SciELO - Scientific Electronic Library Online

 
vol.88 issue3Neurofibromatosis: chronological history and current issuesEvaluation of D-dimer serum levels among patients with chronic urticaria, psoriasis and urticarial vasculitis author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Anais Brasileiros de Dermatologia

Print version ISSN 0365-0596

An. Bras. Dermatol. vol.88 no.3 Rio de Janeiro May/June 2013

http://dx.doi.org/10.1590/abd1806-4841.20131855 

Investigation

Epidemiological aspects of melanoma at a university hospital dermatology center over a period of 20 years*

Aspectos epidemiológicos do melanoma em serviço de dermatologia de hospital universitário em um período de 20 anos

Flavia Vieira Brandão1 

Ana Francisca Junqueira Ribeiro Pereira2 

Bernardo Gontijo3 

Flávia Vasques Bittencourt3 

1M.D. Dermatologist, Hospital Universitário de Brasilia, Brasilia University (UnB) - Brasilia (DF), Brazil.

2M.D. Dermatologist. Hospital das Clínicas, Federal University of Minas Gerais (UFMG) - Belo Horizonte (MG), Brazil.

3M.D. Ph.D. Associate Professor of Dermatology, Federal University of Minas Gerais (UFMG) School of Medicine - Belo Horizonte (MG), Brazil.

ABSTRACT

BACKGROUND:

The incidence of melanoma has been steadily rising in past decades. Although it accounts for only 3% of all skin cancers, it is responsible for 75% of deaths.

OBJECTIVE:

to describe the epidemiological aspects of melanoma in a university hospital setting over a period of 20 years.

METHODS:

A total of 166 patients were analyzed between January 1990 and January 2010 for clinical and histological variables and correlations between them. A 5% level of significance was adopted.

RESULTS:

The majority of patients were Caucasians (74%), females (61%), with a mean age at diagnosis of 55. The predominant histological type was lentigo maligna/lentigo maligna melanoma (35.7%) and the head and neck was the most affected site (30.7%). Among non-Caucasians, the acral region was the most affected. Most tumors were in situ (41.1%). Growth of the lesion was the most frequent complaint (58.1%) and bleeding was most frequently associated with melanomas with a depth > 4mm. There were seven deaths (4.2%), with a high risk among men, non-Caucasians and those under 20 years of age, with a Breslow's depth > 2mm, with lentiginous acral melanoma and with a history of growth and bleeding.

CONCLUSIONS:

Our sample differs from most of the studies in the predominant location (head and neck), histological type (lentigo maligna/ lentigo maligna melanoma) and a major risk of death under the age of 20, which could be with a reflex of regional variation. Broader studies are necessary for validation of the results.

Key words: Epidemiology; Lentigo; Melanoma

RESUMO

FUNDAMENTOS:

A incidência do melanoma cutâneo aumentou nas últimas décadas. Embora represente 3% dos tumores cutâneos, é responsável por 75% dos óbitos. O diagnóstico precoce constitui a principal chance de cura.

OBJETIVO:

Descrever os aspectos epidemiológicos do melanoma em hospital universitário em 20 anos.

MÉTODOS:

Avaliaram-se 166 pacientes no período de janeiro de 1990 a janeiro de 2010, quanto s variáveis epidemiológicas, histológicas e óbitos relacionados ao melanoma e suas correlações. Adotou-se nível de significância de 5%.

RESULTADOS:

A maioria dos pacientes era brancos (74%), mulheres (61%), com média de idade ao diagnóstico de 55 anos. O tipo histológico predominante foi o lentigo maligno/lentigo maligno melanoma (35,7%) e a localização mais frequente foi a cabeça e o pescoço (30,7%). Entre os não-brancos, a região acral foi a mais acometida. Quanto espessura tumoral, a maioria dos melanomas era in situ (41,1%). O crescimento da lesão foi a queixa mais frequente (58%) e o sangramento foi mais associado a melanomas espessos. Ocorreram sete óbitos (4,2%), com maior risco de morte em menores de 20 anos e naqueles com história de sangramento, após análise multivariada.

CONCLUSÃO:

Esta casuística difere da maioria dos estudos em relação localização (cabeça e pescoço), ao tipo histológico (lentigo maligno/lentigo maligno melanoma) e ao maior risco de óbito em menores de 20 anos, o que pode ser devido variação regional. Estudos mais amplos são necessários para validação destes resultados.

Palavras-Chave: Epidemiologia; Lentigo; Melanoma

INTRODUCTION

The incidence of cutaneous melanoma (CM) has increased continuously in recent decades around the world.1 The highest rates, 40 to 60 cases per 100,000 inhabitants, are found in Australia and New Zealand.2 , 3 In the USA, the incidence of CM rose from 8.7 to 26.5 per 100,000 inhabitants from 1975 to 2007, an increase of 300%.4 In Brazil, the estimated incidence of CM for 2010, according to the National Cancer Institute (INCA) varies from 3.04 to 3.72 per 100,000 men and 2.92 to 3.04 per 100,000 women, with the highest rates in the Southern region of the country.5

In recent years, survival rates have improved, probably due to earlier diagnosis.3 , 5 Two thirds of all cases of CM diagnosed in the USA between 1988 and 1999 had a tumor thickness of < 1 mm, while the proportion of CM ≥ 2 mm remained stable.6

Various environmental and constitutional risk factors are found to be associated with CM. Among the former, intermittent exposure to sunlight, especially that which results in burns accompanied by blisters during childhood, has been referred to as the most relevant.7 , 8 As for constitutional risk factors, of note are lighter phototypes (types I and II on the Fitzpatrick scale), congenital nevus, multiple acquired common or atypical melanocytic nevi, a personal and/or family history of CM, xeroderma pigmentosum, a personal history of other skin cancers and immunosuppression.9 , 10

The more refined knowledge of the epidemiological and histological characteristics of CM may be reflected in the approach taken towards it, especially favoring early diagnosis.

