Advertisement

Genetic Alterations in Primary Acral Melanoma and Acral Melanocytic Nevus in Korea: Common Mutated Genes Show Distinct Cytomorphological Features

Open ArchivePublished:November 27, 2017DOI:https://doi.org/10.1016/j.jid.2017.11.017
      Acral melanoma occurring on the palms, soles, and nails is the most common subtype of cutaneous melanoma in Asians. Genetic alterations in acral melanoma and acral melanocytic nevus are not well known. We performed next-generation sequencing and evaluated the correlations between genetic information and the clinicopathologic characteristics from 85 Korean patients with acral melanocytic neoplasms. Of the 64 patients with acral melanoma, most had lesions at the T2 stage or higher, and the heel was the most common anatomical site of melanoma (n = 34 [53.1%]). The five most common mutations were BRAF (22 [34.4%]), NRAS (14, [21.9%]), NF1 (11, [17.2%]), GNAQ (12, [17.2%]), and KIT (7, [10.9%]). In the 21 acral melanocytic nevi, those five gene mutations were also common. Copy number variations were also frequently detected in 75% of acral melanomas and 47.6% of acral melanocytic nevi, and amplification was more common than deletion in both lesions. BRAF mutation was associated with round epithelioid cells and NRAS and NF1 mutations with bizarre cells. NF1 and GNAQ mutations showed elongated and spindle cells with prominent dendrites in acral melanomas. KIT mutations were common in amelanotic acral melanoma. This study suggests that common mutated genes are associated with distinct cytomorphological features in acral melanocytic lesions.

      Abbreviations:

      ALM (acral lentiginous melanoma), CNV (copy number variation), NGS (next-generation sequencing), NM (nodular melanoma), SSM (superficial spreading melanoma)

      Introduction

      Acral melanoma occurs on acral skin, such as the palms, soles, and nails, and is the most common cutaneous melanoma in Asians, Africans, and non-Caucasians, and its absolute incidence is similar among all populations (
      • Chang J.W.
      Acral melanoma: a unique disease in Asia.
      ,
      • Wang Y.
      • Zhao Y.
      • Ma S.
      Racial differences in six major subtypes of melanoma: descriptive epidemiology.
      ). Histopathologically, acral lentiginous melanoma (ALM) showing continuous lentiginous proliferation of atypical melanocytes along the dermal-epidermal junction is the predominant subtype (
      • Kuchelmeister C.
      • Schaumburg-Lever G.
      • Garbe C.
      Acral cutaneous melanoma in Caucasians: clinical features, histopathology and prognosis in 112 patients.
      ). Nodular melanoma (NM) and superficial spreading melanoma (SSM) are not common in acral melanoma (
      • Jung H.J.
      • Kweon S.S.
      • Lee J.B.
      • Lee S.C.
      • Yun S.J.
      A clinicopathologic analysis of 177 acral melanomas in Koreans: relevance of spreading pattern and physical stress.
      ). Acral melanocytic nevus, which sometimes can show atypical features, is important in the differential diagnosis of acral melanoma along with clinicopathological correlations (
      • Bravo Puccio F.
      • Chian C.
      Acral junctional nevus versus acral lentiginous melanoma in situ: a differential diagnosis that should be based on clinicopathologic correlation.
      ,
      • Kim N.H.
      • Choi Y.D.
      • Seon H.J.
      • Lee J.B.
      • Yun S.J.
      Anatomic mapping and clinicopathologic analysis of benign acral melanocytic neoplasms: a comparison between adults and children.
      ). Recently, somatic oncogenic mutations in cutaneous melanoma have emerged as both effective biomarkers and therapeutic targets. Next-generation sequencing (NGS) shows more comprehensive genomic mutations and copy number variations (CNVs) with higher sensitivity compared with non-NGS approaches (
      • Zhang T.
      • Dutton-Regester K.
      • Brown K.M.
      • Hayward N.K.
      The genomic landscape of cutaneous melanoma.
      ). Although there have been recent NGS studies of wide genomic landscapes in cutaneous melanoma, the number of acral melanoma cases is absent or limited (
      Cancer Genome Atlas Network
      Genomic classification of cutaneous melanoma.
      ,
      • Hayward N.K.
      • Wilmott J.S.
      • Waddell N.
      • Johansson P.A.
      • Field M.A.
      • Nones K.
      • et al.
      Whole-genome landscapes of major melanoma subtypes.
      ). Additionally, the exact correlations of genetic changes with clinical and histopathological features have not been characterized because of low mutation rates in acral melanomas (
      • Jin S.A.
      • Chun S.M.
      • Choi Y.D.
      • Kweon S.S.
      • Jung S.T.
      • Shim H.J.
      • et al.
      BRAF mutations and KIT aberrations and their clinicopathological correlation in 202 Korean melanomas.
      ). Moreover, genetic changes in acral melanocytic nevus have not been described.
      In this study, we used NGS to detect mutations and CNVs in 85 patients with primary acral melanoma and acral melanocytic nevus and compared the genetic changes between the two lesion types; we then analyzed the correlations of these mutations with the clinical and histopathological features. We collected acral melanomas with various clinical manifestations at deeper thicknesses with mixed cellularity to determine a wide range of genetic information.

      Results

      Clinicopathological characteristics

      Among the 64 patients with acral melanoma in Korea, the mean age ± standard deviation at diagnosis was 65.5 ± 13.9 years (see Supplementary Table S1 online). There were 26 men and 38 women, with a male to female ratio of 0.7:1.0. Among eight patients with subungual melanoma, six patients had melanoma on the fingernails and two on the toenails. The heel was the most common anatomical site for volar area, and ALM was the most common histopathological subtype, followed by NM and SSM. We also evaluated 21 patients with acral melanocytic nevus: 17 (81%) were women and 4 (19%) were men. The mean age ± standard deviation at diagnosis was 41.8 ± 14.9 years, and compound melanocytic nevus was the most common histopathological subtype. Of the 64 acral melanomas, ulceration was present in 38 (59.4%) patients, and most patients (81.3%) had T2 stage disease or greater, with a mean Breslow thickness of 5.1 mm. T staging was as follows: T0 = 6 patients (9.4%) (melanoma in situ), T1 = 6 patients (9.4%), T3 = 20 patients (31.3%), and T4 = 23 patients (35.9%).

