Research Article

Head & Neck Spindle Cell Melanoma

Cui Ye^1,4, Feiluore Yibulaying³, Lei Feng^2,4, Zhi-cheng Yang^2,4* and Alimujiang Wushou^2,4*
1Department of Orthodontics, Shanghai Stomatological Hospital, Fudan University, China
²Department of Oral and Maxillofacial Surgery, Shanghai Stomatological Hospital, Fudan University, China
³Department of Preventive Medicine, School of Public Health, Fudan University, China
⁴Oral Biomedical Engineering Laboratory, Shanghai Stomatological Hospital, Fudan University, China

^*Corresponding author: Alimujiang Wushou, Department of Oral and Maxillofacial Surgery & Oral Biomedical Engineering Laboratory, Shanghai Stomatological Hospital, Fudan University, China
Zhi-cheng Yang, Department of Oral and Maxillofacial Surgery & Oral Biomedical Engineering Laboratory, Shanghai Stomatological Hospital, Fudan University, China

Published: 22 Oct, 2018
Cite this article as: Ye C, Yibulaying F, Feng L, Yang Z-C, Wushou A. Head & Neck Spindle Cell Melanoma. Clin Surg. 2018; 3: 2167.

Abstract

Purpose: Head Neck Spindle Cell Melanoma (HNSCM) is very rare morphologic subtype of melanoma and unexplored comprehensively. The aim of this study was to investigate HNSCM’s incidence, general demographics, basic clinico-pathologic features, and treatment outcome and disease specific prognostic factors.
Material and Methods: HNSCM cases were identified in Surveillance, Epidemiology, and End Results database (1973-2018).
Results: A total of 2500 HNSCM cases with a median age of 66 years were found in the database. The Male-to-female ratio is 2:1. Statistically significant Overall Survival (OS) and Disease Specific Survival (DSS) differences were found were depending on age, gender, race, age, tumor location, SEER historic stage, T stage, N stage, M stage and pathological differentiation (p<0.05). In the multivariate Cox regression analysis, age >66 years [Hazard ratio (HR) (95%, Confidence Interval, CI) =2.785 (2.224-3.488), p=0.000, for OS; HR (95% CI) =2.140 (1.535-2.983), p=0.000, for DSS, age ≤ 66 years as reference], unmarried status [HR (95% CI) =1.254 (1.031-1.525), p=0.023, for OS, married as reference], T3+T4 stage [HR (95% CI) =1.330 (1.080-1.639), p=0.007, for OS, HR (95% CI) =1.577 (1.106-2.247), p=0.012, for DSS, TX+T1+T2, as reference] and M1 stage [HR (95% CI) =2.658 (1.385-5.101), p=0.003, for OS, HR (95% CI) =6.894 (2.122-22.39), p=0.003, for DSS, M0 as reference] were associated with worse DSS and OS. Besides, the non-skin tumor [HR (95% CI) =0.511 (0.366-0.714), p=0.000, for OS, HR (95% CI) =0.603 (0.366-0.994), for DSS, p=0.047 skin tumor as reference] was associated with favorable DSS.
Conclusion: HNSCM mostly occurred in white people at 60~80 years old with predominance in males. The patient’s age, tumor location, T stage, M stage and marital status were independent prognostic factors for DSS and OS.
Keywords: Spindle cell melanoma; Head neck; SEER analysis

Introduction

Spindle cell melanoma (SCM) is a morphologic variant of melanoma [1]. SCM is a diagnostic challenge for pathologists, because there are many lesions which can be spindle cell rich and SCM lack of classic conventional melanoma characteristic [2]. Histologically, SCM is often confused with other (e.g. mesenchymal or neural) tumors. SCM often leads misdiagnosis and immunohistochemistry is helpful tool in distinguishing SCM from other sarcoma and carcinoma. Specific immunophenotypes is decisive in diagnosis to differentiate SCM from desmoplastic melanoma [2,3].
Melanoma is a common malignancy; however, SCM is very rare subtype [4]. SCM is frequently amelanotic, can occur anywhere on the body [5,6]. The clinical presentation of SCM is non-specific, and the early diagnosis is often delayed. Plus, it has an aggressive biological behavior and presents typically with widespread metastatic disease and ended up worse treatment outcome [7]. Since the majority of cases were found at the advanced stage of tumor, the prognosis is poor despite the availability of common therapeutic option like surgical therapy, immunotherapy, chemotherapy or radiotherapy [8].
To our knowledge, except several sporadic case reports and small incomplete retrospective case series, most of the previous studies have been published on the SCM diagnosis and differential diagnostic viewpoint [9-12]. Due to the rarity of the SCM, especially the head neck region, the limited study population of previous studies, there are no studies available, which have focused on the incidence, clinicopathologic features, treatment outcome and disease independent prognostic factors. Thus, in order to give audiences a concise spectrum of Head Neck SCM (HNSCM), a retrospective clinical case series was performed by using data from registered in the Surveillance, Epidemiology, and End Results (SEER).

