Cytokine gene single nucleotide polymorphisms and susceptibility to and prognosis in cutaneous malignant melanoma

Abstract

Cutaneous malignant melanoma (CMM) is a potentially fatal malignancy in which exposure to UV light is the most important risk factor. Several lines of evidence suggest that CMM patients develop an immune response to their tumours, although, in most cases, anti-tumour immune responses are insufficient to abrogate tumour development. Polymorphism in genes regulating the immune response and cell growth may result in increased susceptibility to and/or poorer prognosis in certain individuals. In this study, we addressed whether single nucleotide polymorphisms (SNPs) associated with differential expression of selected pro- and anti-inflammatory cytokines and growth factors [interleukin (IL)-1beta-35 and -511, IL-2 -330, IL-4 -590, IL-6 -174, IL-8 -251, interferon (IFN)-gamma+874 and transforming growth factor (TGF)beta1 +915] or as markers of candidate cytokine genes (IL-12 +1188) are associated with susceptibility to or known prognostic indicators (e.g. initial tumour growth phase, Breslow thickness, mitotic count in vertical growth phase tumours, tumour regression) in CMM. One hundred and sixty-nine British caucasian CMM patients and 261 controls were included in the study and all SNPs were genotyped by ARMS-PCR. No SNP genotypes or alleles showed significant associations with CMM susceptibility and only the IL-1beta-511 TT genotype was associated with thinner invasive tumours at presentation, as assessed by Breslow thickness at the clinically significant cut-off point of 1.5 mm [occurring in 2/51 (3.9%) thicker vs. 14/78 (17.9%) thinner tumours (P = 0.03; relative risk = 0.29 (95% confidence interval 0.05-0.95)]. These findings suggest that - with the possible exception of IL-1beta- genetic variation associated with differential expression of the selected pro- and anti-inflammatory cytokines is unlikely to play a major role in susceptibility to and prognosis in CMM.

Full text

© 2003 Blackwell Publishing Ltd, European Journal of Immunogenetics 30, 409–414 409
Blackwell Publishing Ltd.
Cytokine gene single nucleotide polymorphisms and susceptibility to and
prognosis in cutaneous malignant melanoma
W. M. Howell,*† S. J. Turner,* J. M. Theaker‡ & A. C. Bateman‡
Summary
Cutaneous malignant melanoma (CMM) is a potentially
fatal malignancy in which exposure to UV light is the
most important risk factor. Several lines of evidence
suggest that CMM patients develop an immune response
to their tumours, although, in most cases, anti-tumour
immune responses are insufficient to abrogate tumour
development. Polymorphism in genes regulating the
immune response and cell growth may result in increased
susceptibility to and/or poorer prognosis in certain
individuals. In this study, we addressed whether single
nucleotide polymorphisms (SNPs) associated with differ-
ential expression of selected pro- and anti-inflammatory
cytokines and growth factors [interleukin (IL)-1β −35
and −511, IL-2 −330, IL-4 −590, IL-6 −174, IL-8 −251,
interferon (IFN)-γ +874 and transforming growth factor
(TGF)β1 +915] or as markers of candidate cytokine genes
(IL-12 +1188) are associated with susceptibility to or
known prognostic indicators (e.g. initial tumour growth
phase, Breslow thickness, mitotic count in vertical growth
phase tumours, tumour regression) in CMM. One hundred
and sixty-nine British caucasian CMM patients and
261 controls were included in the study and all SNPs were
genotyped by ARMS–PCR. No SNP genotypes or alleles
showed significant associations with CMM susceptibility
and only the IL-1β −511 TT genotype was associated with
thinner invasive tumours at presentation, as assessed by
Breslow thickness at the clinically significant cut-off point
of 1.5 mm [occurring in 2/51 (3.9%) thicker vs. 14/78
(17.9%) thinner tumours (P = 0.03; relative risk = 0.29
(95% confidence interval 0.05–0.95)]. These findings
suggest that — with the possible exception of IL-1β —
genetic variation associated with differential expression
of the selected pro- and anti-inflammatory cytokines is
unlikely to play a major role in susceptibility to and
prognosis in CMM.
Introduction
Cutaneous malignant melanoma (CMM) is a serious
and potentially fatal malignancy which is increasing in
incidence among most caucasian populations, where the
most important risk factor is exposure to ultraviolet (UV)
light (Cress & Holly, 1997). Several lines of evidence
suggest that CMM patients develop an immune response
to their tumours (Wolfel et al., 1993) and rare cases of
spontaneous regression have been reported. However, in
most cases, anti-tumour immune responses in CMM are
insufficient to abrogate tumour development. Poly-
morphism in genes regulating the immune response and
cell growth may constitute one mechanism leading to
inter-individual differences in anti-tumour immune respon-
ses, resulting in increased susceptibility to and /or poorer
prognosis in certain individuals. In support of this, a
number of studies have variably suggested that HLA class
II DQB1/DRB1 alleles and/or associated haplotypes may
contribute to susceptibility to and/or prognosis in CMM
(Lee et al., 1996; Bateman et al., 1998; Lulli et al., 1998).
