INTRODUCTION: The eighth edition of American Joint Committee on Cancer guidelines separated cutaneous squamous cell carcinoma in the head and neck (cHNSCCs) into its own entity within head and neck malignancies with the aim of improving tumor risk stratification. The lack of precise prognostic estimates for cHNSCCs prevents clear guidance on the clinical approach to cHNSCCs resulting in heterogeneity of management.
OBJECTIVE: Detailed review of the historical management of cHNSCCs in a single tertiary skin cancer center with the aim of creating a treatment algorithm that provides guidance in selecting the appropriate workup and treatment while minimizing unnecessary treatment and resulting morbidity.
METHODS: Retrospective review of our institutional pathology database from January 2007 to December 2019 of all recorded head and neck cutaneous lesions excised by St Helens & Knowsley Teaching Hospitals NHS Foundation Trust Plastic Surgery Department. Primary outcomes include recurrence rates and cHNSCC-related mortality.
RESULTS: A total of 442 lesions of cHNSCCs from 216 patients excised from the head and neck region were excised during the study period and 164 cHNSCCs had deep margin involvement ± perineural invasion. Median age was 78 years (49–98 years). Median follow-up was 23 months (0–153 months), 9 patients (4%) were lost to follow-up. Seven patients (3%) died before their first 3-monthly follow-up appointment was due. Thirty-nine patients (18%) were immunosuppressed, 10 were solid organ transplant recipients. One hundred thirteen patients (52%) were deceased at the time of review, 25 (22%) died directly or from complications of cHNSCCs. Out of 164 cHNSCCS, majority had deep margin involvement (77%, n = 126), 21% (n = 35) had both deep margin and perineural involvement. Only 5 patients (3%) had cHNSCCs with features of perineural invasion but adequate margins. Forty-five patients (27.4%) in the positive cohort (deep margin involvement ± perineural invasion) went on to develop local, regional, or distant recurrence with a median disease-free interval of 7.7 months (0.5–58 months). Overall recurrence rate was 14.5% (n = 64/442) with a false-negative rate of 6.8% (n = 19/278), The scalp was the most common primary tumor site to recur (45%), followed by the ear (27%) and face (23%). Fifty percent of primary tumors that recurred were moderately differentiated compared with 19% which were poorly differentiated.
CONCLUSIONS: Our study demonstrates increased recurrence risk of both scalp and ear cHNSCCs, which warrant calls for more aggressive surgical measures, that is, primary excision with a wide margin down to periosteum with flap reconstruction with adjuvant radiotherapy. Routine sentinel lymph node biopsy in high-risk cHNSCCs could play a role in locoregional control.