      Mutations in acral melanoma

      Among 83 genes (see Supplementary Table S2 online), the most common mutations were found in BRAF (n = 22 [34.4%]), NRAS (14 [21.9%]), NF1 (11 [17.2%]), GNAQ (12 [18.6%]), and KIT (7 [10.9%]) (Table 1 and Figure 1). One KRAS and three GNAS mutations were also detected. There were 16 (25%) patients with pan-negative results who had none of the mentioned mutations. Most BRAF mutations were BRAF V600E (17 [26.6%]), and BRAF non-V600E mutations (D22N, D594G, G596R, and K601N) were detected in five patients (7.8%) (Figure 2). One KRAS mutation was also detected. An NRAS mutation in codon 61 was detected in three patients, and NRAS G12D, G13R, Y64D, and S173R mutations were also found. NF1 mutations at various gene positions and GNAQ M59L, T96S, and Y101 mutations were detected. KIT mutations were found mostly in exons 11, 13, and 17. The BRAF V600E and NRAS, and BRAF V600E and NF1, mutations were mutually exclusive. The BRAF V600E and KIT, and NRAS and KIT, mutations were also mutually exclusive. However, GNAQ mutations co-occurred with many other mutations: three BRAF V600E, five NF1, and two NRAS mutations. Thus, only one patient exhibited a GNAQ mutation alone. Among six patients of subungual melanoma on the fingernails, KIT, BRAF, and GNAQ mutations were detected in three patients, respectively, one patient had an NRAS mutation, and one had an NF1 mutation. KIT mutation was detected in one subungual melanoma on the toenail. The presence of other commonly mutated genes differed according to the major mutated genes. Patients with BRAF mutations frequently showed mutations in NOTCH1, RET, and MLH1. Patients with NF1 and GNAQ mutations showed mutations in PIK3R1, TP53, and CDH1. NRAS and KIT mutations were associated with fewer mutations in other genes. In patients with pan-negative results, other mutations were also common.
      Table 1Specific mutation types, anatomic sites according to the histopathologic subtypes and cytomorphologic features in acral melanoma
      Mutation (n)Histopathologic

      Subtype (n)
      Anatomical

      Site
      Nucleotide ChangeAmino Acid ChangeExonEffectnCo-mutated Genes (n)Cell Type (n)
      BRAF V600E (17)ALM (5)Palm (1),

      sole (4)
      c.1799T>AV600E15Non-synonymous5Round epithelioid (1)
      Spindle (1)
      GNAQ (1)Round epithelioid + spindle (2)
      GNAQ (1)Round epithelioid

      + elongated & spindle (1)
      NM (6)Sole (6)c.1799T>AV600E15Non-synonymous6Round epithelioid (4)

      Ovoid (1)

      Round epithelioid + spindle (1)
      SSM (6)Sole (6)c.1799T>AV600E15Non-synonymous6Round epithelioid (5)

      Round epithelioid + elongated (1)
      BRAF

      non-V600E (5)
      ALM (3)Finger nail (1)c.1415_1416insATCTGACAAGGTY471_K472

      insYGTVY
      11Non-synonymous1KIT (1)Ovoid + elongated (1)
      Sole (1)c.1803A>TK601N15Non-synonymous1NF1 (1)Round epithelioid

      + ovoid + elongated (1)
      Finger nail (1)c.1781A>GD594G15Non-synonymous1Spindle (1)
      NM (1)Finger nail (1)c.1786G>CG596R15Non-synonymous1NRAS (1)Round epithelioid + ovoid (1)
      SSM (1)Sole (1)c.64G>AD22N1Non-synonymous1KIT (1)Ovoid (1)
      NRAS (14)ALM (7)Sole (7)c.182A>GQ61R3Non-synonymous2Round epithelioid (1)
      GNAQ (1)Elongated & spindle (1)
      c.35G>AG12D2Non-synonymous2NF1(2)Round epithelioid + spindle (2)
      c.519C>AS173R3Non-synonymous1Spindle (1)
      c.37G>CG13R2Non-synonymous1GNAQ (1)Ovoid + elongated (1)
      c.181C>AQ61K3Non-synonymous1Spindle (1)
      NM (5)Sole (4)c.181C>AQ61K3Non-synonymous2Round epithelioid (2)
      c.37G>CG13R2Non-synonymous1NF1 (1)Round epithelioid

      + elongated & spindle (1)
      c.182A>GQ61R3Non-synonymous1Round epithelioid (1)
      Finger nail (1)c.190T>GY64D3Non-synonymous1BRAF non-V600E (1)Ovoid + elongated (1)
      SSM (2)Sole (2)c.183A>CQ61H3Non-synonymous1Round epithelioid (1)
      c.182A>GQ61R3Non-synonymous1Round epithelioid

      + ovoid + elongated (1)
      NF1 (11)ALM (9)Sole (9)c.316G>TE106
      premature stop codon.
      4Non-synonymous1BRAF non-V600E (1)Round epithelioid

      + ovoid + elongated (1)
      c.491T>GL164
      premature stop codon.
      5Non-synonymous1GNAQ (1)Ovoid + spindle (1)
      c.2507A>CE836A21Non-synonymous1Elongated & spindle (1)
      c.6292G>AA2098T42Non-synonymous1NRAS (1)Round epithelioid + spindle (1)
      29497016G-C5Splice site1GNAQ (1)Round epithelioid + spindle (1)
      29556994T-C22Splice site1Round epithelioid + spindle (1)
      29508721T-G6Splice site1GNAQ (1)Spindle (1)
      29585514GTCAAAGGTGAATTATTTTGATAATCTAGC-G32Splice site1NRAS (1)Round epithelioid + spindle (1)
      c.4590_4591insTGTTA1529fsX335Frame shift1GNAQ (1)Ovoid + elongated (1)
      NM (2)Sole (1)c.1312G>AE438K12Non-synonymous1NRAS (1)Round epithelioid

      + elongated & spindle (1)
      Finger nail (1)c.8407C>AP2803T58Non-synonymous1GNAQ (1)Elongated (1)
      GNAQ (12)ALM (10)Sole (10)c.175A>CM59L2Non-synonymous3Elongated & spindle (2)
      BRAF V600E (1)Round epithelioid

      + elongated & spindle (1)
      c.286A>TT96S2Non-synonymous3BRAF V600E (1)Round epithelioid + spindle (1)
      NF1 (1)Round epithelioid + spindle (1)
      NF1 (1)Spindle (1)
      c.303C>AY101
      premature stop codon.
      2Non-synonymous2NF1 (1)Ovoid + spindle (1)
      NF1 (1)Ovoid + elongated (1)
      c.175A>CM59L2Non-synonymous1KIT (1)Spindle (1)
      c.286A>TT96S2Non-synonymous
      c.286A>TT96S2Non-synonymous1GNAQ (1)Elongated (1)
      c.303C>AY101
      premature stop codon.
      2Non-synonymous
      NM (1)Finger nail (1)c.175A>CM59L2Non-synonymous1NF1 (1)Elongated (1)
      c.286A>TT96S2Non-synonymous
      c.303C>AY101
      premature stop codon.
      2Non-synonymous
      SSM (1)Sole (1)c.286A>TT96S2Non-synonymous1BRAF V600E (1)Round epithelioid + elongated (1)
      KIT (7)ALM (3)Finger nail (1)c.1727T>CR19H1Non-synonymous1BRAF non-V600E (1)Ovoid + elongated (1)
      c.56G>AV654A13Non-synonymous
      Finger nail (1)c.2467T>GV560D11Non-synonymous1Spindle (1)
      Sole (1)c.1679T>AK642E13Non-synonymous1GNAQ (1)Spindle (1)
      NM (3)Sole (1)c.1727T>CL576P11Non-synonymous1Ovoid (1)
      c.2449A>CI817L17Non-synonymous
      Finger nail (1)c.1961T>CY823D17Non-synonymous1Elongated (1)
      Toe nail (1)c.1679T>AK642E13Non-synonymous1Round epithelioid + elongated (1)
      SSM (1)Sole (1)c.1727T>CL576P11Non-synonymous1BRAF non-V600E (1)Ovoid (1)
      Pan-negative (16)ALM (11)Sole (9)Spindle (5), ovoid (2)