We have also obtained preliminary data suggesting that
promoter polymorphisms associated with differential
expression of interleukin (IL)-10 may be associated with
both susceptibility to and markers of prognosis in CMM,
with ‘high expression’ genotypes associated with pro-
tection from disease and ‘low expression’ genotypes
associated with susceptibility/disease severity (Howell
et al., 2001). Confirmatory evidence for a role for IL-10
polymorphism in determining tumour progression has
recently been provided by an independent study (Martinez-
Escribano et al., 2002).
Based on the above, the present study was performed
to investigate whether single nucleotide polymorphisms
(SNPs) in a panel of pro- and anti-inflammatory cytokine
and growth factor genes are associated with susceptibility
to and/or markers of prognosis in CMM. The following
SNPs were selected for study: IL-1β −35 (also referred to
as −31 in the literature) and −511, IL-2 −330, IL-4 −590,
IL-6 −174, IL-8 −251, IFNγ +874 and TGFβ1 +915, based
on likely or reported associations between SNP genotype
or associated haplotype and differential expression of
* Department of Human Genetics, University of Southampton,
† Histocompatibility and Immunogenetics Laboratory and
‡ Department of Cellular Pathology, Southampton General Hospital,
Southampton, UK.
Received 8 April 2003; revised 6 August 2003;
accepted 15 September 2003
Correspondence: Dr W. M. Howell, Histocompatibility and
Immunogenetics Laboratory, Duthie Building, Southampton
General Hospital, Tremona Road, Southampton SO16 6YD, UK.
Tel.: + 44-23-8079-6918; Fax: + 44-23-8070-1416; E-mail:
wmh1@soton.ac.uk

410 W. M. Howell et al.
© 2003 Blackwell Publishing Ltd, European Journal of Immunogenetics 30, 409–414
their respective gene product (Herrera et al., 2000;
Hajeer et al., 1998; Rosenwasser et al., 1995; Kornman
et al., 1997; Fishman et al., 1998; John et al., 1998; Sant-
tila et al., 1998; Hull et al., 2000; Pravica et al., 2000; see
also reviews by Bidwell et al., 1999; Haukim et al.,
2002), or utility as marker SNPs for candidate genes
(IL-12 +1188) (Hall et al., 2000). These cytokines were
also selected for study due to their putative role in
melanomagenesis. For example, IL-1β, IL-6 and IL-8
expression is associated with tumour progression in
CMM (Moretti et al., 1999), although IL-6 may act as a
tumour growth inhibitor for early stage melanomas and
as a growth factor for tumour cells at more advanced
stages of disease (Lu & Kerbel, 1992). IL-2 enhances the
growth of human melanoma cell lines derived from
primary (but not from metastatic) tumours (Garcia-Vazquez
et al., 2000), while IL-2 gene transduction — alone or in
combination with IL-4 — may reduce melanoma cell line
tumourigenicity (Hollingsworth et al., 1996). IL-12 gene
transduction has also been shown to have an anti-
tumour effect in mouse melanoma cells (Nagai et al.,
2000), while IFNγ may also be involved in melanoma cell
growth regulation (Krasagakis et al., 1993). Finally,
TGFβ1 was included in the study, since this multifunc-
tional growth and differentiation factor is over-expressed
in melanoma cell lines (Rodeck et al., 1994), increased
expression of TGFβ1 has been demonstrated in invasive
primary melanoma and metastatic nodules (Van Belle
et al., 1996), while elevated serum levels of TGFβ1 have
been demonstrated in patients with disseminated malig-
nant melanoma (Krasagakis et al., 1998).
A possible role for the above SNPs in determining
susceptibility to CMM was investigated in a case-control
study, utilizing DNA extracted from diagnostic histo-
pathological formalin-fixed, wax-embedded tissues from
CMM patients. In addition, within the CMM patient
series, possible cytokine SNP associations with known
clinicopathological markers of CMM prognosis were
investigated. These markers included initial growth phase
(horizontal vs. vertical), mitotic count within vertical
growth phase tumours, Breslow depth of invasive CMM,
presence of tumour infiltrating lymphocytes and presence
or absence of tumour regression (Clark et al., 1989).
Among these factors, Breslow depth at presentation is the
most important prognostic indicator in CMM, with 5-year
survival values ranging from 93% for tumours < 1.5-mm
thick, compared with 67% for 1.5–3.49 mm tumours, to
37% for tumours > 3.5-mm thick (Mackie et al., 1992).
Materials and methods
Patients
Formalin-fixed and paraffin-embedded tissue blocks
from 169 CMM patients (presenting in 1986–93) were
retrieved from the files of the Histopathology Depart-
ment, Southampton General Hospital. All specimens were
re-reviewed by two histopathologists (ACB and JMT)
and the original diagnoses of CMM confirmed. Ethical
approval for the use of this material was granted by the
Southampton and South-west Hants Local Research
Ethics Committee (No. 066/00).
