ARTICLE
The incidence of cutaneous melanoma is increasing mainly due to "thin",
i.e. ≤ 1 mm, forms [1-5] as a probable consequence of the
extension of early diagnosis activities. However, in Australia as in other
parts of the world the incidence of thick melanoma is increasing slightly
or — at least — stable [6]. Since tumour
thickness is the most relevant prognostic factor for skin melanoma [7,
8], the knowledge of factors associated with this subset of lesions is
of straightforward importance in refining criteria for melanoma prevention.
According to studies carried out in predominantly fair-skinned populations,
at high risk for melanoma, thick lesions are diagnosed more often among
old patients [9-15], especially males [9, 10, 13-15], in nodular histotype
[9-12, 14, 15] and in the head and neck [9, 10, 12, 13, 16, 17].
The aim of the present paper was to investigate, in a Mediterranean
area, clinical and demographic factors associated with the diagnosis of
a thick melanoma in a registry-based study and to evaluate the changes
which occurred over time.
Methods
Data were retrieved from the archives of the Tuscany Cancer Registry — RTT,
a population-based cancer registry active in the provinces of Florence
and Prato (about 1,180,000 residents in the 1991 census), central
Italy, since 1985. The description of the criteria for collection, registration
and analysis followed by the Registry has been presented elsewhere [18].
Briefly, the registry collects from the Regional Health Authorities the
database of hospital discharge notes, relative to residents in the Region,
from all the public Italian hospitals. When necessary, general practitioners
are also involved. Pathological reports are collected from all the pathological
departments active in the registry area and from the main ones of the
region. Death certificates of resident subjects are retrieved from the
Regional Mortality Registry.
Breslow thickness was categorised into three classes, < = 1 mm,
1-3.00, > 3 mm. The 3 mm cut-off was chosen in
agreement with previous studies [11, 12, 15].
The association of thickness categories was evaluated according to the
following variables:
Gender: males and females
Age: 0-30, 31-50, 51-70 and > 70 years.
Morphological type: nodular (NM), superficial
spreading (SSM) and nos (not otherwise specified) & other types. The
latter category was grouped because of low numbers and included in 1985-87,
23 not otherwise specified (nos) melanoma, 9 lentigo maligna,
3 epithelioid and spindle cell melanoma, 1 spindle cell, 1 malignant
melanoma in junctional nevus, and in 1995-97, 51 nos melanoma,
8 lentigo maligna, 2 epithelioid cell melanoma, 2 spindle
cell, 1 malignant melanoma in junctional nevus.
Site: Head and neck, trunk, upper limb, lower limb, and other and unspecified.
Residence. Florence and seven surrounding municipalities (about 610,000 inhabitants)
and the other 43 municipalities (about 560,000 residents). In
the city of Florence the Dermatology University Department, has been active
in early diagnosis and in skin health promotion activities since the 1980s.
Most of the municipalities not surrounding Florence are rural areas quite
far from dermatology services promoting melanoma early diagnosis campaigns.
To evaluate the changes which occurred over time in the relationship
between melanoma thickness and the above mentioned variables, the distribution
of thickness according to such variables was evaluated in two periods:
1985-1987 and 1995-1997.
The distribution of Breslow categories for each variable was evaluated
by means of Chi 2 test. When there were less than 5 expected
cases in one cell, the Fisher's exact test was performed.
For the most recent period, 1995-1997, the effect of the above mentioned
variables was also evaluated in a logistic models comparing thick tumours
(> 3 mm) 59 cases to thin ones (< = 1 mm),
227 cases.
Results
In the RTT area 260 cutaneous malignant melanomas were diagnosed
in 1985-1987 [121 (46.5 %) among males and 139 (53.5 %) among
females] and 490 in 1995-1997 [217 (44.3 %) among males and
273 (55.7 %) among females]. No cases were diagnosed among children
(0-14 years) in the two periods. The mean incidence rate (standardised
on the European population) was 6.4 per 100,000/year in both sexes
in 1985-1987 and 10.7 among males and 11.7 among females during
1995-1997. The percentage of histologically verified cases increased from
90.8 % in 1985-1987 to 95.5 % in 1995-1997.