      Round epithelioid (1)

      Round epithelioid + spindle (1)

      Ovoid + elongated (1)

      Spindled + bizarre (1)
      Palm (2)
      NM (3)Sole (2)Ovoid (1)

      Round epithelioid + ovoid (1)
      Toe nail (1)Ovoid + elongated & spindle (1)
      SSM (2)Sole (2)Round epithelioid (2)
      Abbreviations: ALM, acral lentiginous melanoma; NM, nodular melanoma; SSM, superficial spreading melanoma.
      premature stop codon.
      Figure 1
      Figure 1Mutations in acral melanoma and acral melanocytic nevus. Common mutated genes are BRAF, NRAS, NF1, GNAQ, and KIT in acral melanomas and BRAF and GNAQ in acral melanocytic nevi.
      Figure 2
      Figure 2Mutation statuses of BRAF, RAS, NF1, GNAQ, GNAS, and KIT in acral melanoma. Most BRAF mutations are BRAF V600E, and five patients had BRAF non-V600E mutations on D22N, D594G, G596R, and K601N. NRAS mutations in codon 61 were common, and NRAS G12D, G13R, Y64D, and S173R mutations were also found. One KRAS mutation was detected. GNAQ M59L, T96S, and Y101 mutations were commonly detected. NF1 and KIT mutations occurred at various gene positions.

      Mutations in acral melanocytic nevus

      Common mutations in the 21 patients with acral melanocytic nevi included BRAF (14 [66.7%]), NRAS (2 [9.5%]), NF1 (2 [9.5%]), GNAQ (8 [38.1%]), and KIT (3 [14.3%]) (Figure 1). Only two (9.5%) patients had pan-negative results. BRAF V600E and NRAS mutations were mutually exclusive, but five patients had both BRAF and GNAQ mutations; one patient had BRAF, GNAQ, and NF1 mutations; and one patient had BRAF, GNAQ, and KIT mutations. One patient had both GNAQ and KIT mutations, and another patient had both NF1 and KIT mutations. There were some differences in mutated genes between acral melanoma and acral melanocytic nevus. BRAF V600E was more commonly observed in acral melanocytic nevus than in acral melanoma. However, all other mutations were much less commonly found in acral melanocytic nevus than in acral melanoma, except ALK, which was mutated in five acral nevi and one acral melanoma.

      CNVs in acral melanoma and acral melanocytic nevus

      CNVs were found in 48 (75%) of 64 acral melanoma patients and in 10 (47.6%) of 21 acral melanocytic nevi patients. Among 83 genes, 71 gene CNVs (85.5%) were detected in acral melanoma and 56 gene CNVs (67.5%) in acral melanocytic nevi. Gene amplifications and deletions were found in 44 (68.8%) and 22 patients (34.4%) with acral melanomas, respectively, and in 10 (47.6%) and 3 patients (14.3%) with acral melanocytic nevi, respectively (Figure 3). The most common gene amplifications in acral melanoma occurred in SRC (23 [40%]), GNAS (17 [26.6%]), NTRK1 (17 [26.6%]), FGFR3 (12 [18.8%]), NOTCH1 (11 [17.2%]), ARID1B (10 [15.6%]), and EPHB4 (10 [15.6%]). The most common gene deletions in acral melanoma were seen in CDKN2A (6 [9.4%]) and SMAD4 (5 [7.8%]). NRAS and KIT mutations showed fewer CNVs, except SRC. In acral melanocytic nevus, the most common gene amplifications were seen in FGFR3 (6 [28.6%]), ARID1B (6 [28.6%]), and CNKN2A (6 [28.6%]).
      Figure 3
      Figure 3Copy number variations in acral melanoma and acral melanocytic nevus. Copy number variations were found in 48 of 64 acral melanoma patients and in 10 of 21 acral melanocytic nevi patients. Gene amplifications were more common than deletions in both acral melanomas and acral melanocytic nevi. The most common gene amplifications in acral melanoma occurred in SRC, GNAS, NTRK1, FGFR3, NOTCH1, ARID1B, and EPHB4. The most common gene deletions in acral melanoma were seen in CDKN2A and SMAD4. In acral melanocytic nevus, the most common gene amplifications were seen in FGFR3, ARID1B, and CNKN2A.

      Correlations of the clinicopathological features with BRAF, NRAS, NF1, GNAQ, and KIT mutations

