Nevus-Associated Melanomas Clinicopathologic Features Danielle Shitara, MD, 1 Mauricio M. Nascimento, MD, 1 Susana Puig, MD, PhD, 2 Sérgio Yamada, MD, 1 Milvia M. S. S. Enokihara, MD, PhD, 1,3 Nilceo Michalany, MD, 1,3 and Ediléia Bagatin, MD, PhD 1 From the 1 Department of Dermatology, Federal University of São Paulo, São Paulo, Brazil; 2 Melanoma Unit, Dermatology Department, Hospital Clinic of Barcelona, IDIBAPS, Universitat de Barcelona, Barcelona, Spain and CIBER de Enfermedades Raras, Instituto de Salud Carlos III, Barcelona, Spain; and 3 Department of Pathology, Federal University of São Paulo, São Paulo, Brazil. Key Words: Nevus-associated melanomas; De novo melanoma; Breslow; Histologic subtype Am J Clin Pathol October 2014;142:485-491 ABSTRACT Objectives: The clinical significance of nevus-associated melanoma compared with de novo melanomas remains controversial. It has been suggested that nevus-associated melanomas have a higher Breslow thickness and therefore worse prognosis. Over a 10-year period, this study evaluated the incidence of nevus-associated melanoma and its prognostic significance related to clinicopathologic features. Methods: Cross-sectional study from 1995 through 2004 in a dermatopathology referral center. With available data, we evaluated sex, primary location, histologic subtype, Breslow thickness, Clark level, presence of ulceration, associated lesion, and histologic subtype of the associated lesion. Results: Of 135,653 pathologic records from skin biopsy specimens over a 10-year period, 1,190 melanoma records were selected. Nevus-associated melanomas corresponded to 390 (32.8%) melanomas, with thin melanomas having a nevus 1.52 times the association observed with thick melanomas (>1.01 mm; 95% confidence interval, 1.16-1.99; P <.001). Superficial spreading melanoma was the most frequent, while no lentigo maligna melanoma was associated with nevi. The median Breslow thickness of nevus-associated melanomas was lower than that of de novo melanomas. Conclusions: Nevus-associated melanomas, which represent one-third of the melanomas in southeast Brazil, are associated with intermittent sun exposure, superficial spreading melanomas, and lower Breslow thickness. This is one of the largest series describing nevus-associated melanomas in Latin America. The global incidence of melanoma has risen in the past 50 years, despite increasing awareness and public health campaigns. 1,2 In the United States, the incidence rates have risen from 13 per 100,000 in 1990 to 21 per 100,000 in 2006, corresponding to the sixth most frequently diagnosed cancer in men and seventh in women and one of the most frequently diagnosed cancers in youth. 1 In southern Brazil, there has also been an increase in the relative incidence of melanoma, 38% among men and 11% among women, between 1979 and 1987. 3 According to the National Institute of Cancer (Instituto Nacional do Câncer), 4 3,170 new cases of melanoma in men and 2,060 in women were expected in Brazil in 2012, especially in the southern region of Brazil, where the population of European descent is greater. The lack of a centralized melanoma case registry and of mandatory disease notification may account, among other factors, for the nonconformity of national and international data. 5 Among the melanoma risk factors described in the literature, the presence of multiple melanocytic nevi (>100 nevi; relative risk [RR], 6.89; 95% confidence interval [CI], 4.63-10.25) and atypical melanocytic nevi (RR, 6.36; 95% CI, 3.80-10.33) 6,7 stands out, but the influence of preexisting pigmented lesions in melanoma genesis remains controversial. Considering epidemiologic and histologic studies, melanoma seems to arise either in association with a preexisting nevus or de novo, without any associated lesion. 8 The described incidence of melanoma associated with a preexisting lesion may vary from 20% to 50%. 9-16 The clinical significance of nevus-associated melanomas compared with de novo melanomas remains controversial. Some authors suggest that nevus-associated melanomas may have a higher Breslow thickness and therefore worse Am J Clin Pathol 2014;142:485-491 485 485 485