Figure 1

×Close

This text will be overwritten by jQuery

Figure 1
The age and gender distribution of head and neck spindle cell melanoma.

Materials and Methods

Data source
The SEER program is comprised of 18 population-based cancer registries around the United States. Records from SEER registries are publicly available and are managed by the National Cancer Institute and it currently represents approximately 30% of the United States overall population [13]. The data extraction and statistical analysis was performed as described previously [14,15]. Briefly, International Classification of Diseases for Oncology codes 8772/3 were used for data extraction from SEER database with official software SEER*Stat, version 8.3.4.
Statistical analysis
Statistical analysis was carried out by using software of the Statistical Package for Social Sciences, Version 23.0, for Windows (SPSS, Chicago, IL). The chi square test or Fisher exact test was used for categorical variables comparison. The survival curves were developed by using the Kaplan-Meier method, and the log-rank test was performed to evaluate the survival difference. Adjusted Hazard Ratios (HRs) along with 95% Confidence Intervals (CIs) were calculated by using the Cox proportional hazards regression model. When the p value was <0.05, the difference was regarded as statistically significant. All statistical tests were two tailed.

Results

General demographic and clinicopathologic characteristics of study cohort: A total of 2500 patients, consisted of 1704 males and 796 females with a median age of 66 years were found in the SEER database from 1973 to 2017. A peak incidence occurred during sixth to eighth decade of life. The age and gender distribution is shown in (Figure 1). White people account for 97.72% (2443/2500) of study population. The overall median follow-up period was 75.2 months (range, 0-482 months). Nearly 70% of HNSCM originated from HN skin. The basic demographic and clinicopathologic characteristics of the whole patients are summarized in (Table 1).
Overall survival (OS) and cox proportional hazards regression analysis: Kaplan-Meier analysis was applied for time-to-event analysis. Statistically significant OS differences were found were depending on race (p=0.027), age (p=0.000), SEER stage (p=0.000), tumor location (p=0.000), marital status (p=0.005), gender (p=0.000), T stage (p=0.000), N stage (p=0.000), M stage (p=0.000). These results are presented in (Figure 2). The Cox proportional hazards regression model was applied for univariate and multivariate survival analysis. Age, race, gender, marital status, SEER stage, tumor location, pathological grade, T stage, M stage, N stage were associated with OS in the univariate Cox regression analysis. Details are shown in Table 2. In the multivariate Cox regression analysis, age >66 years [Hazard ratio (HR) (95% confidence interval, CI) =2.785 (2.224-3.448), p=0.000, age ≤ 66 years as reference], parameter with unmarried status [HR (95% CI) =1.254 (1.031-1.525), p=0.023, married as reference], T3+T4 stage [HR (95% CI) =1.330 (1.080-1.639), p=0.007, TX+T1+T2 as reference] and M1 stage [HR (95% CI) =2.658 (1.385- 5.101), p=0.003, M0 as reference] were independently associated with worse OS. These details are shown in (Table 3).
The demographic and clinicopathologic features with Disease- Specific Survival (DSS) status: For Disease Specific Survival (DSS) analysis, there were 1393 patients (489 females and 904 males) obtainable in the total study cohort. The median age was 66.2 years and 36.8% of patients with age ≤ 66.2 years. The median follow-up time for DSS was 78.6 months (range, 0-482 months). The early AJCC stage (stage I+II) cases were 555 and advanced stage (stage III+IV) patients were 180. Nearly 80% of patients (1110/1393) were treated by surgery. Summary statistics of DSS status are presented in the (Table 1).
DSS and cox proportional hazards regression analysis: In the survival analysis for DSS, significant survival differences were also identified depending on race (p=0.004), age (p=0.000), SEER stage (p=0.000), tumor location (p=0.000), pathological differentiation (p=0.003), T stage (p=0.000), N stage (p=0.000), M stage (p=0.000) (Figure 3). In the univariate Cox regression analysis, gender, age, SEER stage, race, tumor location, T stage, N stage, M stage was associated with DSS (Table 2). In the multivariate Cox regression analysis, age >66 years [HR (95% CI) =2.140 (1.535-2.983), p=0.000, age ≤ 66 years as reference], SEER historic stage of regional tumor [HR (95% CI) =1.699 (1.076-2.684), p=0.023, localized tumor as reference], T3+T4 stage [HR (95% CI) =1.577 (1.106-2.247), p=0.012, TX+T1+T2 as reference] and M1 stage [HR (95% CI) =6.894 (2.122-22.391), p=0.003, M0 as reference] were adversely associated with DSS. Nonskin tumor [HR (95% CI) =0.603 (0.366-0.994), p=0.047, skin tumor as reference] was favorable associated with better DSS. Details of the univariate and multivariate Cox regression analysis are presented in the (Table 2, 3).