      We evaluated the correlations of five mutations with clinicopathological features. Anatomical mapping results showed no anatomical site difference according to mutation type (Figure 4a). ALM, observed in 35 patients, showed more NF1 mutations (9 [26%]), GNAQ mutations (10 [29%]), and pan-negative cases (11 [31%]) compared with NM and SSM. BRAF (7 [39%]) and NRAS (5 [28%]) mutations were common in NM, which was observed in 18 patients. In the 11 patients with SSM, BRAF mutations (7 [64%]) were common. The 17 patients with BRAF V600E differed significantly from the 47 patients with wild-type BRAF in terms of age (younger), histopathological subtype (non-ALM), Breslow thickness (thinner), and mitotic rate (lower) (see Supplementary Table S3 online). Patients with BRAF V600E also had characteristic architectural features, including abrupt lateral circumscription, high-grade pagetoid scatter, and nesting (see Supplementary Table S4 online). Other genes showed no significant differences between the mutated and wild-type groups. Cytological morphology differed according to the gene mutation, in that BRAF V600E mutations were associated with round epithelioid cells, BRAF V600E mutations with ovoid cells, NRAS and NF1 mutations with bizarre cells, GNAQ and NF1 mutations with elongated and spindle cells, and NF1 mutations with prominent dendrites (Table 2 and Figure 4a, and see Supplementary Table S5 online). Desmoplastic melanoma was found in two patients and exhibited NF1 mutations. Of 14 patients who were diagnosed with amelanotic acral melanoma, 3 (21.4%) had KIT mutations. KIT-mutated and pan-negative melanomas were not correlated with any specific cytomorphological pattern. Patients with common mutations were correlated with higher T stages. Patients with T2 stage or higher had significantly more mutations than did patients with T1 stage or lower (P = 0.007). Patients with mutations in at least two genes showed mixed cellularity, with at least two cell types (P ≤ 0.001). Thus, acral melanomas with deeper Breslow thickness and mixed cellularity were associated with more mutations (see Supplementary Table S6 online). In acral melanocytic nevus, GNAQ mutations were correlated with prominent dendrites (P = 0.014) (Figure 4b).
      Figure 4
      Figure 4Anatomical mapping of acral melanoma on the sole of the foot and mutations correlated with cytological features. (a) Anatomical mapping results show no anatomical site difference according to mutation type. Acral lentiginous melanoma with pan-negativity (scale bar = 200 μm) and acral melanoma with round epithelioid cells and high pagetoid scatter exhibiting BRAFV600E mutations (scale bar = 100 μm). Bizarre cells in NRAS mutations (scale bar = 100 μm), desmoplastic melanoma with NF1 mutations (scale bar = 100 μm), acral melanoma comprising spindle cells with prominent dendrites with NF1 and GNAQ mutations (scale bar = 50 μm), and amelanotic acral melanoma with KIT mutations (scale bar = 100 μm). (b) Clinical, dermoscopic, and histopathological features of acral melanocytic nevus according to gene mutation. Nevoid cells in BRAF- and NRAS-mutated nevus (scale bar = 50 μm) and spindle cells with prominent dendrites in GNAQ-mutated nevus (scale bar = 100 μm).
      Table 2Cytomorphological features of the 64 acral melanomas according to the BRAF, NRAS, NF1, GNAQ, and KIT mutation status, n (%)
      Morphological CharacteristicsBRAFV600E Wild Type (n = 47)BRAFV600E Mutation (n = 17)P-Value
      Fisher exact test was used.
      ,
      P < 0.05 was considered statistically significant.
      BRAF Non-V600E Wild Type (n = 59)BRAF Non-V600E

      Mutation (n = 5)
      P-ValueNRAS Wild Type (n = 50)NRAS Mutation (n = 14)P-Value
      Fisher exact test was used.
      ,
      P < 0.05 was considered statistically significant.
      Epithelioid<0.0010.3550.365
       Absent30 (63.8)2 (11.8)28 (47.5)4 (80.0)27 (54.0)5 (35.7)
       Present17 (36.2)15 (88.2)31 (52.5)1 (20.0)23 (46.0)9 (64.3)
      Ovoid0.0480.012>0.999
       Absent32 (68.1)16 (94.1)47 (79.7)1 (20.0)37 (74.0)11 (78.6)
       Present15 (31.9)1 (5.9)12 (20.3)4 (80.0)13 (26.0)3 (21.4)
      Elongated + spindle0.0190.640>0.999
       Absent14 (29.8)11 (64.7)24 (40.7)1 (20.0)20 (40.0)5 (35.7)
       Present33 (70.2)6 (35.3)35 (59.3)4 (80.0)30 (60.0)9 (64.3)
      Bizarre0.083>0.9990.003
       Absent27 (57.4)14 (82.3)38 (64.4)3 (60.0)37 (74.0)4 (28.6)
       Present20 (42.6)3 (17.7)21 (35.6)2 (40.0)13 (26.0)10 (71.4)
      Prominent dendrite<0.0010.0770.534
       Not prominent13 (27.7)13 (76.5)24 (43.3)0 (0)21 (42.0)5 (35.7)
       Prominent34 (72.3)4 (23.5)35 (56.7)5 (100)29 (58.0)9 (64.3)
      Morphological CharacteristicsNF1

      Wild Type (n = 53)
      NF1

      Mutation (n = 11)
      P-ValueGNAQ

      Wild Type (n = 52)
      GNAQ

      Mutation (n = 12)
      P-ValueKIT

      Wild Type (n = 57)
      KIT

      Mutation (n = 7)
      P-Value
      Epithelioid>0.9990.3370.104
       Absent26 (49.1)6 (54.6)24 (46.1)8 (66.7)26 (45.6)6 (85.7)
       Present27 (50.9)5 (45.4)28 (53.9)4 (33.3)31 (54.4)1 (14.3)
      Ovoid>0.999>0.9990.353
       Absent40 (75.5)8 (72.7)39 (75.0)9 (75.0)44 (77.2)4 (57.1)
       Present13 (24.5)3 (27.3)13 (25.0)3 (25.0)13 (22.8)3 (42.9)
      Elongated + spindle0.0040.0020.695
       Absent25 (47.2)0 (0)25 (48.1)0 (0)23 (40.4)2 (28.6)
       Present28 (52.8)11 (100.0)27 (51.9)12 (100.0)34 (59.6)5 (71.4)
      Bizarre0.0460.742>0.999
       Absent37 (69.8)4 (36.4)34 (65.4)7 (58.3)36 (63.2)5 (71.4)
       Present16 (30.2)7 (63.6)18 (34.6)5 (41.7)21 (36.8)2 (28.6)
      Prominent dendrite0.0020.1730.691
       Not prominent26 (49.1)0 (0.0)25 (48.1)2 (18.2)24 (42.1)2 (28.6)
       Prominent27 (50.9)11 (100.0)27 (51.9)10 (81.8)33 (57.9)5 (71.4)
      1 Fisher exact test was used.
      2 P < 0.05 was considered statistically significant.