Tumour histopathology data
Histopathological prognostic features of each case were
assessed, as defined in the literature and used in previous
studies (Clark et al., 1989; Bateman et al., 1998) with initial
evaluation of the growth phase and thickness (Breslow
depth) of the tumour. Radial growth phase lesions
were defined as those limited in extent to the epidermis
(melanoma in situ) or showing early invasion of the upper
dermis, but with dermal nests of melanocytes no larger
than those at the dermoepidermal junction, and containing
no mitotic figures. Vertical growth phase lesions showed
expansile growth within the dermis, evidenced by nests of
neoplastic melanocytes that were larger than those at the
dermoepidermal junction or the presence of mitotic
figures within dermally located melanocytes (Clark et al.,
1989). For vertical growth phase CMM, the mitotic count
per square millimetre of tumour was assessed as nil, 1–6
or > 6 and the number of tumour infiltrating lymphocytes
(TILs) was evaluated as absent, non-brisk/focal or brisk
(Clark et al., 1989). The presence of tumour regression,
defined as segmental tumour loss, was also recorded.
Clinical follow-up data
The following variables were recorded for each patient,
subject to availability of clinical data: gender, age, site of
CMM, length of clinical follow-up, presence of recurrent
or metastatic tumour, disease-free survival and overall
survival time. The clinicopathological stage of each patient
at initial presentation for whom full data were available
was calculated using the tumour, nodes, metastases system
(Beahrs & Myers, 1983).
Controls
Cancer-free controls consisted of stored DNA samples
derived from 261 cadaveric and non-cadaveric solid organ
and bone marrow donors. All patients and donors were of
Caucasian ethnic origin.
DNA extraction
DNA was extracted from formalin-fixed, paraffin-wax-
embedded tissue blocks as described previously (Bateman
et al., 1998). Briefly, five 10-µm sections were cut from each
tissue block and de-waxed in xylene (Merck Ltd, Poole,
UK) and xylene-ethanol washes. DNA was extracted from
the resulting cellular material by proteinase K digestion.
Cytokine SNP genotyping by ARMS–PCR
All genotyping was performed by allele-specific ARMS–
PCR methodology. Methods for genotyping the IL-1β −511
and IL-8 −251 SNPs have been described previously

Cytokine polymorphisms in malignant melanoma 411
© 2003 Blackwell Publishing Ltd, European Journal of Immunogenetics 30, 409–414
(McCarron et al., 2002). ARMS–PCR methods for geno-
typing the IL-1β −35, IL-2 −330, IL-4 −590, IL-6 −174,
IL-12 +1188, IFNγ +874 and TGFβ1 +915 SNPs were
developed as part of this study. Internal control PCR
primers were included in each reaction and allele specific
and internal control primer sequences and PCR product
sizes are given in Table 1.
All PCR reactions were performed in 10-µL reaction
volumes and final reagent concentrations were as follows:
1 × AS reaction buffer (ABgene, Epsom, UK), 200 µm
each dNTP, 12% (w/v) sucrose, 200 µm cresol red, 1 µm
each specific/common primer, 0.2 µm each internal
control primer (where appropriate), 0.25 units Thermo-
primePLUS DNA polymerase (ABgene) and 25 –100 ng DNA.
Optimised MgCl2 concentrations for each SNP are given
in Table 1. PCR reactions were performed using a 9600
Thermal Cycler (Applied Biosystems, Foster City, CA,
USA), according to the following cycling conditions:
1 min at 96°, followed by 10 cycles of 96° for 15 s, Ta° for
each SNP (Table 1) for 50 s, 72° for 40 s; then 20 cycles
of 96° for 10 s, 60° for 50 s, 72° for 40 s. PCR products
were loaded directly onto 2% agarose gels (containing
0.5 mg/mL ethidium bromide), electrophoresed and
visualised by photography under UV transillumination.
Statistical analysis
The frequencies of each cytokine SNP genotype and allele
were calculated and compared between the CMM patient
series and the controls, using × 2 analysis on 2 × 2 tables,
according to the method of Svejgaard & Ryder (1994).
Fisher’s exact P-values were calculated for analyses in
which one or more variables within 2 × 2 tables was less
than 5. Genotype frequencies were tested for agreement
with Hardy–Weinberg equilibrium using × 2 analysis
based on likelihood theory, using estimates of allele
frequencies. Genotype frequencies were also compared
within patient subgroups according to tumour growth
phase, Breslow depth (invasive CMM), mitotic index
(vertical growth phase CMM), clinicopathological phase
at presentation and the presence of disease recurrence/
metastasis. Statistical analyses were performed using SPSS
Version 10 (SAS Institute, Chicago, IL, USA). Relative
risks (RR) with 95% confidence intervals (CI) were
calculated where a particular genotype/allele showed a
significantly increased or decreased incidence within a
particular patient subgroup.