Figure 2

×Close

This text will be overwritten by jQuery

Figure 2
Overall survival curves of cases with head and neck spindle cell melanoma compared according to (A) age, (B) gender, (C) race, (D) marital status, (E) SEER stage, (F) and (G) tumor location, (H) T stage, (I) N stage and (J) M stage. Log-rank test was utilized to compare curves, and significance is presented on each panel.

Table 1

×

Clinicopathologic parameter		Overall survival				Disease specific survival
		Alive	Dead	Total	p-value	Alive	Dead	Total	p-value
Age	≤ 66 years	686	368	1054	0	612	207	813	0
	> 66 years	552	894	1446		350	224	574
Race	White	1201	1242	2443	0.019	926	427	1353	0.004
	Others	37	20	57		36	4	40
Sex	Male	823	881	1704	0.074	611	293	904	0.106
	Female	415	381	796		351	138	489
Marital  status	Married	746	739	1485	0.387	549	265	814	0.122
	Others	492	523	1015		413	166	579
SEER stage	Localized	851	663	1514	0	684	181	865	0
	Regional	274	332	606		196	131	327
	Distant	57	165	222		37	86	123
	Unknown	56	102	158		45	33	78
Tumor location	Eye & orbit	433	294	727	0	383	118	501	0
	Nose & mouse	20	29	49		19	20	39
	Skin	785	939	1724		560	293	835
T stage	TX	91	89	180	0	62	26	88	0
	T0	28	38	66		20	24	44
	T1	218	80	298		166	21	187
	T2	195	86	281		148	20	168
	T3	174	97	271		119	34	153
	T4	182	160	342		124	55	179
N stage	NO	776	412	1188	0	561	126	687	0
	N1	34	39	73		24	17	41
	N2	20	18	38		13	5	18
	N3	6	9	15		4	3	7
	NX	53	72	125		38	29	67
M stage	M0	840	444	1284	0	604	129	733	0
	M1	39	82	121		27	49	76
	MX	10	24	34		9	2	11
AJCC stage	Stage I	282	96	378	0	206	20	226	0
	Stage II	367	219	586		256	73	329
	Stage III	103	54	157		80	21	101
	Stage IV	41	84	125		29	50	79
Pathological grade	Grade I	30	7	37	0	26	3	29	0.003
	Grade II	19	6	25		15	1	16
	Grade III	12	22	34		9	10	19
	Grade IV	10	13	23		9	7	16
	Unknown	1167	1214	2381		903	410	1313
Treatment	Surgery	999	1041	2040	0.247	772	338	1110	0.433
	Non- surgery	239	221	460		190	93	283

Table 1: The summary of head neck HNSCM patients’ clinico-pathologic characteristics.

Table 1
The summary of head neck HNSCM patients’ clinico-pathologic characteristics.