      Discussion

      We identified genetic alterations in acral melanoma and acral melanocytic nevus and focused only on primary acral melanocytic lesions. In this study, we showed marked associations between several mutations and cytomorphological characteristics.
      We found that BRAF, NRAS, NF1, GNAQ, and KIT are major mutated genes in acral melanocytic lesions, with BRAF V600E mutation being the most commonly detected mutation. The weight-bearing portion of the sole is the most common anatomical site of acral melanoma (
      • Jung H.J.
      • Kweon S.S.
      • Lee J.B.
      • Lee S.C.
      • Yun S.J.
      A clinicopathologic analysis of 177 acral melanomas in Koreans: relevance of spreading pattern and physical stress.
      ,
      • Minagawa A.
      • Omodaka T.
      • Okuyama R.
      Melanomas and mechanical stress points on the plantar surface of the foot.
      ), but there were no differences in gene mutations with respect to anatomical site. Thus, mutation was not found to be associated with a specific anatomical site on the plantar surface. Our study showed a much higher frequency of BRAF V600E mutations in acral melanoma than was found in previous Asian studies (
      • Jin S.A.
      • Chun S.M.
      • Choi Y.D.
      • Kweon S.S.
      • Jung S.T.
      • Shim H.J.
      • et al.
      BRAF mutations and KIT aberrations and their clinicopathological correlation in 202 Korean melanomas.
      ,
      • Sakaizawa K.
      • Ashida A.
      • Uchiyama A.
      • Ito T.
      • Fujisawa Y.
      • Ogata D.
      • et al.
      Clinical characteristics associated with BRAF, NRAS and KIT mutations in Japanese melanoma patients.
      ,
      • Shim J.H.
      • Shin H.T.
      • Park J.
      • Park J.H.
      • Lee J.H.
      • Yang J.M.
      • et al.
      Mutational profiling of acral melanomas in Korean populations.
      ,
      • Si L.
      • Kong Y.
      • Xu X.
      • Flaherty K.T.
      • Sheng S.
      • Cui C.
      • et al.
      Prevalence of BRAF V600E mutation in Chinese melanoma patients: large scale analysis of BRAF and NRAS mutations in a 432-case cohort.
      ). These differences might be explained by the improvement in detection tools, by the selection of deeper melanomas showing variable cytology and mixed cellularity of various morphologies, and by the genes included in the test panels. In acral melanocytic nevus, the BRAF V600E mutation was the most common mutation, which suggests that BRAF mutations are major events in melanocytic nevus, even in this sun-protected area (
      • Takata M.
      • Saida T.
      Genetic alterations in melanocytic tumors.
      ). Studies of Caucasians have shown that BRAF mutations are associated with younger age, SSM, thinner tumor thickness, and lower mitotic rates (
      • Bauer J.
      • Buttner P.
      • Murali R.
      • Okamoto I.
      • Kolaitis N.A.
      • Landi M.T.
      • et al.
      BRAF mutations in cutaneous melanoma are independently associated with age, anatomic site of the primary tumor, and the degree of solar elastosis at the primary tumor site.
      ,
      • Liu W.
      • Kelly J.W.
      • Trivett M.
      • Murray W.K.
      • Dowling J.P.
      • Wolfe R.
      • et al.
      Distinct clinical and pathological features are associated with the BRAF(T1799A(V600E)) mutation in primary melanoma.
      ). We also observed similar features in BRAF-mutated acral melanomas. NRAS showed a higher frequency than in previous Asian acral melanoma reports (
      • Sakaizawa K.
      • Ashida A.
      • Uchiyama A.
      • Ito T.
      • Fujisawa Y.
      • Ogata D.
      • et al.
      Clinical characteristics associated with BRAF, NRAS and KIT mutations in Japanese melanoma patients.
      ,
      • Si L.
      • Kong Y.
      • Xu X.
      • Flaherty K.T.
      • Sheng S.
      • Cui C.
      • et al.
      Prevalence of BRAF V600E mutation in Chinese melanoma patients: large scale analysis of BRAF and NRAS mutations in a 432-case cohort.
      ). NRAS mutations are associated with NM (
      • Lee J.H.
      • Choi J.W.
      • Kim Y.S.
      Frequencies of BRAF and NRAS mutations are different in histological types and sites of origin of cutaneous melanoma: a meta-analysis.
      ) and giant congenital melanocytic nevus (
      • Charbel C.
      • Fontaine R.H.
      • Malouf G.G.
      • Picard A.
      • Kadlub N.
      • El-Murr N.
      • et al.
      NRAS mutation is the sole recurrent somatic mutation in large congenital melanocytic nevi.
      ), and in this study they were detected in five patients with NM and two patients with acral melanocytic nevus. NF1 mutations are detected in 14–17% of all melanomas and are associated with the highest mutation burden and strongest UV signature (
      Cancer Genome Atlas Network
      Genomic classification of cutaneous melanoma.
      ,
      • Cirenajwis H.
      • Lauss M.
      • Ekedahl H.
      • Torngren T.
      • Kvist A.
      • Saal L.H.
      • et al.
      NF1-mutated melanoma tumors harbor distinct clinical and biological characteristics.
      ,
      • Hayward N.K.
      • Wilmott J.S.
      • Waddell N.
      • Johansson P.A.
      • Field M.A.
      • Nones K.
      • et al.
      Whole-genome landscapes of major melanoma subtypes.
      ).
      • Rawson R.V.
      • Johansson P.A.
      • Hayward N.K.
      • Waddell N.
      • Patch A.M.
      • Lo S.
      • et al.
      Unexpected UVR and non-UVR mutation burden in some acral and cutaneous melanomas.
      reported that acral melanomas with high UV signatures occur on subungual sites and harbor BRAF or NF1 mutations. In our study, of 11 patients with NF1 mutations, only one had a left thumbnail subungual melanoma. Thus, our results did not show the association between NF1 mutation and UV signature. NF1 mutations are common in desmoplastic melanoma (
      • Wiesner T.
      • Kiuru M.
      • Scott S.N.
      • Arcila M.
      • Halpern A.C.
      • Hollmann T.
      • et al.
      NF1 mutations are common in desmoplastic melanoma.
      ), and in our study, there were two desmoplastic melanoma cases in whom NF1 mutation was detected. GNAQ mutations were detected in both acral melanoma and acral melanocytic nevus. GNAQ mutations are associated with blue nevus, uveal melanoma, melanoma arising from blue nevus, and melanocytoma (
      • Perez-Alea M.
      • Vivancos A.
      • Caratu G.
      • Matito J.
      • Ferrer B.
      • Hernandez-Losa J.
      • et al.
      Genetic profile of GNAQ-mutated blue melanocytic neoplasms reveals mutations in genes linked to genomic instability and the PI3K pathway.
      ,
      • Whiteman D.C.
      • Pavan W.J.
      • Bastian B.C.
      The melanomas: a synthesis of epidemiological, clinical, histopathological, genetic, and biological aspects, supporting distinct subtypes, causal pathways, and cells of origin.
      ), but they have also been detected in mucosal melanomas (
      • Sheng X.
      • Kong Y.
      • Li Y.
      • Zhang Q.
      • Si L.
      • Cui C.
      • et al.
      GNAQ and GNA11 mutations occur in 9.5% of mucosal melanoma and are associated with poor prognosis.
      ). Most reported GNAQ mutations were detected in codon 209, but our study showed mutations in codons 59, 96, and 101, which we were able to detect because we performed whole GNAQ gene analysis, not just an analysis of hotspot sites. Many GNAQ mutations were found to co-occur with other mutations, and only one patient with acral melanoma had a GNAQ mutation exclusively. GNAQ mutations were most commonly found together with NF1 mutations, suggesting that they may share common or complementary signaling pathways. KIT mutations were detected in 10.9% of acral melanoma patients, and there was a higher frequency of KIT mutations among amelanotic acral melanomas (
      • Choi Y.D.
      • Chun S.M.
      • Jin S.A.
      • Lee J.B.
      • Yun S.J.
      Amelanotic acral melanomas: clinicopathological, BRAF mutation, and KIT aberration analyses.
      ). Pan-negativity was observed in 25% of acral melanoma and 9.5% of acral melanocytic nevus patients, which was lower than we expected.
      CNVs differed between acral melanoma and acral melanocytic nevus. Genes, including CCND1, are frequently amplified from the early stages of acral melanomas (
      • Bastian B.C.
      The molecular pathology of melanoma: an integrated taxonomy of melanocytic neoplasia.
      ), and our study also showed high frequencies of CNVs in acral melanomas and acral melanocytic nevi. A limitation of this study was that we could not detect CCND1 CNV, because it was not included in the targeted gene panel available to us. In acral melanoma, the most commonly amplified gene was SRC, and the most common gene deletions occurred in CDKN2A and SMAD4. In acral melanocytic nevus, commonly amplified genes were FGFR3, ARID1B, and CNKN2A. Notably, SRC family kinases, which are non-receptor tyrosine kinases, and receptor tyrosine kinases (RTKs) play an important role in melanoma signaling pathways (
      • Homsi J.
      • Cubitt C.L.
      • Zhang S.
      • Munster P.N.
      • Yu H.
      • Sullivan D.M.
      • et al.
      Src activation in melanoma and Src inhibitors as therapeutic agents in melanoma.
      ,
      • Easty D.J.
      • Gray S.G.
      • O’Byrne K.J.
      • O’Donnell D.
      • Bennett D.C.
      Receptor tyrosine kinases and their activation in melanoma.
      ). NTRK1 was second most commonly amplified gene in acral melanoma. NTRK1 is an RTK protein for the nerve growth factor (
      • Miranda C.
      • Mazzoni M.
      • Sensi M.
      • Pierotti M.A.
      • Greco A.
      Functional characterization of NTRK1 mutations identified in melanoma.
      ). GNAS, which is a G-protein–coupled receptor protein, was both mutated and amplified in acral melanoma (
      • Shoushtari A.N.
      • Carvajal R.D.
      GNAQ and GNA11 mutations in uveal melanoma.
      ). Thus, RTKs, SRCs, and G-protein–coupled receptor proteins are closely related to these lesions, and their co-activation involves signaling through all major RTK pathways (
      • Easty D.J.
      • Gray S.G.
      • O’Byrne K.J.
      • O’Donnell D.
      • Bennett D.C.
      Receptor tyrosine kinases and their activation in melanoma.
      ). Patients harboring NRAS and KIT mutations showed fewer co-mutations and CNVs. Pan-negative tumors had more frequent amplification of RTKs, suggesting that these tumors could be treated with RTK inhibitors (
      Cancer Genome Atlas Network
      Genomic classification of cutaneous melanoma.
      ).
      • Hayward N.K.
      • Wilmott J.S.
      • Waddell N.
      • Johansson P.A.
      • Field M.A.
      • Nones K.
      • et al.
      Whole-genome landscapes of major melanoma subtypes.
      reported that CDKN2A is a significantly mutated gene in cutaneous melanoma but not in acral melanoma. In this study, no CDKN2A mutation was detected in acral melanomas; however, there was a marked difference in CDKN2A CNV bewteen acral melanoma and acral melanocytic nevus. Deletions were common in acral melanoma, whereas amplification was common in acral melanocytic nevus. Thus, CDKN2A, which is related to cell cycle progression, may be a major CNV in acral melanocytic lesions.
      This study provides substantial information on the histopathological features and mutation status of acral melanomas. ALM was frequently associated with GNAQ and NF1 mutations and pan-negativity, whereas NM and SSM were associated with BRAF and NRAS mutations. The invasive vertical growth phase of acral melanomas may show mixed cytological features of round epithelioid, ovoid, elongated, spindle, and bizarre giant cells. To our knowledge, this is the largest NGS analysis of deeper acral melanomas with cellular heterogeneity, showing a correlation with various genetic changes. Round epithelioid cells with many scattered pagetoid cells and nests were observed in BRAF-mutated acral melanomas. NRAS and NF1 mutations were strongly correlated with bizarre giant cells, GNAQ and NF1 mutations were correlated with elongated and spindle cells, and NF1 mutations were correlated with prominent dendrites. GNAQ-mutated acral melanocytic nevi showed prominent dendrites. Our study suggests that these genes are driver mutations for clonal expansion of specific cytomorphological melanocytes during the early tumor stages, and other genetic alterations are involved in melanoma development. Our findings are also supported by reports that BRAF mutations in melanocytic nevus are fully clonal, and BRAF mutation is an early initiating event in melanocytic neoplasia (
      • Damsky W.E.
      • Bosenberg M.
      Melanocytic nevi and melanoma: unraveling a complex relationship.
      ,
      • Yeh I.
      • von Deimling A.
      • Bastian B.C.
      Clonal BRAF mutations in melanocytic nevi and initiating role of BRAF in melanocytic neoplasia.
      ). Our study suggests that BRAF, NRAS, NF1, GNAQ, and KIT mutations occur initially in acral melanocytic nevus, after which additional mutations and genetic alterations are required for acral melanoma development. Differences in the frequencies of other mutated genes were evident between acral melanoma and acral melanocytic nevus. Mutations and CNVs were much less common in acral melanocytic nevus than in acral melanoma. No mutations in TP53, PTEN, or RB1 have been previously reported in acral or mucosal melanomas (
      • Hayward N.K.
      • Wilmott J.S.
      • Waddell N.
      • Johansson P.A.
      • Field M.A.
      • Nones K.
      • et al.
      Whole-genome landscapes of major melanoma subtypes.
      ); however, these mutations were observed in acral melanomas in this study.
      In conclusion, this study characterized the association between cytomorphological features and several gene mutations in acral melanoma and acral melanocytic nevus, which has not been previously reported to our knowledge. Further studies are needed to more precisely determine the genetic changes and gene-associated histopathologic and pathogenic mechanisms in acral melanocytic lesions.