MATERIAL AND METHODS

One hundred and sixty-six patients from the dermatology center at a university hospital were diagnosed with CM from January of 1990 to January of 2010. The sample was defined for convenience.

The following clinical variables were analyzed: sex, age at diagnosis (< 20, 21-40, 41- 60 and > 60), skin color (white and non-white), body location (head and neck, trunk, upper limbs, lower limbs, acral regions and others), atypical nevi (none, < 10 and > 10), nonmelanoma skin cancer, non-cutaneous cancer, family history of CM, symptoms and/or signs (bleeding, itching, pain and burning sensation, changes in color or size), metastasis and deaths.

The histological variables analyzed were: histological type (superficial spreading melanoma [SSM], nodular melanoma [NM], acral-lentiginous melanoma [ALM], lentigo maligna / lentigo maligna melanoma [LM/LMM] and others), tumor thickness (in situ ≤1 mm, 1.01-2 mm, 2.01-4 mm and >4 mm) and the level of invasion or Clark level (in situ or I, II, III, IV and V).

A significance level of 5% was considered. The variables were compared based on the Chi-squared test and Fisher Exact test.

The progression towards death was evaluated based on survival analysis techniques. Only deaths resulting from CM were considered. A univariate analysis was performed utilizing the log-rank test and the multivariate analysis was conducted using the Cox model.

This study was approved by the Research Ethics Committee, and terms of free and informed consent were duly collected.

RESULTS

In the sample assessed, 101 patients (61%) were women and 65 (39%) were men. The mean age at diagnosis was 55, with standard deviation of 18,2 years. The minimum age was one year and the maximum 92 years (Graph 1). The sample consisted of 120 whites (74%) and 42 non-whites (26%).

GRAPH 1 Histogram of age at diagnosis 

Regarding body location, 30.7% of the patients presented CM on the head and neck, 21.1% on the trunk, 19.3% in acral regions, 15.1% in the upper limbs and 9.6% on the lower limbs. In a comparison between genders, the men presented a higher proportion of MM on the head and neck (35.3%), followed by the trunk (29.2%). Among the female patients, the extremities (upper and lower limbs) were the most frequent site (34.7%), followed by the head and neck (27.7%). The women presented more affected areas on the lower limbs (13.9%) than the men (3.1%), while among men CM affected the trunk more (29.2%) than among women (15.8%). The white patients had a greater proportion of CM on the head and neck (35.9%), followed by the trunk (25.8%). Among non-whites, acral regions corresponded to 50% of the cases (Table 1).

Table 1 Comparison between location and sex, skin color, age at diagnosis, histological type and signs and symptoms 

Location
Characteristics Head and neck Acral Trunk Upper limb Lower limb Others p-value
N % N % N % N % N % N %

Sex
Male 23 35.3 13 20.0 19 29.2 4 6.2 2 3.1 4 6.2 0.0061
Female 28 27.7 19 18.8 16 15.8 21 20.8 14 13.9 3 3.0
Skin color
White 43 35.9 10 8.3 31 25.8 21 17.5 12 10.0 3 2.5 <0.0011
Non-white 7 16.7 21 50.0 4 9.5 4 9.5 4 9.5 2 4.8
Age at diagnosis
1)<20 1 14.2 1 14.3 2 28.6 2 28.6 0 0.0 1 14.3 ...
21-40 8 28.5 5 17.9 7 25.0 4 14.3 1 3.6 3 10.7
41-60 19 30.1 10 15.9 15 23.8 9 14.3 8 12.7 2 3.2
>60 23 33.8 16 23.5 11 16.2 10 14.7 7 10.3 1 1.5
2) ≤40 9 25.7 6 17.1 9 25.7 6 17.1 1 2.9 4 11.5 0.1511
>40 42 32.1 26 19.8 26 19.8 19 14.5 15 11.5 3 2.3
3) ≤60 28 28.6 16 16.3 24 24.5 15 15.3 9 9.2 6 6.1 0.4531
>60 23 33.8 16 23.5 11 16.2 10 14.7 7 10.3 1 1.5
Histological type
1) LM/LMM 29 53.7 3 5.6 8 14.8 10 18.5 4 7.4 0 0.0 ...
SSM 14 26.4 0 0.0 18 34.0 12 22.6 8 15.1 1 1.9
NM 3 33.4 0 0.0 2 22.2 2 22.0 2 22.2 0 0.0
ALM 0 0.0 23 88.5 2 7.7 0 0.0 1 3.8 0 0.0
Others 1 11.1 2 22.2 1 11.1 0 0.0 0 0.0 5 56.6
2) LM/LMM 29 53.7 3 5.6 8 14.8 10 18.5 4 7.4 0 0.0 <0.0011
Others 18 18.6 25 25.8 23 23.7 14 14.4 11 11.3 6 6.2
Signs and symptoms
Yes 31 32.3 21 21.8 16 16.7 14 14.6 11 11.5 3 3.1 0.1441
No 13 32.5 3 7.5 14 35.0 6 15.0 3 7.5 1 2.5
Bleeding
Yes 1 14.3 3 42.8 1 14.3 1 14.3 0 0.0 1 14.3 0.1641
No 43 33.3 21 16.3 29 22.5 19 14.7 14 10.9 3 2.3
Alterations in sensitivity
Yes 5 25.0 4 20.0 5 25.0 3 15.0 2 10.0 1 5.0 0.9081
No 39 33.6 20 17.2 25 21.6 17 14.7 12 10.3 3 2.6
Darkening
Yes 10 33.3 4 13.3 5 16.7 7 23.4 4 13.3 0 0.0 0.5691
No 34 32.0 20 18.9 25 23.6 13 12.3 10 9.4 4 3.8
Growth
Yes 29 36.7 19 24.0 10 12.7 10 12.7 9 11.4 2 2.5 0.0131
No 15 26.3 5 8.8 20 35.1 10 17.5 5 8.8 2 3.5

1: Fisher Exact test, .... No test

Among histological types, LM/LMM was found in 35.7% of the cases, followed by SSM in 35.1%, ALM in 17.2% and NM in 6%. In a comparison respective to body locations, LM/LMM was the most frequent histological type on the head and neck. ALM was present in 88.5% of the CM cases located in acral regions, while SSM was most common on the trunk (34%).