Shitara et al / Nevus-Associated Melanomas prognosis. 12 Other authors 13,14 have obtained different results with a lower Breslow thickness for nevus-associated melanomas. There have been suggestions that the development of nevus-associated melanoma would increase the risk of having another nevus-associated melanoma by 9-fold. 11 We aim to evaluate the incidence of nevus-associated melanoma and its association with clinical and pathologic features over a 10-year period. Materials and Methods This cross-sectional study was carried out in a dermatopathology center in São Paulo, Brazil. The study protocol was submitted and approved by research ethics committees of Universidade Federal de São Paulo/UNIFESP- EPM (number 188.713, January 24, 2013). All data available in pathologic records from skin biopsy specimens over a 10-year period (1995-2004) from a dermatopathology referral center were evaluated, and only melanoma records were included in this study. All melanoma cases were evaluated by a single experienced dermatopathologist (N.M.) using previously described morphologic features. 17-19 The following parameters available in the melanoma pathologic records were considered for clinicopathologic correlation: sex, primary location, histologic subtype, Breslow thickness, Clark level, presence of ulceration, associated lesion, and histologic subtype of the associated lesion. Only data available in the pathologic records were included in the study. Regarding primary location, head and neck were grouped as Head ; trunk and lumbar areas were grouped as Trunk. Histologic subtypes were classified as follows: superficial spreading melanoma (SSM), nodular melanoma (NM), lentigo maligna melanoma (LMM), acral lentiginous melanoma, and other (fusiform melanoma, desmoplastic melanoma, amelanotic melanoma, and melanoma metastasis). Since the distinction of melanoma cells and nevus cells, especially in nevoid melanomas, can be difficult and controversial, we chose not to include any of the nevoid melanoma cases in this study. 20,21 Lesions located in the trunk and extremities were considered located in intermittent sun-exposed areas, while lesions in the head were considered located in chronic sunexposed areas. Genital and acral lesions were considered non sun-exposed areas. Statistical Analysis For each analyzed variable, only data available were considered valid. Associated nevus was described with percentages and 95% CIs. To estimate the relation of associated nevus with selected risk factors, we calculated prevalence ratios (PRs) and 95% CIs, with associated nevus as a dependent variable and various exposure factors as independent variables. Most independent variables were grouped into two or three categories. Variables selected in the univariate analysis at a.20 significance level were included in the multivariate analysis. PRs and their 95% CIs were estimated using a generalized linear model with binomial distribution and log link function. 22 Statistical significance was assessed using the c 2 test. The Mann-Whitney test was used for comparing the Breslow thicknesses of de novo and nevus melanomas. A two-sided P value of less than.05 was considered to indicate statistical significance. All analyses were performed using STATA/SE 12.1 for Windows (StataCorp LP, College Station, TX). Results Of 135,653 pathologic records from skin biopsy specimens over a 10-year period (1995-2004), only the records with the diagnosis of melanoma were selected for the evaluation, totaling 1,190 melanoma records. Unfortunately, not all melanoma records presented all the parameters evaluated, with information missing in some cases. Therefore, only available data were considered valid for statistical analysis. Melanomas were located mainly in trunk and extremities (424 [43.0%] of 985 and 305 [31.0%] of 985, respectively) Table 1. The most common histologic subtype was SSM (517 [62.3%] of 830) (Table 1). Most melanomas without any specified subtype corresponded to in situ melanomas (87.5%). Of all melanomas evaluated, 707 (61.1%) of 1,147 were invasive melanomas, of which 446 (63.1%) of 707 were thin melanomas (Table 1). LMM was significantly associated