Results
Between 142 and 169 CMM cases and between 160 and
261 controls were genotyped for each cytokine SNP,
depending upon DNA availability. Genotype and allele
frequencies for all cytokine SNPs were compared between
CMM patients and controls and results are presented in
Tables 2 and 3. All genotype frequencies in both CMM
patients and controls were distributed in accordance with
Hardy–Weinberg equilibrium (at P = 0.01 for IL-2 −330
in CMM patients and at P = 0.05 for all remaining SNPs).
From these tables it can be seen that there were no
significant differences in genotype or allele frequencies for
any of the SNPs studied.
Genotype and allele frequencies for all cytokine SNPs
were also compared within the patient series, stratified
Table 1. Cytokine SNP PCR primers and conditions
SNP Primer name
Primer sequence
(5′ to 3′)
Product size
(bp)
Final Mg2+
conc. (mM)
Ta (1st 10 cycles)
(°C)
IL-1β −35 IL-1β −35 Common TAGCACCTAGT TGTAAGGAAGA 159 2.75 63.5
IL-1β −35 C CCTACTTCTGCTTTTGAAAGCC
IL-1β −35 T CCTACTTCTGCTTTTGAAAGCT
IL-2 −330 IL-2 −330 Common ACGCCT TCTGTATGAAAC 104 2.0 55.0
IL-2 −330 T TCACATGTTCAGTGTAGTTTTAT
IL-2 −330 G TCACATGT TCAGTGTAGTTTTAG
IL-4 −590 IL-4 −590 Common AGTACAGGTGGCATCT TGGAAA 131 1.9 65.5
IL-4 −590 T CTA AACTTGGGAGAACATTGT T
IL-4 −590 C CTA AACTTGGGAGAACATTGTC
IL-6 −174 IL-6 −174 Common T T TGT TGGAGGGTGAGGGTGG 108 2.75 63.5
L-6 −174 G TTCCCCCTAGT TGTGTCT TGCG
IL-6 −174 C TTCCCCCTAGTTGTGTCT TGCC
IL-12 +1188 IL-12 +1188 Common GACACAACGGAATAGACC 116 1.75 55.0
IL-12 +1188 A AATGAGCATT TAGCATCT
IL-12 +1188 C AATGAGCATT TAGCATCG
IFNγ +874 INFγ +874 Common CATCTACTGTGCCTTCCTGT 116 2.75 61.0
IFNγ +874 T TTCT TACAACACAAAATCAAATCT
IFNγ +874 A TTCT TACAACACAAAATCAAATCA
TGFβ1 +915 TGFβ1 +915 Common CGAGCCGCAGCTTGGACAGGAT 116 1.8 63.5
TGFβ1 +915 G ACTGGTGCTGACGCCTGTCCG
TGFβ1 +915 C ACTGGTGCTGACGCCTGTCCC
Control 63 TGCCAAGTGGAGCACCCAA 796 As per specific SNP
primers 64 GCATCTTGCTCTGTGCAGAT test conditions

412 W. M. Howell et al.
© 2003 Blackwell Publishing Ltd, European Journal of Immunogenetics 30, 409–414
according to known clinical and histopathological
prognostic indicators. This analysis showed that the IL-
1β −511 TT genotype was significantly decreased in
frequency among vertical growth phase tumours with a
Breslow thickness greater than the clinically significant
cut-off point of 1.5 mm, occurring in only 2/51 (3.9%)
thicker vs. 14/78 (17.9%) thinner tumours [P = 0.03; RR
= 0.29 (95% CI 0.05– 0.95)], although this association
failed to retain significance when corrected for the number
of genotypic comparisons made. Numbers of tumours
with a Breslow thickness > 3.5 mm were too small to
assess correlations with cytokine genotypes. No other
cytokine genotypes showed significant associations with
any of the prognostic indicators examined, even prior to
correction of P-values for multiple comparisons.
Discussion
This is the first report of an investigation of the role of
IL-1β, IL-2, IL-4, IL-8, IL-12 and TGFβ1 SNPs in
susceptibility to and/or association with prognostic indi-
cators in CMM, while only a single, small study of IL-6
and IFNγ SNPs and CMM has been published (Martinez-
Escribano et al., 2002). Control genotype and allele
frequencies for all SNPs are in broad agreement with
previously published data from North European Caucasian
populations (e.g. Fishman et al., 1998; John et al., 1998;
Perrey et al., 1998; Mullighan et al., 1999; El-Omar et al.,
2000; Hall et al., 2000; Hull et al., 2000; Reynard et al.,
2000; Karjalainen et al., 2002; Meenagh et al., 2002).
However, none of the nine SNPs studied showed any
association with CMM susceptibility via case-control
comparisons, despite evidence that all of the respective
cytokine gene products show either tumour-associated or
systemic dysregulation of expression in CMM patients
(Lu & Kerbel, 1992; Krasagakis et al., 1993; Rodeck
et al., 1994; Krasagakis et al., 1998; Moretti et al., 1999),
or a tumour response to cytokine administration (Holl-
ingsworth et al., 1996; Garcia-Vazquez et al., 2000; Nagai
et al., 2000). Martinez-Escribano et al. (2002) also failed to
demonstrate any association between the IL-6 −174 and
IFNγ +874 SNPs and CMM susceptibility in a small series
of 42 Spanish Caucasian CMM patients and 48 controls.