      Materials and Methods

      Patients

      A total of 85 patients in Korea, 64 with primary acral melanoma and 21 with acral melanocytic nevus, were enrolled from the Department of Dermatology at Chonnam National University Hospital in Gwangju and Chonnam National University Hwasun Hospital in Hwasun, South Korea, from 2004 through 2016. The study protocol was approved by the institutional review board of the hospital, and written informed patient consent was obtained.

      Clinicopathological analysis

      We reviewed the clinical characteristics of all 85 patients, including age, sex, anatomical site of the lesion, and presence of internal malignancies. We performed anatomical mapping of the plantar melanoma as described in a previous study (
      • Jung H.J.
      • Kweon S.S.
      • Lee J.B.
      • Lee S.C.
      • Yun S.J.
      A clinicopathologic analysis of 177 acral melanomas in Koreans: relevance of spreading pattern and physical stress.
      ). We also reviewed all melanoma hematoxylin and eosin slides to analyze the histopathological subtype, ulceration, Breslow thickness, and mitotic rate. The histopathologic subtypes included ALM, SSM, NM, and desmoplastic melanoma. We diagnosed those in which black to brownish pigmentation was absent as amelanotic acral melanoma. The histopathologic differentiation between acral melanocytic nevi and early acral melanomas is sometimes very difficult; thus, we included only cases of acral melanocytic nevi with stringent criteria including typical clinical, dermoscopic, and histopathologic features showing dermal nevoid cells with maturation.

