Cutaneous head and neck melanoma (CHNM): A population-based study of the prognostic impact of tumor location

详细信息    查看全文

• 作者：Per Helsing ; MD^a ; ^{phelsing@ous-hf.no" class="auth_mail" title="E-mail the corresponding author} ; Trude Eid Robsahm ; PhD^b ; Linda Vos ; PhD^b ; Syed Mohammad Husain Rizvi ; MD^a ; Lars Andreas Akslen ; MD ; PhD^c ; ^d ; Marit Bragelien Veierø ; d ; PhD^e
• 关键词：head and neck melanoma ; mortality ; population-based ; prognostic factors ; scalp/neck location ; visibility
• 刊名：Journal of the American Academy of Dermatology
• 出版年：2016
• 出版时间：November 2016
• 年：2016
• 卷：75
• 期：5
• 页码：975-982.e2
• 全文大小：375 K

文摘

Most studies of cutaneous head and neck melanomas (CHNM) have reported poorer survival in CHNM compared with other sites, especially on the scalp/neck.

Objective

We sought to compare patient and tumor characteristics between CHNM and cutaneous trunk and extremity melanomas and between CHNM locations (face/ear vs scalp/neck, anterior vs posterior), and to study prognostic factors in patients with CHNM.

Methods

We studied all CHNM (n = 1074) from 8120 cases of cutaneous melanomas diagnosed in Norway in 2008 to 2012.

Results

Compared with cutaneous trunk and extremity melanomas, CHNM were more frequently found in men, more often nodular and lentigo maligna cutaneous melanomas, and diagnosed at higher T stage (P ≤ .01). CHNM located on posterior sites were diagnosed at significantly higher T stage, and were significantly more often diagnosed with ulceration and at more advanced stage compared with CHNM located on anterior sites (P < .001). T stage and clinical stage were the only significant prognostic factors for melanoma-specific and overall death in the multivariable analysis (P < .001).

Limitations

Low number of cases and the relatively high frequency of missing values are limitations.

Conclusion

More advanced CHNM were diagnosed on posterior compared with anterior locations, but location was not a significant prognostic factor for cutaneous melanoma–specific or overall death in the multivariable models.