Table 2

×

Parameters		Overall survival		Disease specific survival
		HR (95% CI)	p -value	HR (95% CI)	p -value
Age	≤ 66 years	1.00 Reference	0	1.00 Reference	0
	> 66 years	3.424(3.009-3.896)		2.281(1.881-2.765)
Race	White	1.00 Reference	0.029	1.00 Reference	0.024
	Others	0.611(0.393-0.951)		0.322(0.120-0.863)
Sex	Female	1.00 Reference	0	1.00 Reference	0.017
	Male	1.340(1.188-1.512)		1.280(1.045-1.567)
Marital status	Married	1.00 Reference	0.005	1.00 Reference	0.808
	Others	1.174(1.050-1.314)		0.976(0.804-1.185)
SEER stage	Localized	1.00 Reference		1.00 Reference
	Regional	1.821(1.593-2.081)	0	2.563(2.042-3.216)	0
	Distant	4.529(3.804-5.392)	0	9.084(6.987-11.812)	0
Tumor location	Skin	1.00 Reference		1.00 Reference
	Eye & orbit	3.242(2.209-4.759)	0	3.779(2.436-6.088)	0
	Nose & mouth	2.322(2.033-2.653)	0	2.099(1.692-2.603)	0
T stage	T0	1.00 Reference		1.00 Reference
	T1	0.255(0.173-0.376)	0	0.109(0.061-0.196)	0
	T2	0.329(0.225-0.483)	0	0.119(0.066-0.215)	0
	T3	0.458(0.315-0.667)	0	0.267(0.158-0.450)	0
	T4	0.691(0.485-0.985)	0.041	0.400(0.247-0.646)	0
	TX	0.708(0.484-1.035)	0.075	0.417(0.239-0.727)	0.002
N stage	NO	1.00 Reference		1.00 Reference
	N1	1.943(1.399-2.699)	0	3.090(1.861-5.130)	0
	N2	1.877(1.171-3.011)	0.009	2.340(0.956-5.726)	0.063
	N3	2.881(1.487-5.579)	0.002	2.373(0.755-70457)	0.139
	NX	2.608(2.028-3.353)	0	3.489(2.327-5.232)	0
M stage	M0	1.00 Reference		1.00 Reference
	M1	4.310(3.397-5.468)	0	8.112(5.807-11.331)	0
	MX	2.217(1.470-3.343)	0	0.847(0.209-3.424)	0.816
AJCC Stage	Stage I	1.00 Reference		1.00 Reference
	Stage II	1.892(1.487-2.406)	0	2.914(1.776-4.780)	0
	Stage III	1.778(1.274-2.483)	0.001	3.058(1.657-5.644)	0
	Stage IV	7.186(5.342-9.668)	0	18.683(11.061-31.557)	0
Pathological grade	Grade I	1.00 Reference		1.00 Reference
	Grade II	0.811(0.271-2.424)	0.708	0.369(0.038-3.569)	0.389
	Grade III	3.429(1.462-8.041)	0.005	4.753(1.307-17.282)	0.018
	Grade IV	3.414(1.358-8.582)	0.009	5.082(1.310-19.706)	0.019

Table 2: Univariate Cox regression analysis of characteristics associated with OS and DSS.

Table 2
Univariate Cox regression analysis of characteristics associated with OS and DSS.

Discussion

The incidence of HN melanoma is 22% to 34% of all melanoma [16-18]. As a morphologic variant of melanoma, the incidence of HNSCM is rare and no detailed data are available [4]. To our knowledge, this is the first study about the incidence and survival analysis of HNSCM. Current results demonstrated nearly 70% of HNSCM occurred in the head neck skin with peak incidence in the sixth to eighth decade of life and predilection of males.
Regarding to general parameters in the survival analysis, remarkable survival differences were identified in age, gender and race regarding OS and DSS (Figures 2A-2C, and 3A-3C). This in accordance with previous studies on melanoma [19,20]. The explanations behind prognosis may be men have less hair on the scalp and more UV exposure outdoors than women, and white people are more sensitive compared with colored people. Besides, significant differences were shown in marital status in OS (Figure 2D), which may indicate that the spouse’s care and support are good for physical and mental recovery.
Because of unspecific hidden clinical presentation of HNSCM, early detection and diagnosis are always delayed. Thus, overall prognosis of HNSCM is poor [7,8]. Supporting these, advanced stage of HNSCM had the worse outcome than early stage tumor in our analysis (Figure 3D). In SEER stage survival analysis; there were also remarkable differences on OS and DSS (Figure 2E, 2G, 3D) as AJCC stage. Tumor locations had a great effect on the OS and DSS. In our study, we found that HNSCM derived from skin had a significantly worse outcome than non-skin tumor. This may be due to the greater metastases tendency [8]. Tumors located in the nose & mouth and skin obviously had poor OS and DSS comparing in the eye and orbit. The results are in accordance with previous studies [21,22]. This is likely because superficial and bare tumor sites like eyes & orbit had an earlier diagnosis and treatment than deep-seated unexposed tumor sites.
HNSCM are frequently found at the advanced stage and it is a challenge to the head and neck oncologist to correctly evaluate this aggressive tumor and choose the best suitable treatment protocol. In the current investigation, surgery is only treatment modality we can obtain the SEER database. However, surgery types are unavailable, even types of neck dissections were not recorded too. Owing to the complex anatomic structure of the head and neck region, radical surgery with clear margins for HNSCM is difficult, especially tumor at the advanced stage. Because, extensive excision in the head neck region requires wide-range of surgical reconstruction. No controlled surgical trials or multiple treatments have been carried out for HNSCM [23]. Such a circumstance, how to best plan a surgery based combined treatment protocol for different stage of HNSCM with currently available therapies such as neoadjuvant chemotherapy, immunotherapy or radiotherapy is difficult and challenging question to answer. Therefore, multi-cancer randomized clinical trials are urgently needed.
As a retrospective analysis, a few limitations of this study should be noticed. Firstly, not all of the cases have complete information, thus, the strength of the study is impaired. Secondly, the limited number of patients did not enable us to perform more specific subgroup comparison analysis for different tumor location or other important clinicopathological parameters. Thirdly, besides surgical treatment, other treatment modalities were lacking. Those missing treatment modalities should not be neglected, because they may have a huge impact for patient’s survival. Lastly, the follow-up time is uneven and not long enough.