with chronic sun-exposed areas (P <.001), while SSM and NM were related to intermittent sun-exposed areas. Almost three-fourths of the melanomas (729 [74.0%] of 985) were located in intermittent sun-exposed areas. Of the general melanoma sample, 390 (32.8%) of 1,190 (95% CI, 30.1%-35.5%) corresponded to nevus-associated melanomas. Regarding primary location, acral and head did not show any association with melanoma arising in a nevus (P =.706). Nevertheless, the trunk was significantly more likely to be associated with a nevus compared with the head (PR, 2.92; 95% CI, 2.15-3.97; P <.001; Table 1). No significant association was established between nevusassociated melanoma and sex (P =.548). When considering sun-exposed areas and nevus association, melanomas in chronic sun-exposed areas were associated with a lower risk of preexisting nevus compared with melanomas in intermittent sun-exposed areas (Table 1). Of all histologic subtypes, SSM represented most cases of nevus-associated melanoma (P <.001; 486 Am J Clin Pathol 2014;142:485-491 Downloaded 486 from https://academic.oup.com/ajcp/article-abstract/142/4/485/1766613
Table 1 Distribution of Melanoma Sample and Prevalence Ratio of Associated Lesions According to Clinical and Pathologic Variables Nevus Absent, Nevus Present, Crude PR Adjusted PR Variable No. (%) No. (%) P Value (95% CI) (95% CI) Histologic subtype n = 589 n = 241 <.001 P <.001 P <.001 SSM 302 (51.3) 215 (89.2) 3.02 (2.08-4.38) 2.65 (1.61-4.35) LMM a 124 (21.0) 0 1.00 1.00 NM 113 (19.2) 21 (8.7) ALM 33 (5.6) 3 (1.3) Other 17 (2.9) 2 (0.8) Location n = 668 n = 317 <.001 P <.001 P =.008 Acral 49 (7.4) 10 (3.2) 1.00 1.00 Head 165 (24.7) 29 (9.1) Genital b 1 (0.1) 2 (0.6) Extremities 219 (32.8) 86 (27.1) 1.83 (1.30-2.57) 1.27 (0.83-1.96) Trunk 234 (35.0) 190 (60.0) 2.92 (2.15-3.97) 1.70 (1.15-2.51) Solar exposure n = 668 n = 317 <.001 P <.001 No 50 (7.5) 12 (3.8) 1.00 Intermittent 453 (67.8) 276 (87.0) 1.95 (1.17-3.28) Chronic 165 (24.7) 29 (9.2) 0.77 (0.42-1.42) Breslow thickness n = 776 n = 382.210 P <.001 In situ melanomas 312 (40.2) 139 (36.4) Invasive melanomas 464 (59.8) 243 (61.3) Thin ( 1 mm) c 267 (58.6) 179 (74.6) <.001 1.52 (1.16-1.99) Thick (>1 mm) c 189 (41.4) 61 (25.4) <.001 1.00 1.01-2.00 mm 98 (51.9) 49 (80.3) 2.01-4.00 mm 63 (33.3) 10 (16.4) >4.00 mm 28 (14.8) 2 (3.3) Sex n = 789 n = 388.502 P =.548 Female 417 (52.9) 197 (51.0) 1.00 Male 372 (47.1) 191 (49.0%) 1.06 (0.88-1.27) Clark level d n = 760 n = 381 <.001 P =.002 I 315 (41.4) 140 (36.7) 1.75 (1.19-2.58) II 146 (19.2) 79 (20.7) III 194 (25.5) 134 (35.2) IV 105 (13.8) 28 (7.3) 1.00 ALM, acral lentiginous melanoma; CI, confidence interval; LMM, lentigo maligna melanoma; NM, nodular melanoma; Other, fusiform melanoma, amelanotic melanoma, desmoplastic melanoma, and metastasis of melanoma; PR, prevalence ratio; SSM, superficial spreading melanoma;. a For PR value, LMM cases were not taken into account. b For PR value, the only three cases of genital melanoma were not taken into account. c Eleven melanomas did not present a measurable Breslow thickness (8 in general melanoma sample and 3 in nevus-associated melanoma sample). d For PR value, Clark V cases were not taken into account. Table 1). After multivariate analysis, primary location and histologic subtype were the variables independently related to associated nevus (Table 1). Of 390 nevus-associated melanomas, 243 (62.3%) were invasive melanomas Image 1 and Image 2. Applying the Mann-Whitney test, the median Breslow thickness for de novo melanomas (0.85; interquartile range, 0.50-1.75) was greater than that for nevus-associated melanomas (0.70; interquartile range, 0.45-1.08; P <.001) Table 2. Discussion Because of Brazil s ethnic diversity, the most common histologic subtype of melanomas varies according to the region in question, with SSM being the most frequent histologic subtype among the central-southern populations. The melanoma samples for this study were collected in São Paulo, located in southeast Brazil, where the white population is predominant, mainly due to European immigration. Our findings corroborate the international literature, in which SSM was the most common histologic subtype observed (517 [62.3%] of 830; Table 1), and the primary location was the trunk, followed by extremities. 