Results for IL-8 −251 contrast with our recent findings
in prostate cancer, where the IL-8 −251 TT (‘low expres-
sion’) genotype was associated with protection from
prostate cancer (McCarron et al., 2002), consistent with
a pro-angiogenic role for IL-8 in this latter malignancy.
TGFβ1 also plays a key role in tumour growth and
angiogenesis (Blobe et al., 2000), although, again, no
association with the TGFβ1 +915 expression-related
SNP was observed in the present study. These findings con-
trast with significant associations between IL-10 and vas-
cular endothelial growth factor (VEGF) expression-related
Table 2. Cytokine genotype frequencies in CMM patients and healthy
controls
Genotype
CMM
n (%)
Controls
n (%)
IL-1β −35 T T 66 (39.1) 66 (40.2)
TC 78 (46.2) 76 (46.3)
CC 25 (14.8) 22 (13.4)
n = 169 n = 164
IL-1β −511 T T 19 (12.3) 39 (14.9)
TC 73 (47.4) 135 (51.7)
CC 62 (40.3) 87 (33.3)
n = 154 n = 261
IL-2 −330 T T 79 (57.7) 86 (53.8)
TG 44 (32.1) 61 (38.1)
GG 14 (10.2) 13 (8.1)
n = 137 n = 160
IL-4 −590 T T 0 (0.0) 4 (1.9)
TC 23 (15.0) 39 (18.8)
CC 130 (85.0) 165 (79.3)
n = 153 n = 208
IL-6 −174 GG 48 (29.8) 79 (35.3)
GC 79 (49.1) 101 (45.1)
CC 34 (21.1) 44 (19.6)
n = 161 n = 224
IL-8 −251 T T 39 (27.5) 76 (32.3)
TA 74 (52.1) 105 (44.7)
AA 29 (20.4) 54 (23.0)
n = 142 n = 235
IL-12 +188 AA 95 (65.5) 139 (60.7)
AC 42 (29.0) 77 (33.6)
CC 8 (5.5) 13 (5.7)
n = 145 n = 229
IFNγ +874 T T 31 (21.8) 46 (20.7)
TA 71 (50.0) 107 (48.2)
AA 40 (28.2) 69 (31.1)
n = 142 n = 222
TGFβ1 +915 GG 135 (87.1) 133 (83.1)
GC 19 (12.3) 26 (16.3)
CC 1 (0.6) 1 (0.6)
n = 155 n = 160
Table 3. Cytokine allele frequencies in CMM patients and healthy
controls
Allele CMM n (%) Controls n (%)
IL-1β −35 T 210 (62.1) 208 (63.4)
C128 (37.9) 120 (36.6)
IL-1β −511 T 111 (36.0) 213 (40.8)
C197 (64.0) 309 (59.2)
IL-2 −330 T 202 (73.7) 233 (72.8)
G72 (26.3) 87 (27.2)
IL-4 −590 T 23 (7.5) 47 (11.3)
C283 (92.5) 369 (88.7)
IL-6 −174 G 175 (54.3) 259 (57.8)
C147 (45.7) 189 (42.2)
IL-8 −251 T 152 (53.5) 257 (54.7)
A132 (46.5) 213 (45.3)
IL-12 +1188 A 232 (80.0) 355 (77.5)
C58 (20.0) 103 (22.5)
IFNγ +874 T 133 (46.8) 199 (44.8)
A151 (53.2) 245 (55.2)
TGFβ1 +915 G 289 (93.2) 292 (91.2)
C21 (6.8) 28 (8.8)
All case-control comparisons in Tables 2 and 3 non-significant at
P = 0.05.

Cytokine polymorphisms in malignant melanoma 413
© 2003 Blackwell Publishing Ltd, European Journal of Immunogenetics 30, 409–414
SNPs in the same CMM patient series, consistent with IL-
10 and VEGF acting as anti- and pro-antiogenic factors,
respectively (Howell et al., 2001, 2002a). These results do
not suggest a role for IL-10 in suppressing anti-tumour
immune responses, as genotypes associated with low IL-
10 expression were a risk factor for CMM. In addition,
we have also shown that tumour necrosis factor (TNF)α-
and lymphotoxin (LT)α-associated SNPs are unlikely to
confer a major risk for CMM susceptibility (Howell et al.,
2002b), focusing attention on IL-10 and VEGF SNPs as
anti/pro-angiogenic risk factors for CMM.
Taken together, these results suggest that it is unlikely
that any of the nine IL-1β, IL-2, IL-4, IL-6, IL-8, IL-12,
IFNγ or TGFβ1 SNPs studied plays a major role in deter-
mining susceptibility to CMM. However, a role for other
polymorphisms in the genes studied cannot be excluded.