The present analysis was restricted to those cases for which the information
on Breslow thickness was available, 182/260 (70.0 %) in 1985-87 and
387/490 (79 %) in 1995-1997, respectively.
In Table I the distribution
of the clinical and demographic variables according to the Breslow levels
is shown for the two periods.
In 1985-1987 thick melanomas (> 3 mm) were diagnosed predominantly
among males (males to females ratio
M/F = 1.7) particularly in older subjects (especially > 70 years),
in nodular melanoma and among residents in municipalities not surrounding
the city of Florence. There were no significant differences according
to skin site.
From 1985-87 to 1995-97 there had been a general shift towards
thinner tumours for all the analysed variables. The overall percentage
of thin tumours (< = 1 mm) was 29.1 % in
1985-87 and 58.7 % in 1995-97; the proportion of intermediate
thick level (1-3 mm) reduced from 42.9 % in 1985-87 to
26.1 % in 1995-97 and the thick ones (> 3 mm) decreased
from 28.0 % to 15.2 %.
In 1995-1997, thick melanomas were still lower among females (12.9 %)
than among males (18.5 %) but not significantly. The highest prevalence
of thick melanomas was among older subjects (33.7 % of melanoma diagnosed
in subjects older than 70 years) and especially for nodular type
(21/38, 55.3 %). No differences were evidenced according to skin
site and place of residence.
In Table II the results
of a logistic multivariate analysis performed on 286 cases diagnosed
in 1995-1997 and with Breslow thickness < = 1 or > 3 mm
are shown. The risk of having a diagnosis of a thick melanoma, > 3 mm
(n.59), was evaluated in comparison of the diagnosis of a thin melanoma, < = 1 mm
(n.227). Melanomas thicker than 3 mm, after adjustment for other
variables, were highly likely to be nodular (OR = 112.3, 95 %
C.I. 25.7-490.1) or of other type (excluded SSM) (OR = 8.8;
95 % C.I. = 3.6-21.5) than SSM, and to develop in subjects
older than 70 years (OR = 16.4; 95 % C.I. 4.0-67.3)
than in those aged 51-70 years. There were no significant differences
according to gender, residences and sites.
Discussion
The incidence rates of skin melanoma in the RTT area have doubled over
the analysed period from 6.1 per 100,000 (European standard) in 1985
to 12.3 in 1997. Moreover, we observed a striking improvement in
thickness distribution. In fact, the proportion of thin melanoma (< = 1 mm)
doubled from 29.7 % in 1985-87 to 58.7 % in 1995-97 and
that for thick (> 3 mm) ones almost halved from 28.0 %
to 15.2 %. However, a generally accepted worrisome point is that
the incidence rates of thick forms are stable or even slightly increasing.
Therefore, the knowledge of variables associated with late melanoma diagnosis
could be of straightforward importance to better gear melanoma prevention
strategies [6]. Also in RTT area, incidence rates of thick lesions showed
no reduction over time (data not shown).
The present series showed that in central Italy during 1995-1997 the
most relevant determinants of thick melanoma were nodular type and older
age at diagnosis, as evidenced in Australia [11] and in Sweden [14].
In 1995-1997 case-series we found that 50.9 % of melanomas thicker
than 3 mm were in patients older than 70 years and that 35.6 %
of thick tumours belonged to nodular type. Such figures agreed with previously
published data from several countries with higher melanoma incidence rates
[9-12, 14, 15]. Nodular type was the strongest determinant of the risk
of thick melanoma diagnosis, being associated to a very high OR value
(OR = 112).
Interestingly, during the period analysed, from 1985 to 1997, there
were some changes among the factors involved in the presentation of thick
melanoma.