      Diagnostic criteria of cytological features

      We divided histopathological findings into architectural and cytological features, as described by
      • Viros A.
      • Fridlyand J.
      • Bauer J.
      • Lasithiotakis K.
      • Garbe C.
      • Pinkel D.
      • et al.
      Improving melanoma classification by integrating genetic and morphologic features.
      and subsequently validated (
      • Broekaert S.M.
      • Roy R.
      • Okamoto I.
      • van den Oord J.
      • Bauer J.
      • Garbe C.
      • et al.
      Genetic and morphologic features for melanoma classification.
      ). Briefly, the architectural features analyzed were the epidermal contour, lateral circumscription, pagetoid scatter, and nest formation. Tumor cell size was graded as small, medium, and large. Cell morphological features were described as round epithelioid, ovoid, elongated, and spindle cells and the presence of bizarre cells and prominent dendrites. Round epithelioid cells were defined as large cells with approximately equal long and short diameters, ovoid cells as those with a long diameter approximately one third longer than the short diameter, elongated cells as those with a long diameter one third to twofold longer than the short diameter, and spindle cells as those with a long diameter longer than two times short diameter (see Supplementary Figure S1 online). In this analysis, we considered elongated and spindle cells as one group, because differentiation between these cell types is sometimes difficult, and they are often intermingled together. Acral melanomas are frequently composed of a combination of various cell types. Therefore, we classified the cell types in an individual tumor and determined the proportions of each cell type. The major cell types, defined as those composing more than 10% of the total cell population, were included for the statistical analysis. Bizarre giant cells was defined as giant cells with atypical large nuclei, including anaplastic cells, monster-like cells, and multinucleated cells. Bizarre giant cells were simply described as present or absent, because they were scattered throughout the tumor and interspersed with the dominant cell types. To identify epidermal melanocytes with prominent dendrites, immunohistochemical staining of melan-A was performed in 42 acral melanoma patients in whom the prominent dendrites could not be determined by hematoxylin and eosin staining (see Supplementary Figure S2 online). In the 21 patients with acral melanocytic nevi, we analyzed cell morphological features and correlated these features with the genetic mutations.

      Next-generation sequencing

      In total, 85 tissue specimens were macrodissected from 8- to 10-μm–thick unstained archival formalin-fixed paraffin-embedded sections. Tumor presence was verified by hematoxylin and eosin staining. Areas containing viable tumors were marked on the slides. Compared with nontumor tissue components, the dissected areas contained at least 50% tumor nuclei. Paraffin was removed by xylene treatment, and DNA was purified using the QIAamp DNA FFPE Tissue Kit (Qiagen, Hilden, Germany). A targeted panel was used to capture 83 whole genes, including all coding exons, for the detection of single nucleotide variants, insertions/deletions, and CNVs (see Supplementary Table S2).
      Genomic DNA (200–500 ng) was prepared to construct libraries using the SureSelect gene target panel (Agilent, Santa Clara, CA), according to the manufacturer’s protocol. Briefly, the genomic DNA sample was randomly fragmented by ultrasonication (Covaris, Woburn, MA), followed by adapter ligation, purification, hybridization, and PCR. Captured libraries were assessed for quality using the Agilent 2100 Bioanalyzer and were loaded onto the Illumina HiSeq 2500 (TheragenEtex Bio Institute, Suwon, Korea) according to the manufacturer’s recommendations. Raw image files were processed by HCS1.4.8 (Illumina) for base-calling using the default parameters, and the sequences of each sample were generated as 101-base pair paired-end reads (see Supplementary Materials and Methods online). Single nucleotide variants with a variant allele frequency of 3% or greater and insertion/deletion frequency of 10% or greater were ultimately selected. CNVs were analyzed using the depth of coverage for each target region between tumor and preprocessed normal data. The copy number cutoff values of 7 or greater and 0 were used for amplification and homo-deletion, respectively.

      Statistical analysis

      Statistical analyses were performed using the SPSS version 22.0 (SPSS Inc., Chicago, IL). For normally distributed data, we used t tests and analysis of variance to determine significant differences between group means. The distributions of epidemiological, clinical, and histopathological data were compared using Fisher exact test. The correlations between the mutation type and clinicopathological features were also analyzed using Fisher exact t test. A P-value less than 0.05 was considered statistically significant.

      Conflict of Interest

      The authors state no conflict of interest.

      Acknowledgments

      This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education ( NRF-2016R1D1A3B03930554 ) and by a grant ( HCRI17903-21 ) from the Chonnam National University Hwasun Hospital Biomedical Research Institute .