Correlated with skin color, the white patients presented a predominance of SSM (43.5%), followed by LM/LMM (35.2%). Among the non-whites, 40% presented with ALM, followed by LM/LMM (37.5%) (Table 2 and Graph 2).

Table 2 Comparison between histological type and skin color and age at diagnosis 

Histological type
Characteristics LM /LMM SSM NM ALM Others p-value
N % N % N % N % N %

Skin color
White 38 35.2 47 43.5 9 8.3 10 9.3 4 3.7 <0.0011
Non-whites 15 37.5 6 15.0 0 0.0 16 40.0 3 7.5
Age at diagnosis
1) <20 2 33.3 2 33.3 0 0.0 1 16.7 1 16.7 ...
21- 40 5 19.2 11 42.3 2 7.7 5 19.2 3 11.5
41 -60 19 33.9 20 35.7 5 8.9 8 14.3 4 7.1
>60 28 44.4 20 31.8 2 3.2 12 19.1 1 1.6
2) ≤40 7 21.9 13 40.6 2 6.3 6 18.8 4 12.5 0.2031
>40 47 39.5 40 33.6 7 5.9 20 16.8 5 4.2
3) ≤60 26 29.5 33 37.5 7 8.0 14 15.9 8 9.1 0.1071
>60 28 44.4 20 31.7 2 3.2 12 19.0 1 1.6

1: Fisher Exact test, .... No test

GRAPH 2 Distribution of histological type by skin color 

Roughly 20% of the patients had atypical nevi, with 16.5% presenting less than 10 and 3.4% more than 10 such lesions.

A family history of CM was present in 16 (10%) of the patients.

Regarding tumor thickness, 41.1% were in situ, 31.1% presented a tumor thickness < 1 mm, 11.3% were between 1.01 and 2 mm and 9.9% were betweenn 2.01 and 4mm, 6.6% were > 4 mm. In a comparison of histological types, LM/LMM was the most frequent among in situ CM or CM with a tumor thickness ≤ 1 mm (92.3%), followed by SSM (73.1%). There was no statistically significant difference in the correlations between tumor thickness and gender, skin color, age and body location (Table 3).

Table 3 Comparison between tumor thickness and sex, skin color, age at diagnosis, histological type and signs and symptoms 

Tumor thickness (mm)
Characteristics In situ <1.0 1.01-2,0 2.01- 4.0 >4.0 p-value
N % N % N % N % N %

Sex
Male 20 35.7 18 32.1 10 17.9 5 8.9 3 5.4 0.3751
Female 42 44.2 29 30.5 7 7.4 10 10.5 7 7.4
Skin color
White 40 36.1 39 35.1 14 12.6 10 9.0 8 7.2 0.2931
Non-white 21 55.3 8 21.1 3 7.8 4 10.5 2 5.3
Age at diagnosis
1) <20 4 66.6 1 16.7 0 0.0 0 0.0 1 16.7 ...
21 - 40 11 42.3 9 34.6 2 7.7 4 15.4 0 0.0
41 - 60 19 34.5 21 38.2 8 14.5 4 7.3 3 5.5
>60 28 43.8 16 25.0 7 10.9 7 10.9 6 9.4
2) ≤40 15 46.9 10 31.2 2 6.3 4 12.5 1 3.1 0.7521
>40 47 39.5 37 31.1 15 12.6 11 9.2 9 7.6
3) ≤60 34 39.1 31 35.6 10 11.5 8 9.2 4 4.6 0.5631
Histological type
1) SSM 12 23.1 26 50.0 9 17.3 4 7.7 1 1.9 ...
LM/LMM 35 67.3 13 25.0 1 1.9 3 5.8 0 0.0
NM 0 0.0 1 14.3 1 14.3 3 42.8 2 28.6
ALM 13 52.0 4 16.0 3 12.0 3 12.0 2 8.0
2) LM/LMM 35 67.3 13 25.0 1 1.9 3 5.8 0 0.0 <0.0011
Others 25 29.1 31 36.0 14 16.3 10 11.6 6 7.0
Location
1) Head and neck 24 52.2 11 23.9 5 10.9 5 10.9 1 2.2 ...
Acral 13 44.8 5 17.3 3 10.3 3 10.3 5 17.3
Trunk 12 36.3 11 33.3 5 15.2 3 9.1 2 6.1
Upper limb 9 36.0 12 48.0 1 4.0 2 8.0 1 4.0
Lower limb 4 26.7 7 46.6 3 20.0 1 6.7 0 0.0
Others 0 0.0 1 33.3 0 0.0 1 33.3 1 33.3
2) Head and neck 24 52.2 11 23.9 5 10.9 5 10.9 1 2.2 0.2931
Others 38 36.2 36 34.3 12 11.4 10 9.5 9 8.6
Signs and symptoms
Yes 37 41.1 27 30.0 10 11.1 10 11.1 6 6.7 0.2871
No 15 40.5 17 46.0 3 8.1 2 5.4 0 0.0
Bleeding
Yes 0 0.0 1 20.0 0 0.0 1 20.0 3 60.0 <0.0011
No 52 42.6 43 35.2 13 10.7 11 9.0 3 2.5
Alterations in sensitivity
Yes 5 27.8 6 33.3 3 16.7 4 22.2 0 0.0 0.1831
No 47 43.1 38 34.9 10 9.2 8 7.3 6 5.5
Darkening
Yes 10 33.3 9 30.0 5 16.7 3 10.0 3 10.0 0.2841
No 42 43.3 35 36.1 8 8.2 9 9.3 3 3.1
Growth
Yes 36 48.6 22 29.7 5 6.8 6 8.1 5 6.8 0.0991
No 16 30.2 22 41.5 8 15.1 6 11.3 1 1.9