Males had a higher risk of having a diagnosis of thick melanoma than
females during 1985-87; but this effect disappeared in 1995-97. Results
on gender as a risk factor for thick melanoma vary among studies, and
the lack of differences between sexes has been also evidenced among more
than 700 cases in the US [9] and among more than 900 Australian
cases when tumour type was taken into account [11]. Presumably, the well
known lower awareness of males on their health and on skin signs, in comparison
with females [19], was also present in central Italy during the mid-1980s
leading to a worse melanoma thickness distribution for males. It is probable
that the improvement in education about melanoma recorded in the last
decade, and indirectly supported by the finding of an increasing percentage
of "thin" forms, may have bridged the gap between genders, strongly
reducing the effect associated with male sex as predictor of late diagnosis.
Our data did not support the finding of an higher risk for the head/neck
site and, in agreement with that evidenced also by Chamberain [11] and
Hanrahan [15], tumour site was not related to thickness. Therefore, two
widely accepted predictors of late diagnosis according to previous studies,
i.e. male gender and anatomic site, were no longer found to be associated
with thick melanoma when allowance for other variables was made and once
focussing on more recent data. According to our study, multivariate analysis
of recent population-based incidence data showed
that only histotype (NM and other non SSM forms) and age older than 70 were
significant predictors of thick lesions.
Interestingly, no clear-cut effect was indeed found associated with
residence place, defined as municipality of Florence and surrounding areas
versus other peripheral municipalities. In Florence, the main municipality
of the RTT area, a University Dermatology Department is active. This institution
has been carrying out both early diagnosis activities and educational
campaigns in the population [20]. Presumably, during the late-1980s, the
catchment population for such activities was mainly resident in the municipality
of Florence or in the surrounding ones, this may contribute to explain
the higher risk for thicker melanoma during 1985-87 in subjects living
far from this institution. The extension of such activities, the greater
awareness of general practitioners and the increased expertise of dermatologists
have contributed to a global improvement of diagnosis all over the area.
In fact, the shift towards thinner melanomas was evidenced in both sexes,
in all the age-groups, for all types, in all sites and also in all municipalities
of residence.
A strong point of this study is that the present series derived from
a population-based registry, therefore it described a mean value for each
subject in the resident population, independently from the place of diagnosis.
However, since the RTT collects pathology reports from all the pathology
services in the registry area and from the main ones of the region, it
is possible that the expertise among general pathologists varies among
services and in comparison with dermopathologists (weakness point). This
may be the reason for the relevant percentage of melanomas without type
specification (51/387, 13.2 %) and/or without information on Breslow
thickness (103/490, 21.0 % in 1995-97).
The results of the present study showed that also in central Italy the
main target population for improved future preventive activity should
be older subjects, but not predominantly males. However, some limitations
in the improving melanoma screening in old subjects should be taken into
account. As patients' awareness of melanoma signs and symptoms is associated
with reduced Breslow depth, older age is associated with a decreased likelihood
of performing skin self-examination [19]. Regarding future efforts in
improving detection of nodular histotype, it is accepted that NMs are
associated with sensation and bleeding more often than SSMs [15]. Unfortunately,
these are late signs of suspicion, associated with already well developed
disease. As previously shown, the most widely used rule for early diagnosis
of melanoma, i.e. the ABCD rule, has been developed with the purpose of
improving the early diagnosis of superficial spreading melanomas, while
it scarcely applies to nodular type [21]. Therefore, the future challenge
to fighting thick melanoma is how to better involve older subjects in
prevention activities. The investigation and the refinement of symptoms
and signs associated with early diagnosis of nodular melanoma, if existing
at all, may additionally represent a key point for the improvement of
the educational message about melanoma prevention for the future.
Article accepted on 17/10/2002REFERENCES
1
Hall HI, Miller DR, Rogers JD, Bewerse B. Update on the incidence
and mortality from melanoma in the United States. J Am Acad Dermatol
1999; 40: 35-42.
2
Stang A, Stang K, Stegmaier C, Hakulinen T, Jockel KH. Skin melanoma
in Saarland: incidence, survival and mortality 1970-1996. Eur J Cancer
Prev 2001; 10: 407-15.
3
Holme SA, Malinovsky K, Roberts DL. Malignant melanoma in South Wales:
changing trends in presentation (1986-98). Clin Exp Dermatol 2001;
26: 484-9.