      Supplementary Material

      References

        • Bastian B.C.
        The molecular pathology of melanoma: an integrated taxonomy of melanocytic neoplasia.
        Annu Rev Pathol. 2014; 9: 239-271
        • Bauer J.
        • Buttner P.
        • Murali R.
        • Okamoto I.
        • Kolaitis N.A.
        • Landi M.T.
        • et al.
        BRAF mutations in cutaneous melanoma are independently associated with age, anatomic site of the primary tumor, and the degree of solar elastosis at the primary tumor site.
        Pigment Cell Melanoma Res. 2011; 24: 345-351
        • Bravo Puccio F.
        • Chian C.
        Acral junctional nevus versus acral lentiginous melanoma in situ: a differential diagnosis that should be based on clinicopathologic correlation.
        Arch Pathol Lab Med. 2011; 135: 847-852
        • Broekaert S.M.
        • Roy R.
        • Okamoto I.
        • van den Oord J.
        • Bauer J.
        • Garbe C.
        • et al.
        Genetic and morphologic features for melanoma classification.
        Pigment Cell Melanoma Res. 2010; 23: 763-770
        • Cancer Genome Atlas Network
        Genomic classification of cutaneous melanoma.
        Cell. 2015; 161: 1681-1696
        • Chang J.W.
        Acral melanoma: a unique disease in Asia.
        JAMA Dermatol. 2013; 149: 1272-1273
        • Charbel C.
        • Fontaine R.H.
        • Malouf G.G.
        • Picard A.
        • Kadlub N.
        • El-Murr N.
        • et al.
        NRAS mutation is the sole recurrent somatic mutation in large congenital melanocytic nevi.
        J Invest Dermatol. 2014; 134: 1067-1074
        • Choi Y.D.
        • Chun S.M.
        • Jin S.A.
        • Lee J.B.
        • Yun S.J.
        Amelanotic acral melanomas: clinicopathological, BRAF mutation, and KIT aberration analyses.
        J Am Acad Dermatol. 2013; 69: 700-707
        • Cirenajwis H.
        • Lauss M.
        • Ekedahl H.
        • Torngren T.
        • Kvist A.
        • Saal L.H.
        • et al.
        NF1-mutated melanoma tumors harbor distinct clinical and biological characteristics.
        Mol Oncol. 2017; 11: 438-451
        • Damsky W.E.
        • Bosenberg M.
        Melanocytic nevi and melanoma: unraveling a complex relationship.
        Oncogene. 2017; 36: 5771-5792
        • Easty D.J.
        • Gray S.G.
        • O’Byrne K.J.
        • O’Donnell D.
        • Bennett D.C.
        Receptor tyrosine kinases and their activation in melanoma.
        Pigment Cell Melanoma Res. 2011; 24: 446-461
        • Hayward N.K.
        • Wilmott J.S.
        • Waddell N.
        • Johansson P.A.
        • Field M.A.
        • Nones K.
        • et al.
        Whole-genome landscapes of major melanoma subtypes.
        Nature. 2017; 545: 175-180
        • Homsi J.
        • Cubitt C.L.
        • Zhang S.
        • Munster P.N.
        • Yu H.
        • Sullivan D.M.
        • et al.
        Src activation in melanoma and Src inhibitors as therapeutic agents in melanoma.
        Melanoma Res. 2009; 19: 167-175
        • Jin S.A.
        • Chun S.M.
        • Choi Y.D.
        • Kweon S.S.
        • Jung S.T.
        • Shim H.J.
        • et al.
        BRAF mutations and KIT aberrations and their clinicopathological correlation in 202 Korean melanomas.
        J Invest Dermatol. 2013; 133: 579-582
        • Jung H.J.
        • Kweon S.S.
        • Lee J.B.
        • Lee S.C.
        • Yun S.J.
        A clinicopathologic analysis of 177 acral melanomas in Koreans: relevance of spreading pattern and physical stress.
        JAMA Dermatol. 2013; 149: 1281-1288
        • Kim N.H.
        • Choi Y.D.
        • Seon H.J.
        • Lee J.B.
        • Yun S.J.
        Anatomic mapping and clinicopathologic analysis of benign acral melanocytic neoplasms: a comparison between adults and children.
        J Am Acad Dermatol. 2017; 77: 735-745
        • Kuchelmeister C.
        • Schaumburg-Lever G.
        • Garbe C.
        Acral cutaneous melanoma in Caucasians: clinical features, histopathology and prognosis in 112 patients.
        Br J Dermatol. 2000; 143: 275-280
        • Lee J.H.
        • Choi J.W.
        • Kim Y.S.
        Frequencies of BRAF and NRAS mutations are different in histological types and sites of origin of cutaneous melanoma: a meta-analysis.
        Br J Dermatol. 2011; 164: 776-784
        • Liu W.
        • Kelly J.W.
        • Trivett M.
        • Murray W.K.
        • Dowling J.P.
        • Wolfe R.
        • et al.
        Distinct clinical and pathological features are associated with the BRAF(T1799A(V600E)) mutation in primary melanoma.
        J Invest Dermatol. 2007; 127: 900-905
        • Minagawa A.
        • Omodaka T.
        • Okuyama R.
        Melanomas and mechanical stress points on the plantar surface of the foot.
        N Engl J Med. 2016; 374: 2404-2406
        • Miranda C.
        • Mazzoni M.
        • Sensi M.
        • Pierotti M.A.
        • Greco A.
        Functional characterization of NTRK1 mutations identified in melanoma.
        Genes Chromosomes Cancer. 2014; 53: 875-880
        • Perez-Alea M.
        • Vivancos A.
        • Caratu G.
        • Matito J.
        • Ferrer B.
        • Hernandez-Losa J.
        • et al.
        Genetic profile of GNAQ-mutated blue melanocytic neoplasms reveals mutations in genes linked to genomic instability and the PI3K pathway.
        Oncotarget. 2016; 7: 28086-28095
        • Rawson R.V.
        • Johansson P.A.
        • Hayward N.K.
        • Waddell N.
        • Patch A.M.
        • Lo S.
        • et al.
        Unexpected UVR and non-UVR mutation burden in some acral and cutaneous melanomas.
        Lab Invest. 2017; 97: 130-145
        • Sakaizawa K.
        • Ashida A.
        • Uchiyama A.
        • Ito T.
        • Fujisawa Y.
        • Ogata D.
        • et al.
        Clinical characteristics associated with BRAF, NRAS and KIT mutations in Japanese melanoma patients.
        J Dermatol Sci. 2015; 80: 33-37
        • Sheng X.
        • Kong Y.
        • Li Y.
        • Zhang Q.
        • Si L.
        • Cui C.
        • et al.
        GNAQ and GNA11 mutations occur in 9.5% of mucosal melanoma and are associated with poor prognosis.
        Eur J Cancer. 2016; 65: 156-163
        • Shim J.H.
        • Shin H.T.
        • Park J.
        • Park J.H.
        • Lee J.H.
        • Yang J.M.
        • et al.
        Mutational profiling of acral melanomas in Korean populations.
        Exp Dermatol. 2017; 26: 883-888
        • Shoushtari A.N.
        • Carvajal R.D.
        GNAQ and GNA11 mutations in uveal melanoma.
        Melanoma Res. 2014; 24: 525-534
        • Si L.
        • Kong Y.
        • Xu X.
        • Flaherty K.T.
        • Sheng S.
        • Cui C.
        • et al.
        Prevalence of BRAF V600E mutation in Chinese melanoma patients: large scale analysis of BRAF and NRAS mutations in a 432-case cohort.
        Eur J Cancer. 2012; 48: 94-100
        • Takata M.
        • Saida T.
        Genetic alterations in melanocytic tumors.
        J Dermatol Sci. 2006; 43: 1-10
        • Viros A.
        • Fridlyand J.
        • Bauer J.
        • Lasithiotakis K.
        • Garbe C.
        • Pinkel D.
        • et al.
        Improving melanoma classification by integrating genetic and morphologic features.
        PLoS Med. 2008; 5: e120
        • Wang Y.
        • Zhao Y.
        • Ma S.
        Racial differences in six major subtypes of melanoma: descriptive epidemiology.
        BMC Cancer. 2016; 16: 691
        • Whiteman D.C.
        • Pavan W.J.
        • Bastian B.C.
        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
        • Wiesner T.
        • Kiuru M.
        • Scott S.N.
        • Arcila M.
        • Halpern A.C.
        • Hollmann T.
        • et al.
        NF1 mutations are common in desmoplastic melanoma.
        Am J Surg Pathol. 2015; 39: 1357-1362
        • Yeh I.
        • von Deimling A.
        • Bastian B.C.
        Clonal BRAF mutations in melanocytic nevi and initiating role of BRAF in melanocytic neoplasia.
        J Natl Cancer Inst. 2013; 105: 917-919
        • Zhang T.
        • Dutton-Regester K.
        • Brown K.M.
        • Hayward N.K.
        The genomic landscape of cutaneous melanoma.
        Pigment Cell Melanoma Res. 2016; 29: 266-283