Table 3

×

Parameters		Overall survival		Disease specific survival
		HR (95% CI)	p-value	HR (95% CI)	p-value
Age	≤ 66 years	1.00 Reference	0	1.00 Reference	0
	> 66 years	2.785(2.224-3.488)		2.140(1.535-2.983)
Marital  status	Married	1.00 Reference	0.023	1.00 Reference	0.972
	Others	1.254(1.031-1.525)		0.994(0.717-1.379)
SEER stage	Localized	1.00 Reference		1.00 Reference
	Regional	1.264(0.979-1.633)	0.073	1.699(1.076-2.684)	0.023
	Distant	2.006(1.185-3.396)	0.01	1.759(0.603-4.992)	0.289
Tumor  location	Skin	1.00 Reference	0	1.00 Reference	0.047
	Non-skin	0.511(0.366-0.714)		0.603(0.366-0.994)
T stage	TX+T1+T2	1.00 Reference	0.007	1.00 Reference	0.012
	T3+T4	1.330(1.080-1.639)		1.577(1.106-2.247)
M stage	M0	1.00 Reference	0.003	1.00 Reference	0.001
	M1	2.658(1.385-5.101)		6.894(2.122-22.39)

Table 3: Multivariate Cox regression analysis of characteristics associated with DSS and OS.

Table 3
Multivariate Cox regression analysis of characteristics associated with DSS and OS.

Figure 3

×Close

This text will be overwritten by jQuery

Figure 3
Disease specific survival curves of cases with head and neck spindle cell melanoma compared according to (A) age, (B) gender, (C) race, (D) SEER stage, (E) and (F) tumor location, (G) AJCC stage, (H) T stage, (I) N stage and (J) M stage. Log-rank test was utilized to compare curves, and significance is presented on each panel.

Conclusion

Within the limitations of this study, we report a first largest case series about HNSCM’s demographic trends, clinico-pathologic features, outcome and disease specific independent prognostic factors. Our results demonstrate that HNSCM mostly occurred in white people with predominance in the male and reached a peak incidence during the sixth to eighth decade of life. The patient’s age, tumor location, T stage, M stage, marital status and SEER stages were independent prognostic factors for DSS and OS.

Funding

This study was supported by The Scientific Research Foundation of Shanghai Stomatological Hospital, Fudan University (SSDCZ-2016-01) and Shanghai Natural Science Foundation (17ZR1424900).