For example, at least 16 SNPs have been identified in the
IL-12 gene (Haukim et al., 2002), although many of these
may have one rare allele and/or have no effect on gene
expression or gene function. Full genotyping and haplo-
typing for informative SNPs would need to be undertaken
in order to formally exclude a role for polymorphism in
each gene studied.
An examination of possible associations between these
SNPs and known clinical and histopathological prognostic
indicators in CMM revealed only a single association,
with the IL-1β −511 TT genotype associated with thinner
invasive tumours, at the clinically significant cut-off point
of 1.5 mm, although this association did not retain sig-
nificance when corrected for multiple comparisons. The
influence of the IL-1β −511 SNP on IL-1β expression is
uncertain, although it has been reported that peripheral
blood mononuclear cells of IL-1β −511 T genotype
produce a non-significantly elevated level of IL-1β (Santtila
et al., 1998). If this is so, the IL-1β −511 TT association
with thinner invasive tumours is consistent with reports
that IL-1 inhibits the growth of melanoma cells, except at
very low doses (Garcia de Galdeano et al., 1999).
However, possible IL-1β −511 genotype correlations with
expression may be due to linkage disequilibrium with
IL-1β −35 or other SNPs (El-Omar et al., 2000) and the
IL-1β −35 SNP was not associated with Breslow thickness
of invasive CMM in this study. Accordingly, a role for IL-
1β polymorphism in the development of CMM cannot be
firmly excluded or included by this study and merits
further investigation in a larger patient series. In addition,
an investigation of IL-1β genotype in combination with
other genotypes (e.g. IL-10 and VEGF), which we have
previously shown to be associated with tumour Breslow
thickness, is indicated, although numbers of patients in
the present study are insufficient for such an analysis
(Howell et al., 2001; Howell et al., 2002a). Because CMM
is a relatively uncommon cancer, accumulation of sufficient
cases for detailed immunogenetic analysis and to provide
a DNA resource bank for future genetic epidemiological
studies on this tumour type will require a collaborative
network involving several clinical centres and laboratories,
with a parallel collection of controls, appropriately
matched for age, sex, ethnicity and sun exposure.
In conclusion, subject to the above limitations, results
from this preliminary study suggest that polymorphisms
associated with pro- and anti-inflammatory cytokine
genes, along with TGFβ1-associated polymorphisms, are
unlikely to play a major role in susceptibility to and/or
prognosis in CMM, although analysis of additional SNPs
and informative haplotypes for candidate cytokine genes
and combinations of genes is required for a more defini-
tive analysis. Such analysis may be merited in order to
determine whether IL-1β-associated polymorphisms play
a significant role in determining tumour invasiveness in
this malignancy.
Acknowledgement
This work was supported by a grant from the Association
for International Cancer Research, St, Andrews, Scotland
(No. 99–121).
References
Bateman, A.C., Turner, S.J., Theaker, J.M. & Howell, W.M. (1998)
HLA-DQB1*0303 and *0301 alleles influence susceptibility to
and prognosis in cutaneous malignant melanoma in the British
caucasian population. Tissue Antigens, 52, 67.
Beahrs, O.H. & Myers, M.H. (1983) Manual for staging of cancer.
American Joint Committee on Cancer, 2nd edn. p. 117.
JB Lippincott, Philadelphia.
Bidwell, J., Keen, L., Gallagher, G., Kimberley, R., Huizinga, T.,
McDermott, M.F. et al. (1999) Cytokine gene polymorphism in
human disease: on-line databases. Genes and Immunity, 1, 39.
Blobe, G.C., Schiemann, W.P. & Lodish, H.F. (2000) Role of trans-
forming growth factor β in human disease. New England Journal
of Medicine, 342, 1350.
Clark, W.H. Jr, Elder, D.E., Guerry, D., Braitman, L.E., Trock, B.J.,
Schultz, D., Synnestvedt, M. & Halpern, A.C. (1989) Model
predicting survival in stage I melanoma based on tumour
progression. Journal of the National Cancer Institute, 81, 1893.
Cress, R.D. & Holly, E.A. (1997) Incidence of cutaneous melanoma
among non-Hispanic whites, Hispanics, Asians and blacks: an
analysis of California cancer registry data, 1988. Cancer Causes
and Control, 8, 246.
El-Omar, E.M., Carrington, M., Chow, W.-H., McColl, K.E.L.,
Bream, J.H., Young, H.A., Garcia de Galdeano, A., Boyano, D. &
Canavate, L. (1999) Effect induced by interleukin-1 on the
behaviour of B16F10 melanoma cells. Oncology Reports, 6, 225.
Fishman, D., Faulds, G., Jeffery, R., Mohamed-Ali, V., Yudkin, J.S.,
Humphries, S. & Woo, P. (1998) The effect of novel
polymorphisms in the interleukin-6 (IL-6) gene on IL-6
transcription and plasma IL-6 levels, and an association
with systemic-onset juvenile chronic arthritis. Journal of Clinical
Investigation, 102, 1369.