4
Van der Rhee HJ, van der Spek-Keijser LM, van Westering R, Coebergh
JW. Increase in and stabilization of incidence and mortality of primary
cutaneous malignant melanoma in Western Netherlands, 1980-95. Br J
Dermatol 1999; 140: 463-7.
5
Garbe C, McLeod GR, Buettner PG. Time trends of cutaneous melanoma
in Queensland, Australia and Central Europe. Cancer 2000; 89: 1269-78.
6
Demierre MF. Thin melanomas and regression, thick melanomas and older
men: prognostic implications and perspectives on secondary prevention.
Arch Dermatol 2002; 138: 678-82.
7
Balch CM, Soong CJ, Gershenwald JE, Thompson JF, Reintgen DS, Cascinelli
N, et al. Prognostic factors analysis of 17,600 melanoma patients:
validation of the American Joint Committee on Cancer melanoma staging
system. J Clin Oncol 2001; 19: 3622-34.
8
Garbe C, Ellwanger U, Tronnier M, Brocker EB, Orfans CE. The new American
Joint Committee on cancer staging system for cutaneous melanoma. A critical
analysis based on data of the German central malignant melanoma registry.
Cancer 2002; 94: 2305-7.
9
Kopf AW, Welkovich B, Frankel RE, Stoppelmann EJ, Bart RS, Rogers
GS, et al. Thickness of malignant melanoma: global analysis of
related factors J Dermatol Surg Oncol 1987; 3: 345-90.
10
Roder DM, Luke CG, McCaul KA, Esterman AJ. Trends in prognostic factors
of melanoma in South Australia., 1981-1992: implications for health promotion.
Med J Aust 1995; 162: 25-9.
11
Chamberlain AJ, Fritschi L, Giles GG, Dowling JP, Kelly JW. Nodular
type and older age as the most significant associations of thick melanoma
in Victoria, Australia. Arch Dermatol 2002; 138: 609-14.
12
Hersey P, Sillar RW, Howe CG, Burton RC, Darbar SV, Foster HM, et
al. Factors related to the presentation of patients with thick primary
melanoma. Med J Aust 1991; 154: 583-7.
13
Osborne JE, Hutchinson PE. Clinical correlates of Breslow thickness
of malignant melanoma. Br J Dermatol 2001; 144: 476-83.
14
Bergenmar M, Ringborg U, Mansson Brahme E, Brandberg Y. Nodular histogenetic
type — the most significant factor for thick melanoma:
implications for prevention. Melanoma Res 1998; 8: 403-11.
15
Hanrahan PF, Hersey P, D'Este CA. Factors involved in presentation
of older people with thick melanoma. Med J Aust 1998; 169: 410-4.
16
Cohen HJ, Cox E, Manton K, Woodbury M. Malignant melanoma in the elderly.
J Clin Oncol 1987; 5: 100-6.
17
McHenry PM, Hole DJ, MacKie RM. Melanoma in people aged 65 and
over in Scotland, 1979-1989. BMJ 1992; 304: 746-9.
18
Buiatti E, Balzi D, Barchielli A, Carli S, Crocetti E, Giovannetti
L. Registro Tumori Toscano. In: Cancer Incidence in Five Continents.
Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J, (editors). Cancer
Incidence in Five Continents vol. VII. Lyon: IARC Scientific Publications,
1997, p 526-9.
19
Oliveria SA, Christos PJ, Halpern AC, Fine JA, Barnhill RL, Berwick
M. Evaluation of factors associated with skin self-examination. Cancer
Epidemiol Biomarkers Prev 1999; 8: 971-8.
20
Carli P, De Giorgi V, Nardini P, Mannone F, Palli D, Giannotti B.
Melanoma detection rate and concordance between self-skin examination
and clinical evaluation in patients attending a pigmented lesion clinic
in Italy. Br J Dermatol 2002; 146: 261-6.
21
Carli P, Borgognoni L, Reali UM, Giannotti B. Clinico-pathological
features of small diameter malignant melanoma. Eur J Dermatol 1994;
4: 440-2.
|
Printable version