2018-19 Section 7-Neoplastic and Inflammatory Diseases of the Head and Neck eBook

Otolaryngology–Head and Neck Surgery 156(5)

Figure 2. Survival outcomes: cohort comparisons. PNI, perineural invasion.

had a very low proportion of patients who were immuno- compromised (3.3%) despite the strong association that has been described between CSCC and immunosuppression. 27 In our series, PNI was a strong independent predictor of recurrence (HR = 2.62, P = .028), and patients with PNI were 2.74 times more likely to present with nodal disease. PNI as a risk factor for recurrence, nodal spread, and poor survival has been widely reported in series of early-stage CSCCs and in series specific to advanced head and neck CSCCs. 4,5,7,18,19,20,22-24 The first definitive report of the negative implications of PNI in CSCCs was published by Goepfert in 1984, where he described more than double the rate of regional (35% vs 15%) and distant (15% vs 3.3%) metastases and decreased survival in patients with head and neck CSCCs with evidence of PNI. 18 Likewise, Schmults et al demonstrated a significantly increased risk of local fail- ure (HR = 2.2) and disease-specific death (HR = 3.6) in patients with early-stage disease and PNI. 4 However, several more recent studies of advanced head and neck CSCCs have not demonstrated this association between PNI and recur- rence or survival. 10,23-26 Poorly differentiated histology has consistently been demonstrated to be a powerful predictor of recurrence and poor survival in a number of dermatologic series of skin cancer patients, with HRs of 2.5 to 3.3 for local recurrence,

head and neck CSCCs treated by head and neck surgeons remains poorly defined. Furthermore, the current AJCC staging system is skewed toward early-stage disease and may be limited in its ability to differentiate outcomes within this high-risk group of patients. 5,16 Given the increasing incidence of CSCC and the high morbidity and mortality of advanced head and neck CSCCs, more data are needed to help guide treatment recommendations and surveillance guidelines and to appropriately counsel patients on prognosis. 2,7-10,23,25 Our study included 212 patients with primary CSCCs of the head and neck that were locally advanced, recurrent, or otherwise deemed too complex for in-office treatment by a dermatologist. Our observed 5-year recurrence rate was 46.8%. This is in contrast to dermatologic series where recurrence rates for all subsites are between 4.6% and 7%; however, it is consistent with other series of advanced head and neck CSCCs where 5-year recurrence rates have been described between 36% and 63%. 4,10,12,23,25 Our cohort of patients contained a large proportion of recurrent tumors (52.4%) and tumors with PNI (36 %), and 96.6% of patients had a tumor with DOI greater than the AJCC-defined high- risk threshold of 2 mm. This is in significant contrast to most large dermatologic series of CSCCs, where rates of PNI and history of recurrence are typically \ 5%. 4,11,12 We

89