1: Fisher Exact test, .... No test

In relation to the Clark level, 42% of the patients presented at level 1 or in situ, 13% at level II, 20% at level III, 21% at level IV and 4% at level V. The LM/LMM histological type was the most common at Clark levels I and II (84%). Among cases of level V CM, ALM was the most common histological type (11.5%) (Table 4).

Table 4 Comparison between Clark's level and sex, skin color, age at diagnosis, histological type, signs and symptoms and location 

Clark's level
Characteristics I II III IV V p-value
N % N % N % N % N %

Sex
Male 20 37.0 8 14.8 12 22.2 11 20.4 3 5.6 0.7541
Female 42 45.2 11 11.8 16 17.2 21 22.6 3 3.2
Skin color
White 41 38.0 14 13.0 25 23.1 25 23.1 3 2.8 0.0971
Non-white 21 53.8 5 12.8 3 7.7 7 18.0 3 7.7
Age at diagnosis
1) <20 4 80.0 1 20.0 0 0.0 0 0.0 0 0.0 ...
21 - 40 11 44.0 1 4.0 8 32.0 5 20.0 0 0.0
41 -60 19 34.6 11 20.0 11 20.0 13 23.6 1 1.8
>60 28 45.1 6 9.7 9 14.5 14 22.6 5 8.1
2) ≤40 15 50.0 2 6.6 8 26.7 5 16.7 0 0.0 0.3841
>40 47 40.2 17 14.5 20 17.1 27 23.1 6 5.1
3) ≤60 34 40.0 13 15.3 19 22.3 18 21.2 1 1.2 0.1861
Histological type
1) SSM 13 26.0 6 12.0 18 36.0 13 26.0 0 0.0 ...
LM/LMM 34 68.0 8 16.0 5 10.0 1 2.0 2 4.0
NM 0 0.0 0 0.0 1 12.5 7 87.5 0 0.0
ALM 13 50.0 3 11.5 2 7.8 5 19.2 3 11.5
2) LM/LMM 34 68.0 8 16.0 5 10.0 1 2.0 2 4.0 <0.0011
Others 26 30.6 9 10.6 22 25.9 25 29.4 3 3.5
Signs and symptoms
Yes 37 42.5 11 12.6 17 19.6 17 19.6 5 5.7 0.5301
No 15 41.6 6 16.7 10 27.8 5 13.9 0 0.0
Bleeding
Yes 0 0.0 0 0.0 1 16.7 3 50.0 2 33.3 0.0021
No 52 44.5 17 14.5 26 22.2 19 16.2 3 2.6
Alterations in sensitivity
Yes 5 29.4 3 17.7 3 17.7 6 35.2 0 0.0 0.3091
No 47 44.4 14 13.2 24 22.6 16 15.1 5 4.7
Darkening
Yes 10 33.3 5 16.7 7 23.3 6 20.0 2 6.7 0.7031
No 42 45.2 12 12.9 20 21.5 16 17.2 3 3.2
Growth
Yes 36 50.0 9 12.5 12 16.7 10 13.9 5 6.9 0.0391
No 16 31.4 8 15.7 15 29.4 12 23.5 0 0.0
Location
1) Head and neck 23 50.0 7 15.2 5 10.9 9 19.5 2 4.4 ...
Acral 13 43.4 4 13.3 1 3.3 8 26.7 4 13.3
Trunk 13 43.3 2 6.7 7 23.3 8 26.7 0 0.0
Upper limb 9 36.0 5 20.0 7 28.0 4 16.0 0 0.0
Lower limb 4 26.7 1 6.6 7 46.7 3 20.0 0 0.0
Others 0 0.0 0 0.0 1 100.0 0 0.0 0 0.0
2) Head and neck 23 50.0 7 15.2 5 10.9 9 19.5 2 4.4 0.4241
Others 39 38.6 12 11.9 23 22.8 23 22.8 4 3.9

1: Fisher Exact test, .... No test

Symptoms of and/or alterations in the primary lesion were reported by 96 (70.6%) of the patients. Among these, the most frequent report was of growth of the lesion (58.1%), followed by alterations in color by 22%, alterations in sensitivity by 14.7% and bleeding by 5.2%. Darkening was the most common alteration in color, while itching was the most common symptom regarding sensitivity. Reports of bleeding were most common for Clark levels IV and V (83.3%), along with tumor thicknesses > 4 mm (60%). Growth was most commonly reported for level I tumor invasions (50%). Alterations in sensitivity and color were associated with in situ and thin CM (Breslow's depth ≤ 1 mm), with 61.1% and 63.3% of cases, respectively, but for the last two characteristics there was no statistical significance (Tables 3 and 4).

Metastasis was detected in 19 (13.4%) patients. Lymph node metastasis was found in 12 (40%) patients, cutaneous in seven (24%), cerebral in five (17%), lung in four (13%), blood in one (3%) and bone, also in one patient (3%).