References

1. Longacre TA, Egbert BM, Rouse RV. Desmoplastic and spindle-cell malignant melanoma: an immunohistochemical study. Am J Surg Pathol. 1996;20:1489-500.
2. Weissinger S, Keil P, Silvers D, Klaus BM, Möller P, Horst BA, et al. A diagnostic algorithm to distinguish desmoplastic from spindle cell melanoma. Mod Pathol. 2014;27(4):524-34.
3. Livestro DP, Muzikansky A, Kaine EM, Flotte TJ, Sober AJ, Mihm MC Jr. Biology of desmoplastic melanoma: a case-control comparison with other melanomas. J Clin Oncol. 2005;23(27):6739-46.
4. Sundersingh S, Majhi U, Narayanaswamy K, Balasubramanian S. Primary spindle cell melanoma of the urinary bladder. Indian J Pathol Microbiol. 2011;54(2):422-4.
5. Sheff J, Pane T. Spindle cell melanoma arising from decades-old burn scar. Plast Reconstr Surg. 2009;124(5).
6. Dainichi T, Kobayashi C, Fujita S, Shiramizu K, Ishiko T, Kiryu H, et al. Interdigital amelanotic spindle-cell melanoma mimicking an inflammatory process due to dermatophytosis. J Dermatol. 2007;34(10):716-9.
7. Piao Y, Guo M, Gong Y. Diagnostic challenges of metastatic spindle cell melanoma on fine‐needle aspiration specimens. Cancer. 2008;114(2):94-101.
8. Conley J, Lattes R, Orr W. Desmoplastic malignant melanoma (a rare variant of spindle cell melanoma). Cancer. 1971;28(4):914-36.
9. Prieto V, Woodruff J. Expression of HMB45 antigen in spindle cell melanoma. J Cutan Pathol. 1997;24(9):580-1.
10. Prieto V, Woodruff J. Expression of basement membrane antigens in spindle cell melanoma. J Cutan Pathol. 1998;25(6):297-300.
11. Kobayashi G, Cobb C. A case of amelanotic spindle-cell melanoma presenting as metastases to breast and axillary lymph node: diagnosis by FNA cytology. Diagn Cytopathol. 2000;22(4):246-9.
12. Mitamura Y, Ito T, Nakano-Nakamura M, Uchi H, Furue M. S100A6 and c-Kit-Positive Spindle Cell Melanoma of the Dorsal Foot. Case Rep Dermatol. 2014;6(2):140-4.
13. Cronin KA, Ries LA, Edwards BK. The Surveillance, Epidemiology, and End Results (SEER) Program of the National Cancer Institute. Cancer. 2014;120 Suppl 23:3755.
14. Feng L, Cai D, Muhetaer A, Yin-Long Y, Fei R, Mumingjiang Y, et al. Spindle cell carcinoma: the general demographics, basic clinico-pathologic characteristics, treatment, outcome and prognostic factors. Oncotarget. 2017;8(26):43228-36.
15. Wushou A, Jiang YZ, Liu YR, Shao ZM. The demographic features, clinicopathologic characteristics, treatment outcome and disease-specific prognostic factors of solitary fibrous tumor: a population-based analysis. Oncotarget. 2015;6(39):41875-83.
16. Koto M, Demizu Y, Saitoh JI, Suefuji H, Tsuji H, Okimoto T, et al. Multicenter Study of Carbon-Ion Radiation Therapy for Mucosal Melanoma of the Head and Neck: Subanalysis of the Japan Carbon-Ion Radiation Oncology Study Group (J-CROS) Study (1402 HN). Int J Radiat Oncol Biol Phys. 2017;97(5):1054-60.
17. Kılıç S, Unsal AA, Chung SY, Samarrai R, Kılıç SS, Baredes S, et al. Geographic region: Does it matter in cutaneous melanoma of the head and neck? Laryngoscope. 2017;127(12):2763-9.
18. Baykal C, Atci T, Polat Ekinci A, Buyukbabani N. An update on cutaneous melanoma in Turkey: evaluation of 19‐year data in a single tertiary centre and review of the literature. J Eur Acad Dermatol Venereol. 2017;31:236-40.
19. Helsing P, Robsahm TE, Vos L, Rizvi SMH, Akslen LA, Veierød MB. Cutaneous head and neck melanoma (CHNM): A population-based study of the prognostic impact of tumor location. J Am Acad Dermatol. 2016;75(5):975-2.
20. Sample A, He YY. Mechanisms and prevention of UV‐induced melanoma. Photodermatol Photoimmunol Photomed. 2018;34(1):13-24.
21. Khan F, Strohl A, Allen PD, Doerr TD. Desmoplastic Melanoma of the Head and Neck: Incidence and Survival, 1992-2013. Otolaryngol Head Neck Surg. 2017;157(4):648-56.
22. Paolino G, Cardone M, Didona D, Moliterni E, Losco L, Corsetti P, et al. Prognostic factors in head and neck melanoma according to facial aesthetic units. G Ital Dermatol Venereol. 2017.
23. Gladfelter P, Darwish NH, Mousa SA. Current status and future direction in the management of malignant melanoma. Melanoma Res. 2017;27:403-10.