1,5,23 LMM was the histologic subtype most related to chronic sun-exposed areas (P <.001), while SSM and NM were related to intermittent sun-exposed areas, corroborating the idea suggested by Whiteman et al 24 that LMM is widely accepted to be caused by chronic sun exposure. 25 The described nevus melanoma incidence may vary from 20% to 50% in the literature, considering clinical and epidemiologic studies. Regarding studies in which histologic evaluation of melanomas was conducted, benign nevus was found in association with 20.6% to 42% of melanomas, in agreement with the results obtained from this study (390 [32.8%] of 1,190 melanomas). 12-14,26,27 The variables that were statistically significant in association with nevus-associated melanomas in univariate analysis were primary location, histologic subtype, and Breslow thickness, but after multivariate analysis, the only variables Am J Clin Pathol 2014;142:485-491 487 487 487
Shitara et al / Nevus-Associated Melanomas B A Image 1 Nevus-associated melanoma. A, Melanoma associated with a nevus showing melanoma cells in the epidermis and associated dermal nevus (H&E, 100). B, Same lesion showing positivity of melanoma cells and negativity of nevus cells (immunohistochemistry HMB-45, 200). C, Same lesion showing intense positivity of intraepidermal melanoma cells and weakly positive nevus cells (Melan A, 100). C A B Image 2 Nevus-associated melanoma. A, Melanoma on the left side with large epithelioid atypical cells and nevus on the right side in dermis with small nonatypical cells (H&E, 100). B, Detail of melanoma and nevus collision area (H&E, 200). 488 Am J Clin Pathol 2014;142:485-491 Downloaded 488 from https://academic.oup.com/ajcp/article-abstract/142/4/485/1766613
Table 2 Median Breslow Thickness Related to Presence or Absence of Associated Nevus Associated Nevus Breslow Thickness, mm Yes No Median 0.70 0.85 Interquartile range 0.45-1.08 0.50-1.75 Minimum-maximum 0.25-8.00 0.20-15.00 No. of observations 240 456 that remained independently associated with nevusassociated melanoma were primary location (P =.008) and histologic subtype (P <.001; Table 1). When considering the anatomic distribution of nevusassociated melanomas, most melanomas were located on intermittent sun-exposed areas (trunk and extremities; Table 1). 13,28 The finding that the trunk is the most frequent primary location is in agreement with previous studies, 12,13 being 2.92 times more likely to be associated with a nevus-associated melanoma than other locations (95% CI, 2.15-3.97; P <.001; Table 1). What stands out is that acral and head did not show any association with nevi in the majority of melanomas (49 [83.1%] of 59 and 165 [85.1%] of 194, respectively; Table 1). It has been suggested that this difference is unlikely to be due to the anatomic distribution of nevi, because nevus density, as a function of surface area, has been reported to be higher on the face than on the trunk. 13 Of all histologic subtypes, the association of a nevus with SSM was 3.02 times the association observed with other histologic subtypes (95% CI, 2.08-4.38; P <.001; Table 1). It has been suggested that histologic subtype rather than thickness seems to be the primary factor, with SSM more likely to arise in a nevus than other types of melanoma. 13,29 Indeed, some authors found that NM was less related to nevus even after controlling for Breslow thickness. 29 In addition, no LMM was related to nevus, despite being the melanoma subtype characterized by slow horizontal spreading along the basal cell layer, 19 corroborating previous findings that it corresponds to the histologic subtype least likely to be associated with a nevus (Table 1). 13,16,30 This may partially be due to the fact that LMM shows distinct clinical and epidemiologic characteristics and may arise from a different pathogenic pathway than does nevus-associated melanoma. 13,31 LMM is more frequent in chronically sundamaged areas, while nevus-associated melanomas would be more frequent in intermittent sun-exposed areas. 