There were seven deaths resulting from CM. We have included here patients with giant congenital melanocytic nevi (GCMN) and xeroderma pigmentosum (PX). In the univariate analysis, there was a greater chance of death for men, those under the age of 20, non-whites, those with a Breslow's depth > 2 mm, with the ALM histological type and with a history of growth and bleeding. However, following the multivariate analysis, only those under the age of 20 and those with a history of bleeding continued to run a high relative risk of death, approximately 13 and eight times higher, respectively (Table 5).

Table 5 Multivariate model - evolution to death 

Final model Coefficient Standard error p-value Relative risk Confidence interval 95%
Age at diagnosis
<20 2.6 1.3 0.043 13.1 1.1 a 156.5
21 - 40 0.6 1.4 0.690 1.8 0.1 a 28.7
41 - 60 -0.4 1.4 0.790 0.7 0.04 a 11.1
>60 1.0
Bleeding
Yes 2.1 1.0 0.029 8.3 1.2 a 54.9
No 1.0

DISCUSSION

In countries with a lower incidence of CM, such as European countries and Brazil, the majority of cases occur among females and, in contrast, in countries with the high incidence, such as Australia, New Zealand and the USA, there is a predominance of the disease among males or even equality in terms of gender distribution.2

In this study, 61% of the patients were female, which is in line with the majority of studies that have been conducted in Brazil.11 - 16

In relation to age, CM is characterized by its incidence among young people. The average age at diagnosis is 52, which is 10 years less than what is reported for more common cancers such as breast, lung and prostate cancer.17 In this study, a mean age of 55 years was found, which is in consonance with Brazilian and international literature.11 , 13 , 14 , 18 - 20

Among risk factors for CM are phenotypical traits such as light or fair skin, hair and eyes, with studies indicating a risk of developing this neoplasm 10 times higher among Caucasians than other ethnicities.9 , 10 In the Southern region of Brazil, the majority of studies deal with a population that is almost 100% white.12 , 15 In this study, 74% of the patients were white and 26% were non-whites. This ratio of non-whites is higher than that found in the literature studied.11 , 16 , 21 , 22

Regarding histological type, the international literature reports that among white individuals, SSM predominates, varying from 37.7% in Chile to 43.6% in Argentina, 60% in Spain, 62% in Switzerland and 73.6% in Australia.19 , 23 - 26 In studies involving non-white individuals, ALM is the most prevalent.10 , 27 Forman et al, in the USA, however, found that 56% of the cases were LM/LMM and speculated whether this difference in relation to other studies is due to a change in risk factors or to a regional peculiarity.28

In the Brazilian literature, in the Southern region, where there is a prevalence of people with a light skin tone, SSM vary from 35.3 to 68.9%.12 , 14 , 29 - 35 In other regions in Brazil, where there is a greater proportion of non-whites, the SSM percentage is lower.16 , 22 Fernandes et al found a predominance of LM/LMM (29.8%), as did the present study.21 Despite the fact that the difference in proportion between LM/LMM (35.7%) and SSM (35.1%) found in this sample was small, the frequency of LM/LMM is still higher than that reported in the literature (5 to 10%), just as SSM is less frequent when compared with studies in which a white population is predominant (70%).10 This difference may be due to the frequency of non-whites found in this study (26%) and possibly to sun exposure patterns, although this last variable has not been measured.

In the current sample, SSM was most frequent on the trunk and on the extremities and ALM was responsible for almost 90% of the MM in acral regions. LM/LMM was the most common histological type on the head and neck, which is in concordance with the literature.1 A higher proportion of ALM was found (17.2%) and, once again, the most plausible explanation is the elevated frequency of non-whites in this sample.

The most common body location varies with gender and with the histological type. In the majority of studies, CM occurs with greater frequency on men's dorsal area and on women's legs.10

In Brazilian literature, two studies11 , 33 found a predominance of CM on the head and neck, as was found in the present study.

The risk of CM is increased in the presence of atypical melanocytic nevi.10 Gandini and collaborators, in a meta-analysis study, highlighted that finding an atypical nevus increased the risk of CM by 2.4 times, while 10 or more atypical nevi increased the risk by 32 times.36 In the present study, approximately 20% of the patients presented atypical nevi, which is a rate similar to that described in the literature.12 , 37

The risk of CM is higher in patients that suffer from non-melanoma skin cancer, including basal-cell carcinoma (BCC) and squamous-cell carcinoma (SqCC), since all three neoplasms are related to sun exposure.38 In the present study, the finding of 27,7% of the patients with a history of non-melanoma skin cancer is higher than that found in the literature.39 , 40 One possible explanation would be the chronic exposure to the sun, which may also explain the elevated frequency of CM on the head and neck, even though this feature has not been evaluated by this study.

A family history of CM is an important risk factor. Patients with a history of CM8 among their relatives face a risk twice as high, while three or more family members with CM may result in a risk that is 35 to 70 times higher.10 A history of CM in the family was observed in 10% of the cases, which is similar to the rate described by the majority of the authors.1 , 8 - 10 , 19 , 41

Breslow's tumor thickness is described in various studies as the most important prognostic factor for CM.2 , 42

According to the most recent version of the American Joint Committee on Cancer (AJCC), MMs are classified, based on Breslow's depth, as thin (≤ 1 mm), intermediate (1,01-4 mm) and thick (>4 mm).42 In recent years some papers, from many areas of the world, have mentioned an increase in the incidence of cases of thin CM, as well as the stabilization of the incidence of thick ones.2 , 18 , 20 , 43 Regarding the Brazilian literature, some authors have also found a high proportion of cases of thin CM, which is, therefore, in concordance with the findings in the international literature.14 , 15 , 22 , 44

In the sampling presented here, 72% of the cases of CM were restricted to the epidermis (in situ) or had a tumor thickness of ≤ 1 mm. One possible explanation for the predominance of CM in initial stages results from the diagnosis being made, in the majority of cases, by dermatologists, which contributed to the early identification of the tumor.45