Garcia de Galdeano, A., Boyano, D. & Canavate, L. (1999) Effect
induced by interleukin-1 on the behaviour of B16F10 melanoma
cells. Oncology Reports, 6, 225.
Garcia-Vazquez, M.D., Boyano, M.D., Canavate, M.L.,
Gardeazabal, J., de Galdeano, A.G., Lopez-Michelena, T. et al.
(2000) Interleukin-2 enhances the growth of human melanoma
cells derived from primary but not from metastatic tumours.
European Cytokine Network, 11, 654.
Hajeer, A.H. (1998) Two novel bi-allelic polymorphisms in the IL-2
gene. European Journal of Immunogenetics, 25, 419.
Hall, M.A., McGlinn, E., Coakley, G., Fisher, S.A., Boki, K.,
Middleton, D. et al. (2000) Genetic polymorphism of IL-12 p40
gene in immune-mediated disease. Genes and Immunity, 1, 219.

414 W. M. Howell et al.
© 2003 Blackwell Publishing Ltd, European Journal of Immunogenetics 30, 409–414
Haukim, N., Bidwell, J.L., Smith, A.J.P., Keen, L.J., Gallagher, G.,
Kimberly, R. et al. (2002) Cytokine gene polymorphism in human
disease: on-line databases (suppl. 2). Genes and Immunity, 3,
313.
Herrera, J., Lissowska, J., Yuan, C.-C., Rothman, N., Lanyon, G.,
Martin, M., Fraumeni, J.F. Jr & Rabkin, C.S. (2000)
Interleukin-1 polymorphisms associated with increased risk of
gastric cancer. Nature, 204, 396.
Hollingsworth, S.J., Darling, D., Gaken, J., Hirst, W., Patel, P.,
Kuiper, M., Towner, P., Humphreys, S., Farzaneh, F. &
Mufti, G.J. (1996) The effect of combined expression of
interleukin 2 and interleukin 4 on the tumour genicity of B16F10
melanoma. British Journal of Cancer, 74, 6.
Howell, W.M., Turner, S.J., Bateman, A.C. & Theaker, J.M. (2001)
IL-10 promoter polymorphisms influence tumour development in
cutaneous malignant melanoma. Genes and Immunity, 2, 25.
Howell, W.M., Bateman, A.C., Turner, S.J., Collins, A. &
Theaker, J.M. (2002a) Influence of vascular endothelial growth
factor single nucleotide polymorphisms on tumour development
in cutaneous malignant melanoma. Genes and Immunity, 3, 229.
Howell, W.M., Bateman, A.C., Turner, S.J. & Theaker, J.M. (2002b)
Influence of TNFα and LTα single nucleotide polymorphisms on
susceptibility to and prognosis in cutaneous malignant melanoma
in the British population. European Journal of Immunogenetics,
29, 17.
Hull, J., Thomson, A. & Kwiatowski, D. (2000) Association of
respiratory syncytial virus bronchioloitis with the interleukin
8 gene region in UK families. Thorax, 55, 1023.
John, S., Turner, D., Donn, R., Sinnott, P., Worthington, J.,
Ollier, W.E., Hutchinson, I.V. & Hajeer, A.H. (1998) Two novel
bi-allelic polymorphisms in the IL-2 gene. European Journal of
Immunogenetics, 25, 419.
Karjalainen, J., Nieminen, M.N., Aromaa, A., Klaukka, T. &
Hurme, M. (2002) The IL-1β genotype carries asthma
susceptibility only in men. Journal of Allergy and Clinical
Immunology, 109, 514.
Kornman, K.S., Crane, A., Wang, H.Y., di Giovine, F.S.,
Newman, M.G., Pirk, F.W., Wilson, T.G. Jr, Higginbottom, F.L. &
Duff, G.W. (1997) The interleukin-1 genotype as a severity factor
in adult periodontal disease. Journal of Clinical Periodontology,
24, 72.
Krasagakis, K., Garbe, C. & Orfanos, C.E. (1993) Cytokines in
human melanoma cells: synthesis, autocrine stimulation and
regulatory functions. Melanoma Research, 3, 425.
Krasagakis, K., Tholke, D., Farthmann, B., Eberle, J., Mansmann, U.
& Orfanos, C.E. (1998) Elevated plasma levels of transforming
growth factor (TGF)-β1 and TGF-β2 in patients with
disseminated malignant melanoma. British Journal of Cancer, 77,
1492.
Lee, J.E., Lu, M., Mansfield, J.F., Platsoucas, C.D., Reveille, J.D. &
Ross, M.I. (1996) Malignant melanoma: relationship of the
human leukocyte antigen class II gene DQB1*0301 to disease
recurrence in American Joint Committee on Cancer stage I or II.
Cancer, 78, 758.
Lu, C. & Kerbel, R.S. (1992) Interleukin 6 undergoes transition from
paracrine growth inhibitor to autocrine stimulator during human
melanoma progression. Journal of Cell Biology, 120, 1281.
