the skin as soon as the
grafts are performed. This has never been done in any previous
study reported in the literature. This suggests that
dermatologists, who are better trained to detect early cutaneous
lesions, should indeed be involved in such a process.
Co-factors did not seem to play a role, due to the fact that
both arms were similar in terms of type of graft, post surgery
follow up, immunological status, kidney function, phototype and age
of the graft. No concomitant viral infections were reported by the
investigator, in any group, which could have influenced the results
of the study. Finally we have not identified any sub-groups of
patients.
Our study also shows that the time of onset of epithelial
lesions is unpredictable. We observed that the onset of skin cancer
seems to be linked to a previous history of actinic keratoses (risk
increased by 17). In conclusion, this first study to be performed
in immuno-suppressed patients using a double-blind, randomized,
placebo-controlled methodology, demonstrates that 200 μg/day
of selenium in the form of selenium enriched yeast, has no
preventative effect against the occurrence of warts or various
keratoses, nor for skin cancer but nevertheless it does demonstrate
the value of starting early dermatological follow-up after the
graft.
Acknowledgments
Participating investigators: Prof. P. Amblard (Grenoble), Prof. S.
Aractingi (Paris), Dr D. Augias (Montpellier), Prof. J.C. Béani
(Grenoble), Dr P. Bécherel (Paris), Dr F. Berthod (Grenoble), Dr D.
Bessis (Montpellier), Dr E. Bourrat (Paris), Prof. J.
Chevrant-Breton (Rennes), Prof. A. Claudy (Lyon), Dr S. Dalac
(Dijon), Dr M. D’Incan (Clermont-Ferrand), Prof. M. Faure (Lyon),
Prof. C. Frances (Paris), Prof. J.J. Guilhou (Montpellier), Prof.
G. Guillet (Brest), Dr O. Jumbou (Nantes), Dr J. Kanitakis (Lyon),
Prof. D. Lambert (Dijon), Prof. C. Lebbé (Paris), Prof. M.T. Leccia
(Grenoble), Dr B. Legoux (Nantes), Prof. D. Leroy (Caen), Dr S.
Louvet (Caen), Dr P. Monseux (Charleville-Mézières), Prof. P. Morel
(Paris), Dr F. Mouly (Paris), Dr B. Sassolas (Brest), Dr M.
Schollhammer (Brest), Pr. P. Souteyrand (Clermont-Ferrand), Dr E.
Wetterwald (Paris).
This trial was funded by Labcatal Pharmaceuticals, France.
References
1 Gupta AK, Cardella CJ, Haberman HF. Cutaneous
malignant neoplasm in patients with renal transplants. Arch
Dermatol 1986; 122: 1288-93.
2 Barr BBB, Mc Laren K, Smith IW, et al.
Human papillomavirus infection and skin cancer in renal allograft
recipients. Lancet 1989; 1: 124-9.
3 Boyd AS, Stasko T, Cameron GS, et al.
Histologic features of actinic keratoses in solid organ transplant
recipients and healthy controls. J Am Acad Dermatol 2001; 45:
217-21.
4 Ulrich C, Schmook T, Nindl I, et al.
Cutaneous precancers in organ transplant recipients: an old enemy
in a new surrounding. Br J Dermatol 2003; 149(suppl 66): 40-2.
5 Euvrard S, Kanitakis J, Claudy A. Skin cancers
after organ transplantation. N Engl J Med 2003; 348: 1681-91.
6 Bouwes-Bavinck JN, Hardie DR, Green A,
et al. The risk of skin cancer in renal transplant recipients
in Queensland, Australia: a follow-up study. Transplantation 1996;
61: 715-21.
7 Espana A, Martinez-Gonzalez MA,
Garcia-Granero M, et al. A prospective study of incident
nonmelanoma skin cancer in heart transplant recipients. J Invest
Dermatol 2000; 115: 1158-60.
8 Rafferty TS, Beckett GJ, Walker C, et al.
Selenium protects primary human keratinocytes from apoptosis induce
by exposure to ultraviolet radiation. Clin Exp Dermatol 2003;
28(3): 294-300.
9 Pence BC, Delver E, Dunn DM. Effects of dietary
selenium on UVB-induced skin carcinogenesis and epidermal
antioxydant status. J Invest Dermatol 1994; 102: 759-61.
10 Celerier P, Richard A, Litoux P, Dreno B.
Modulatory effects of selenium and strontium salts on
keratinocytes-derived inflammatory cytokines. Arch Dermatol Res
1995; 287(7): 680-2.
11 Rafferty T, Norval M, El-Ghorr A, et al.
Dietary selenium levels determine epidermal langerhans cell numbers
in mice. Biol Trace Elem Res 2003; 92(2): 161-72.
12 Birt DF, Lawson TA, Julius AD, Kunice CE,
Salamsi S. Inhibition by dietary selenium of colon cancer
induced in the rat by bis (2-oxopropyl) nitrosamine. Cancer Res
1982; 42: 4456-9.
13 Overvad K, Thorling EB, Bjerring P,
Ebbesen P. Selenium inhibits UV-light-induced skin
carcinogenesis in Hairless mice. Cancer Lett 1985; 27: 163-70.
14 Clark LC, Combs Jr. GF, Turnbill BW,
et al. Effects of selenium supplementation for cancer
prevention in patients with carcinoma of the skin a randomized
controlles trial. JAMA 1996; 276: 1957-63.
15 Hercberg S, Galan P, Prezosi P, et al.
The SU.VI.MAX study, a randomized, placebo-controlled trial of the
health effects of antioxidant vitamins and minerals. Arch Intern
Med 2004; 164: 2335-42.
16 Fleming J, Ghose A, Harrison PR. Molecular
mechanism of cancer prevention by selenium compounds. Nutr Cancer
2001; 40(1): 42-9.
17 Menter DG, Sabichi AL, Lippman SM. Selenium
effects on prostate cell growth. Cancer Epidemiol Biomarkers Prev
2000; 9: 1171-82.
18 Rayman MP. Selenium in cancer prevention: a review of
the evidence and mechanism of action. Proc Nutr Soc 2005; 64(4):
527-42.
19 Brash DE. Roles of the transcription factor p53 in
keratinocyte carcinomas. Br J Dermatol 2006; 154(suppl 1):
8-10.
20 Bandura L, Drukala J, Wolnicka-Glubisz A,
et al. Differential effects of selenite and selenate on human
melanocytes, keratinocytes, and melanoma cells. Biochem Cell Biol
2005; 83: 196-211.
|
Printable version
Version PDF