The Clark level, or the tumor invasion level, was initially considered an independent prognostic factor. In the 2002 version of the AJCC's staging, this index was computed in order to evaluate the prognosis of lesions with a Breslow's depth of < 1 mm.42 Now, the Clark level is considered in the staging of thin CMs only when information on ulceration and the mitotic index is not available.42 In the present study, 55.1% of the patients presented Clark levels I and II, as found in the international literature, which reveals the predominance of lesions with lower invasion levels.20 , 46

Changes in color and size of a previous lesion and the appearance of a new lesion are the earliest characteristics that are noted by patients and that contribute to the diagnosis. Of these, growth has been highlighted as the most frequent clinical sign, followed by alterations in color.19 , 24 , 47 While in some studies growth has been associated with greater tumor thickness, in another this association was not detected.19 , 24 , 47

Alterations in color have been related to thinner CM and bleeding from thicker lesions.19 , 24 , 47 There is disagreement between authors regarding alterations in sensitivity, especially itching. For some, this is correlated with thin tumors while others associate it with thicker ones.10 , 19 , 24 , 47 In this study, 70.6% of the patients reported symptoms and/or signs in the lesion. The most common complaint was growth of the lesion, followed by darkening, itching and, less frequently, bleeding. This bleeding was associated with cases of thicker CM and was similar to the findings of the studies mentioned.19 , 24 , 47

Alterations in color, sensitivity and growth were more frequent in cases of thinner CM, though with no statistical significance.

Metastases denote a poor prognosis, with survival times estimated in months. As a rule, the greater the thickness of the tumor, the greater the chances of metastasis.46 Cases of CM diagnosed with a Breslow's thickness < 1.5 mm show 10-year survival rates above 90%.48 The CM disseminates via lymphatic and/or hematogenic routes. The former occurs very early in the evolution of CM. Distant metastases may be either non-visceral (skin, subcutaneous tissue and nonregional lymph node), or visceral, most frequently affecting the lungs (18- 36%), followed by the liver (1420%), brain (12-20%), bones (11-17%) and gastrointestinal tract (1-7%).10

In the sample assessed, metastases were found in 19 (13.4%) patients, with the lymph nodes being the most common (40%). In 80% of the patients suffering from a metastasized disease, the tumor thickness was > 1.5 mm, which is similar to the findings from the literature referred to above.

In addition to tumor thickness, other factors associated with survival rates are: sex, age, primary body location, histological type, skin color and staging.10

The men presented a higher proportion of cases of thick CM than the women, resulting in worse survival rates.3 , 15 , 18 , 24 Patients over the age of 60 also had poor prognoses with an elevated proportion of cases of thick CM.3 , 18 , 49 Balch et al recorded that, even when controlling confusing factors such as gender and tumor thickness, age was an independent prognostic factor in CM.46 This may be attributed to a decline in the immunological response which occurs withage.46 , 49

Regarding body location, the majority of the authors consider the trunk to be the location with the poorest prognosis.3 , 15 , 18 , 25 , 47 In relation to the histological type, some investigations have observed worse survival rates with NM15 , 18 while others have recorded that ALM was associated with a poorer prognosis.47

As for skin color, Chang et al recorded worse survival rates for non-white individuals both in the initial and later stages, though this association did not have an impact on the multivariate analysis.18

In this sample a greater chance of death was found among men, those under the age of 20, nonwhites, those with the ALM histological type, those with a Breslow's depth > 2 mm and those with a history of growth and bleeding. However, following the multivariate analysis, only the patients under the age of 20 and with a history of bleeding remained at high risk.

Children and adolescents suffer from CM less frequently than adults, which may delay diagnosis, thereby worsening survival rates.10 , 41 , 50 The two cases diagnosed in children occurred in patients bearing GCMN, which is known to carry a poor prognosis, since two thirds of cases of CM in people with GCMN are of a non-epidermic origin, which leads to a late diagnosis and, consequently, reduced survival rates.41

Bleeding is most common in cases of thick CM according to the literature and to the results found in this study, which denotes a poor prognosis.19 , 24 , 47

CONCLUSIONS

In conclusion, in this sample, cases of CM were more common in women, with a mean age of 55 years at diagnosis. Whites represent the majority of patients, but the proportion of non-whites is bigger than other Brazilian states, justifying the higher frequency of ALM than cited in other studies. The most frequent histological type was LM/LMM, that could be because of regional factors and matches with predominant location. Comparing sex and body site, CM was more common in trunk of men and limbs of women. In agree with national and international tendencies, the majority of CM was in situ or thin. The number of deaths was small, with more chances of dying in those under 20 years and with a history of bleeding, after a multivariate analysis. In that way, more attention should be taken to children and teenagers, mostly those with predisposing factors such GCMN, PX and atypical nevus, besides CM in these population being rare, the diagnosis of CM with less than 1 mm of growth, is still the best chance of cure for the most lethal type of skin cancer.

This sample analyzed is small and came from a single institution, meaning that these findings cannot be generalized for the entire population of the state. This data may not reflect the reality of CM among the population of this state. Further studies are necessary in order to validate the results obtained. However, this study represents the largest sampling of patients with melanoma in this state, with data collected over the longest period of time.