24,25 It has been suggested that there may be different pathways in melanoma genesis and that one determinant of melanoma development would be the propensity of the host to develop nevi. 13,24,28,31-33 In the first pathway, melanocytes in nevusprone individuals would require modest amounts of sunlight only in the early stages of tumorigenesis, with host factors driving tumor progression thereafter, and in another pathway, melanocytes of people with a low propensity to develop nevi would require long-term accumulation of sun exposure for melanoma to develop. In the latter pathway, melanomas would increase with age and affect the chronically sunexposed skin of the head and neck, face, and distal and dorsal extremities, with the surrounding skin typically showing marked solar elastosis and typically not in association with a precursor melanocytic nevus or in patients with numerous melanocytic nevi. Microscopically, these melanomas show a radial growth phase comprising melanocytes arranged as solitary units with poor lateral circumscription. On the other hand, the nevus-related pathway would affect younger patients, decreasing with older age, arising primarily on the intermittently sun-exposed skin of the trunk and proximal extremities on individuals with multiple melanocytic nevi. Microscopically, these melanomas are composed of larger melanocytes that are arranged primarily in nests and display upward intraepidermal scatter in their radial growth phase. Our findings that nevus-associated melanomas were more frequent on the trunk and extremities, in intermittent sunexposed areas (P <.001), and of the SSM subtype, while most melanomas located on the head were not associated with a nevus and no LMM was found in association with a nevus, corroborate previous findings that there may be more than one pathway to melanoma genesis. Roughly two-thirds of nevus-associated melanomas corresponded to invasive melanomas. Comparing thin and thick melanomas, thin melanomas were significantly associated with a nevus (PR, 1.52; 95% CI, 1.16-1.99; P <.001), in agreement with previous publications, 14,34 even though this parameter did not remain significant after multivariate analysis. The greater incidence of nevusassociated melanoma in thin melanomas may be justified partially by the fact that melanoma progression destroys nevus cells, so early diagnosis may increase the chance of finding melanoma with associated nevus. 30,35 Nevertheless, other authors did not find an association of an eclipsing of nevus cells by growing melanomas. 36 The distribution of nevus subtypes observed in this study is pursuant to the findings of Marks et al, 30 in that more than half of nevi associated with melanoma were common acquired nevi Table 3. Nevertheless, we found a smaller percentage of atypical (or dysplastic) nevi. This difference may be explained by the lack of consensus regarding the diagnosis of atypical nevi in the literature. 16,37-40 The prognostic significance of nevus-associated melanoma remains controversial in the literature. In a prospective study, Weatherhead et al 12 suggested that nevus-associated melanoma had a higher Breslow thickness and therefore a worse prognosis compared with de novo melanomas. However, Garcia-Cruz et al 35 could not demonstrate a significant difference between Am J Clin Pathol 2014;142:485-491 489 489 489
Shitara et al / Nevus-Associated Melanomas Table 3 Subtypes of Associated Lesions the prognosis of nevus-associated and de novo melanomas. Kaddu et al, 14 after evaluation of 667 nevus-associated melanomas and 519 de novo control melanomas, found lower Breslow thickness and a 5-year survival rate of 93.5% for nevus-associated melanomas, even though the latter did not show statistical significance. Purdue et al 13 also obtained a lower Breslow thickness for nevus-associated melanomas. The present study corroborates the findings of Kaddu et al 14 and Purdue et al. 13 The median Breslow thickness of nevus-associated melanomas was lower than that of de novo melanomas (Table 2), suggesting a more favorable prognosis based on the Breslow thickness of the primary melanoma, 3,41,42 although a 5-year survival assessment was not possible in this study due to the follow-up data being unavailable. This is one of the largest series describing melanomas associated with nevus described in Latin America. Melanoma arising in a nevus represents one-third of the melanomas in southeast Brazil, being associated with thin, superficial spreading melanomas, located mainly on the trunk, corroborating the data in the international literature. 