Lulli, P., Grammatico, P., Brioli, G., Catricala, C., Morellini, M.,
Roccella, M. et al. (1998) HLA-DR and -DQ alleles in Italian
patients with melanoma. Tissue Antigens, 51, 276.
Mackie, R., Hunter, J.A., Aitchison, T.C., Hole, D., Mclaren, K.,
Rankin, R., Blessing, K., Evans, A.T., Hutcheson, A.W. & Jones,
D.H. (1992) Cutaneous malignant melanoma, Scotland, 1979.
The Scottish Melanoma Group. Lancet, 339, 971.
Martinez-Escribano, J.A., Moya-Quiles, M.R., Muro, M.,
Montes-Ares, O., Hernandez-Caselles, T., Frias, J.F. &
Alvarez-Lopez, M.R. (2002) Interleukin-10, interleukin-6 and
interferon-γ gene polymorphisms in melanoma patients.
Melanoma Research, 12, 465.
McCarron, S.L., Edwards, S., Evans, P.R., Gibbs, R.,
Dearnaley, D.P., Dowe, A. et al. (2002) Influence of cytokine gene
polymorphisms on the development of prostate cancer. Cancer
Research, 62, 3369.
Meenagh, A., Williams, F., Ross, O.A., Patterson, C., Gorodezky, C.,
Hammond, M., Leheny, W.A. & Middleton, D. (2002) Frequency
of cytokine polymorphisms in populations from Western Europe,
Africa, Asia, the Middle East and South America. Human
Immunology, 63, 1055.
Moretti, S., Pinzi, C., Spallanzani, A., Berti, E., Chiarugi, A.,
Mazzoli, S., Fabiani, M., Vallecchi, C. & Herlyn, M. (1999)
Immunohistochemical evidence of cytokine networks networks
during progression of human melanocytic lesions. International
Journal of Cancer, 84, 160.
Mullighan, C.G., Marshall, S.E., Bunce, M. & Welsh, K.I. (1999)
Variation in immunoregulatory genes determines the clinical
phenotype of common variable immunodeficiency. Genes and
Immunity, 1, 137.
Nagai, H., Hara, I., Horikawa, T., Fujii, M., Kurimoto, M.,
Kamidono, S. & Ichihashi, M. (2000) Antitumor effects on mouse
melanoma elicited by local secretion of interleukin-12 and their
enhancement by treatment with interleukin-18. Cancer
Investigation, 18, 206.
Perrey, C., Pravica, V., Sinnott, P.J. & Hutchinson, I.V. (1998)
genotyping for polymorphisms in interferon-γ, interleukin-10,
transforming growth factor-β1 and tumour necrosis factor-α
genes: a technical report. Transplant Immunology, 6, 193.
Pravica, V., Perrey, C., Stevens, A., Lee, J.-H. & Hutchinson, I.V.
(2000) A single nucleotide polymorphism in the first intron of the
human IFN-γ gene: absolute correlation with a polymorphic CA
microsatellite marker of high IFN-γ production. Human
Immunology, 61, 863.
Reynard. M.P., Turner, D. & Navarrete, C.V. (2000) Allele
frequencies of polymorphisms of the tumour necrosis factor-α,
interleukin-10, interferon-γ and interleukin-2 genes in a North
European Caucasoid group from the UK. European Journal of
Immunogenetics, 27, 241.
Rodeck, U., Bossler, A., Graeven, U., Fox, F.E., Nowell, P.C.,
Knabbe, C. & Kari, C. (1994) Transforming growth factor-β
production and responsiveness in normal human melanocytes and
melanoma cells. Cancer Research, 54, 575.
Rosenwasser, L.J., Klemm, D.J., Dresback, J.K., Inamura, H.,
Mascali, J.J., Klinnert, M. & Borish, L. (1995) Promoter
polymorphism in the chromosome 5 gene cluster in asthma and
atopy. Clinical and Experimental Allergy, 25, 74.
Santtila, S., Savinainen, K. & Hurme, M. (1998) Presence of the IL-
1RA allele 2 (IL1RN*2) is associated with enhanced IL-1β
production in vitro. Scandinavian Journal of Immunology, 47,
195.
Svejgaard, A. & Ryder, L.P. (1994) HLA and disease associations:
detecting the strongest association. Tissue Antigens, 43, 18.
Van Belle, P., Rodeck, U., Nuamah, I., Halpern, A.C. & Elder, D.E.
(1996) Melanoma-associated expression of transforming growth
factor-β isoforms. American Journal of Pathology, 148, 1887.
Wolfel, T., Hauer, M., Klehmann, E., Brichard, V., Ackermann, B.,
Knuth, A., Boon, T. & Zumbuschenfelde, K.H.M. (1993)
Analysis of antigens recognised on human melanoma cells
by A2-restricted cytolytic T lymphocytes (CTL). International
Journal of Cancer, 55, 237.