REFERENCES

1. Markovic SN, Erickson LA, Rao RD, Weoenig RH, Pockaj BA, Bardia A, et al. Malignant melanoma in the 21st sentury, part 1: epidemiology, risk factors, screening, prevention, and diagnosis. Mayo Clin Proc. 2007;82:364-80. [ Links ]

2. Garbe C, Leiter U. Melanoma epidemiology and trends. Clin Dermatol. 2009;27:3-9. [ Links ]

3. Liang JJ, Robinson E, Martin RC. Cutaneous melanoma in New Zealand: 2000-2004. ANZ J Surg. 2010;80:312-6. [ Links ]

4. Altekruse SF, Kosary CL, Krapcho M, Neyman N, Aminou R, Waldron W, et al, editors. SEER Cancer Statistics Review, 1975-2007 (Vintage 2009 Populations), National Cancer Institute. [ cited 2010 mar 13]. Available from: http://seer.cancer.gov/csr/1975_2007/. [ Links ]

5. Inca.gov [Internet]. Estimativa 2010: Incidência de Câncer no Brasil [acesso 20 nov 2010]. Disponível em: http://www.inca.gov.br/estimativa/2010/index.asp?link=conteudo_view.asp&ID=5Links ]

6. Desmierre MF, Chung C, Miller DR, Geller AC. Early detection of thick melanomas in the United States: beware of the nodular subtype. Arch Dermatol. 2005;141:745-50. [ Links ]

7. Geller AC, Swetter SM, Brooks K, Demierre MF, Yaroch AL. Screening, early detection, and trends for melanoma: current status (2000-2006) and future directions. J Am Acad Dermatol. 2007;57:555-72. [ Links ]

8. Berwick M, Erdei E, Hay J. Melanoma epidemiology and public health. Dermatol Clin. 2009;27:205-14. [ Links ]

9. Belfort FA, Waisntein AJA. Etiopatogenia - Melanócito ao Melanoma. In: Belfort FA, Waisntein AJA, editores. Melanoma: diagnóstico e tratamento. São Paulo: Lemar; 2010. p. 37-45. [ Links ]

10. Paek SC, Sober AJ, Tsao HT, Mihm Jr MC, Johnson TM. Cutaneous melanoma. In: Wolf K, Goldsmith LA, Katz SI, Gilchrest BA, Paller AS, Leffell DJ, editors. Fitzpatrick's Dermatology in general medicine. New York: Mc Graw Hill; 2008. p.1134-57 [ Links ]

11. Lucas EA, Deps PD, Lima JGB, Toriblo R, Gomes CC. Malignant melanoma: retrospective study in the period 1982-1992, in the University Hospital of UFES. Arq Bras Med. 1994;68:67-70. [ Links ]

12. Bakos L, Wagner M, Bakos R, Leite CSM, Sperhacke CL, Dzekaniak KS, et al. Sunburn, sunscreen, and phenotypes: some risk factors for cutaneous melanoma in southern Brazil. Int J Dermatol. 2002; 41:557-62. [ Links ]

13. Gon AS, Minelli L, Guembarovski AL. Primary cutaneous melanoma in Londrina. An Bras Dermatol. 2001;76:413-26. [ Links ]

14. Borges SZ, Bakos L, Cartell A, Wagner M, Agostini A, Lersch E. Distribution of clinical-pathological types of cutaneous melanomas and mortality rate in the region of Passo Fundo, RS, Brazil. Int J Dermatol. 2007; 46:679-86. [ Links ]

15. Dimatos DC, Duarte FO, Machado RS, Vieira VJ, Vasconcelos ZAA, Bins-Ely J et al. Melanoma cutâneo no Brasil. Arq Cat Med. 2009;38:S14-19 [ Links ]

16. Ferrari Júnior NM, Muller H, Ribeiro M, Maia M, Sanches Júnior JA. Cutaneous melanoma: descriptive epidemiological study. Sao Paulo Med J. 2008; 126:41-7. [ Links ]

17. Berwick M, Wiggins C. The current epidemiology of cutaneous malignant melanoma. Front Biosci 2006; 11:1244-54. [ Links ]

18. Chang AE, Karnell LH, Menck HR. The National Cancer Data Base report on cutaneous and noncutaneous melanoma: summary of 84,836 cases from the past decade. The American College of Surgeons Commission on Cancer and the American Cancer Society. Cancer.1998;83:1664-78. [ Links ]

19. Cabrera R, Silva S, Días de Medina J, Hoell I, Gugliemetti A, Rohmann I. Clinical study of 113 cases of malignant melanoma. Rev Med Chil. 1994;122:900-6. [ Links ]

20. Buettner PG, Leiter U, Eigentler TK, Garbe C. Development of prognostic factors and survival in cutaneous melanoma over 25 Years. An analysis of the Central Malignant Melanoma Registry of the German Dermatological Society. Cancer. 2005;103:616-24. [ Links ]

21. Fernandes NC, Cardoso ICL, Maceira J, Perez M. Melanoma: estudo retrospectivo de 47 casos. An Bras Dermatol. 1996;71:381-85. [ Links ]

22. Fernandes NC, Calmon R, Maceira JP, Cuzzi T, Silva CSC. Cutaneous melanoma: prospective study of 65 cases. An Bras Dermatol. 2005;80:25-34. [ Links ]

23. Loria D, Matos E. Risk factors for cutaneous melanoma: a case-control study in Argentina. Int J Dermatol. 2001;40:108-14. [ Links ]

24. Nagore E, Oliver V, Botella-Estrada R, Moreno-Picot S, Guillén C, Fortea JM. Clinicopathological analysis of 1571 cutaneous malignant melanoma in Valencia, Spain: factors related to tumour thickness. Acta Derm Venerol. 2006;86:50-6. [ Links ]

25. Lindholm C, Anderson R, Dufmats M, Hansson J, Ingvar C, Möller T, et al. Invasive cutaneous malignant melanoma in Sweden, 1990-1999. A prospective, populationbased study of survival and prognostic factors. Cancer. 2004;101:2067-78. [ Links ]

26. Garbe C, MMleod GR, Buettner PG. Time trends of cutaneous melanoma in Queensland, Australia and Central Europe. Cancer. 2000;89:1269-78. [ Links ]