13,14 Address reprint requests to Dr Shitara: Dept of Dermatology, Federal University of São Paulo, São Paulo, Brazil, R Borges Lagoa, 504 zip 04038-001, São Paulo-SP, Brazil; danielleshitara@yahoo.com.br. The research was funded by grants from FAPESP (São Paulo Research Foundation/Fundação de Amparo à Pesquisa do Estado de São Paulo), process number 2012-15238-0. The sponsors had no role in the design and conduct of the study; the sponsors also had no role in the collection, analysis, and interpretation of data. References Associated Lesion, No. (%) Subtype Without Atypia Atypical Junctional nevus 79 (20.2) 51 (13.1) Compound nevus 72 (18.4) 37 (9.5) Intradermal nevus 37 (9.5) 1 (0.4) Congenital nevus 6 (1.5) Not specified 94 (24.1) 13 (3.3) 1. Geller AC, Annas GD. Epidemiology of melanoma and nonmelanoma skin cancer. Semin Oncol Nurs. 2003;19:2-11. 2. Giblin AV, Thomas JM. Incidence, mortality and survival in cutaneous melanoma. J Plast Reconstr Aesthetic Surg. 2007;60:32-40. 3. Anger M, Friedhofer H, Fukutaki MF, et al. Primary cutaneous melanoma: an 18-year study. Clinics (Sao Paulo). 2010;65:257-263. 4. INCA. Estimativa 2012 Instituto Nacional de Cancer 2012. http://www.inca.gov.br/estimativa/2012/index.asp?id=5. Accessed August 22, 2014. 5. Chiba FB, Schettini AP, Delfino AC, et al. Clinical and epidemiological profile of cutaneous malignant melanomas in two referral institutions in the city of Manaus, Brazil. An Bras Dermatol. 2011;86:1239-1241. 6. Garbe C, Buttner P, Weiss J, et al. Risk factors for developing cutaneous melanoma and criteria for identifying persons at risk: multicenter case-control study of the Central Malignant Melanoma Registry of the German Dermatological Society. J Invest Dermatol. 1994;102:695-699. 7. Nijsten T, Leys C, Verbruggen K, et al. Case-control study to identify melanoma risk factors in the Belgian population: the significance of clinical examination. J Eur Acad Dermatol Venereol. 2005;19:332-339. 8. Thomas NE, Groben P. Invasive superficial spreading melanomas arising from clinically normal skin. J Am Acad Dermatol. 2004;51:466-470. 9. Togawa Y, Nakamura Y, Kamada N, et al. Melanoma in association with acquired melanocytic nevus in Japan: a review of cases in the literature. Int J Dermatol. 2010;49:1362-1367. 10. Tsao H, Bevona C, Goggins W, et al. The transformation rate of moles (melanocytic nevi) into cutaneous melanoma: a population-based estimate. Arch Dermatol. 2003;139:282-288. 11. Echeverria B, Botella-Estrada R, Serra-Guillen C, et al. Increased risk of developing a second primary cutaneous nevus-associated melanoma in patients previously diagnosed with the disease [in Spanish]. Actas Dermosifiliogr. 2010;101:710-716. 12. Weatherhead SC, Haniffa M, Lawrence CM. Melanomas arising from naevi and de novo melanomas does origin matter? Br J Dermatol. 2007;156:72-76. 13. Purdue MP, From L, Armstrong BK, et al. Etiologic and other factors predicting nevus-associated cutaneous malignant melanoma. Cancer Epidemiol Biomarkers Prev. 2005;14:2015-2022. 14. Kaddu S, Smolle J, Zenahlik P, et al. Melanoma with benign melanocytic naevus components: reappraisal of clinicopathological features and prognosis. Melanoma Res. 2002;12:271-278. 15. Crucioli V, Stilwell J. The histogenesis of malignant melanoma in relation to pre-existing pigmented lesions. J Cutan Pathol. 1982;9:396-404. 16. Skender-Kalnenas TM, English DR, Heenan PJ. Benign melanocytic lesions: risk markers or precursors of cutaneous melanoma? J Am Acad Dermatol. 1995;33:1000-1007. 17. Veronese LA, Corrente JE, Marques MEA. Melanoma maligno cutâneo: sistema de pontos (scoring system) para auxílio no diagnóstico histopatológico. J Bras Patol Med Lab. 2006;42:375-383. 