27. Cress RD, Holly EA. Incidence of cutaneous melanoma among non-Hispanic Whites, Hispanics, Asians, and Blacks: analysis of California Cancer Registry data, 1988-93. Cancer Causes Control. 1997;8:246-52. [ Links ]

28. Forman SB, Ferringer TC, Peckham SJ, Dalton SR, Sasaki GT, Libow LF, et al. Is superficial spreading melanoma still the most common form of malignant melanoma. J Am Acad Dermatol. 2008;58:1013-20. [ Links ]

29. Bakos L. Melanomas malignos e etnia. An Bras Dermatol. 1991;66:299-302. [ Links ]

30. Ponzio HA, Bernadi CDV, Favaretto AL, Brancher MM. Rate of malignant melanoma in Service of Dermatology of ISCMPA/UFRGS. An Bras Dermatol. 1998;73:S8. [ Links ]

31. Moreno M. Perfil dos pacientes com melanoma no oeste de Santa Catarina, Brasil. Grupo Brasileiro de Melanoma. Boletim informativo do GBM. 2005;3:3. [Acesso 13 mar 2010] Disponível em: http://www.gbm.org.br/gbm/boletim/2005/infor31.aspx. [ Links ]

32. Battisti R, Nunes DH, Weber A, Schweitzer LC, Sgrott I. Evaluation of the epidemiological profile and the mortality rate of the patients with primary cutaneous melanoma in Florianopolis - SC, Brazil. An Bras Dermatol. 2009; 84:335-42. [ Links ]

33. Nasser N. Epidemiologia do melanoma maligno em Blumenau-SC. An Bras Dermatol. 1993;68:17-20. [ Links ]

34. Minelli L, Pereira VL. Melanoma - Estudo casuístico do Instituto de Câncer de Londrina. An Bras Dermatol. 1983;58:81-4. [ Links ]

35. Lebsa-Weber A, Nunes DH, Filho JJS, Pinto CJC. Assessment of 496 pathological reports of melanoma diagnosed in the city of Florianopolis, SC, Brazil An Bras Dermatol. 2007;82:227-32. [ Links ]

36. Gandini S, Sera F, Cattaruzza MS, Pasquini P, Abeni D, Boyle P, et al. Meta-analysis of risk factors for cutaneous melanoma: I. Common and atypical naevi. Eur J Cancer. 2005;41:28-44. [ Links ]

37. Ferrone CR, Ben Porat L, Panageas KS, Berwick M, Halpern AC, Patel A, et al. Clinicopathological features of and risk factors for multiple primary melanomas. JAMA. 2005;294:1647-54. [ Links ]

38. Armstrong BK, Kricker A. The epidemiology of UV induced skin cancers. J Photochem Photobiol B. 2001;63:8-18. [ Links ]

39. Wolff J, Wollina U. Second malignancies in melanoma patients in Thuringia. J Eur Acad Dermatol Venereol. 2000;14:479-83. [ Links ]

40. van der Velden HM, van Hossum NM, Blokx WA, Boezeman JB. Gerritsen MJ. Clinical characteristics of cutaneous melanoma and second primary malignancies in a dutch hospital-based cohort of cutaneous melanoma patients. Dermatol Res Pract. 2009. doi: 10.1155/2009/479183 [ Links ]

41. Jen M, Murphy M, Grant-Kels J. Childhood melanoma. Clin Dermatol. 2009;27:529-36. [ Links ]

42. Balch CM, Gershenwald JE, Soong SJ, Thompson JF, Atkins MB, Byrd DR, et al. Final Version of 2009 AJCC Melanoma Staging and Classification. J Clin Oncol. 2009;27:6199-206. [ Links ]

43. Geller AC, Swetter SM, Brooks K, Demierre MF, Yaroch AL. Screening, early detection, and trends for melanoma: current status (2000-2006) and future directions. J Am Acad Dermatol. 2007;57:555-72. [ Links ]

44. Castro LGM, Toyama CL, Gomes AP, Freire MA, Brito TF. Câncer de pele em clínica particular em São Paulo - SP. An Bras Dermatol. 1996;71:471-6. [ Links ]

45. Fisher NM, Schaffer JV, Berwick M, Bolognia JL. Breslow depth of cutaneous melanoma: impact of factors related to surveillance of the skin, including prior skin biopsies and family history of melanoma. J Am Acad Dermatol. 2005;53:393-406. [ Links ]

46. Balch CM, Soong SJ, 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. [ Links ]

47. Negin BP, Riedel E, Oliveria SA, Berwick M, Coit DG, Brady MS. Symptoms and signs of primary melanoma. Important indicators of Breslow Depth. Cancer. 2003;98:344-8. [ Links ]

48. Lens MB, Dawes M. Global perspectives of contemporary epidemiological trends of cutaneous malignant melanoma. Br J Dermatol. 2004;150:179-85. [ Links ]

49. Hedge UP, Chakraborty N, Kerr P, Grant-Kels JM. Melanoma in the elderly patient: relevance of the aging immune system. Clin Dermatol. 2009;27:537-44. [ Links ]

50. Downard CD, Rapkin LB, Gow KW. Melanoma in children and adolescents. Surg Oncol. 2007;16:215-20. [ Links ]

* Work performed at the Dermatology Service at Federal University of Minas Gerais Hospital (HC-UFMG) - Belo Horizonte (MG), Brasil.

Received: May 06, 2012; Accepted: August 07, 2012

MAILING ADDRESS: Flávia Vieira Brandão, CCSW 3, Lote 5, Bloco A, apto 205 - Sudoeste, 70680-350 - Brasília - DF, Brazil, E-mail: flaviavieirabrandao@yahoo.com.br

Financial Support: None.

Conflict of Interest: None.

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.