18. Veronese LA, Corrente JE, Marques MEA. Critérios histopatológicos para diagnóstico de melanoma maligno cutâneo: análise comparativa de sua frequência em lesões benignas e melanomas de pequena espessura (<2mm). J Bras Patol Med Lab. 2007;43:363-368. 19. Barnhill R. Textbook of Dermatopathology. New York, NY: McGraw-Hill; 1998. 20. Stadler R, Garbe C. Nevus associated malignant melanomas diagnostic validation and prognosis [in German]. Hautarzt. 1991;42:424-429. 21. McNutt NS. Triggered trap : nevoid malignant melanoma. Semin Diagn Pathol. 1998;15:203-209. 22. McCullagh P, Nelder JA. Generalized Linear Models. New York, NY: Chapman & Hall; 1989. 490 Am J Clin Pathol 2014;142:485-491 Downloaded 490 from https://academic.oup.com/ajcp/article-abstract/142/4/485/1766613
23. Naser N. Cutaneous melanoma: a 30-year-long epidemiological study conducted in a city in southern Brazil, from 1980-2009. An Bras Dermatol. 2011;86:932-941. 24. Whiteman DC, Watt P, Purdie DM, et al. Melanocytic nevi, solar keratoses, and divergent pathways to cutaneous melanoma. J Natl Cancer Inst. 2003;95:806-812. 25. Cohen LM. Lentigo maligna and lentigo maligna melanoma. J Am Acad Dermatol. 1995;33:923-940. 26. Carli P, Massi D, Santucci M, et al. Cutaneous melanoma histologically associated with a nevus and melanoma de novo have a different profile of risk: results from a case-control study. J Am Acad Dermatol. 1999;40:549-557. 27. Massi D, Carli P, Franchi A, et al. Naevus-associated melanomas: cause or chance? Melanoma Res. 1999;9:85-91. 28. Whiteman DC, Parsons PG, Green AC. p53 expression and risk factors for cutaneous melanoma: a case-control study. Int J Cancer. 1998;77:843-848. 29. Gruber SB, Barnhill RL, Stenn KS, et al. Nevomelanocytic proliferations in association with cutaneous malignant melanoma: a multivariate analysis. J Am Acad Dermatol. 1989;21(pt 1):773-780. 30. Marks R, Dorevitch AP, Mason G. Do all melanomas come from moles? a study of the histological association between melanocytic naevi and melanoma. Australas J Dermatol. 1990;31:77-80. 31. Whiteman DC, Pavan WJ, Bastian BC. The melanomas: a synthesis of epidemiological, clinical, histopathological, genetic, and biological aspects, supporting distinct subtypes, causal pathways, and cells of origin. Pigment Cell Melanoma Res. 2011;24:879-897. 32. Greenwald HS, Friedman EB, Osman I. Superficial spreading and nodular melanoma are distinct biological entities: a challenge to the linear progression model. Melanoma Res. 2012;22:1-8. 33. Takata M, Murata H, Saida T. Molecular pathogenesis of malignant melanoma: a different perspective from the studies of melanocytic nevus and acral melanoma. Pigment Cell Melanoma Res. 2010;23:64-71. 34. Sagebiel RW. Melanocytic nevi in histologic association with primary cutaneous melanoma of superficial spreading and nodular types: effect of tumor thickness. J Invest Dermatol. 1993;100:322S-325S. 35. Garcia-Cruz A, Florez A, de la Torre-Fraga C, et al. Observational cross-sectional study comparing Breslow thickness of melanoma arising from naevi and melanoma de novo. Br J Dermatol. 2009;161:700-702. 36. Harley S, Walsh N. A new look at nevus-associated melanomas. Am J Dermatopathol. 1996;18:137-141. 37. Ackerman AB. What naevus is dysplastic, a syndrome and the commonest precursor of malignant melanoma? a riddle and an answer. Histopathology. 1988;13:241-256. 38. Consensus conference: precursors to malignant melanoma. JAMA. 1984;251:1864-1866. 39. Clark WH Jr, Elder DE, Guerry DT, et al. A study of tumor progression: the precursor lesions of superficial spreading and nodular melanoma. Hum Pathol. 1984;15:1147-1165. 40. Barnhill RL, Cerroni L, Cook M, et al. State of the art, nomenclature, and points of consensus and controversy concerning benign melanocytic lesions: outcome of an international workshop. Adv Anat Pathol. 2010;17:73-90. 41. Vollmer RT, Seigler HF. Using a continuous transformation of the Breslow thickness for prognosis in cutaneous melanoma. Am J Clin Pathol. 2001;115:205-212. 42. Barnhill RL, Fine JA, Roush GC, et al. Predicting fiveyear outcome for patients with cutaneous melanoma in a population-based study. Cancer. 1996;78:427-432. Am J Clin Pathol 2014;142:485-491 491 491 491