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

First page Previous page 112 Next page Last page

Harris et al

Table 3. Cox Multivariate Analysis of Recurrence. a

Table 4. Factors Associated with Nodal Disease: Multinomial Logistic Regression. a

Factor

P Value

Factor

P Value

Female

1.34 1.93 1.60

.64

Female

0.38

.27 .81

Age . 70 y

.071

Age . 70 y

1.0 NA

Immunosuppressed

.59

Primary site b Ear

Immunosuppressed

—

Primary site b Ear

1.09 0.93 1.69 1.78 1.31 2.40 2.21

.89 .91 .46 .56 .73 .25

16.19 15.04 13.50

.016 .017 .037

Cheek/temple

Lip

Neck Nose

—

1.97 2.28 1.74

.66 .59 .26

Periorbital

Periorbital Recurrent

Recurrent

.041

Diameter, cm b 2-4

0.99 0.88

.98 .82

1.47 1.25

.54 .78

. 4

DOI, cm b 1-2

. 4

DOI, cm b 1-2

1.03 1.26 1.53 2.62 1.31 2.34 0.72

.94 .70 .46

1.19 0.64 0.53 2.74 0.89

.76 .57 .13 .05 .82

. 2

LVI 1

PNI 1

.028

PNI 1

Regional disease

.54

Poorly differentiated

.047

Adjuvant radiation

.39

Abbreviations: DOI, depth of invasion; LVI, lymphovascular invasion; NA, not applicable; OR, odds ratio; PNI, perineural invasion. a Bold indicates statistical significance, P \ .05. b Reference groups: age, 70 years; scalp, 0-2 cm, 0-1 cm.

Abbreviations: DOI, depth of invasion; HR, hazard ratio; LVI, lymphovascu- lar invasion; PNI, perineural invasion. a Bold indicates statistical significance, P \ .05. b Reference groups: age, 70 years; scalp, 0-2 cm, 0-1 cm.

.04) along with the presence of regional nodal disease. 8 In our multivariate analysis, we found that patients in our cohort were more than twice as likely to experience recur- rence (HR = 2.21, P = .041) if they had been previously treated. Despite strong evidence for the association between history of recurrence and future recurrence, the nature of this association remains poorly understood. These results imply that important pathophysiologic features of these tumors are not well quantified by existing objective mea- sures of tumor behavior, such as PNI, lymphovascular inva- sion, and nodal diseases. Further research directed at understanding this relationship is clearly needed. Ear and lip primaries were identified as risk factors for local and nodal recurrence in a dermatologic series and a meta-analysis. 4,15 Within the head and neck literature, ear pri- maries were also identified as a high-risk site for recurrence by a single study, although this was not described in a number of other studies examining similar types of patients. 21 In our series, we did not see a difference in recurrence among primary sites; however, we did find that patients with tumors of the ear, cheek or temple, and lip were more likely to present with regionally metastatic disease. Although the AJCC and National Comprehensive Cancer Network consider tumor size . 2 cm and DOI . 2 mm to be high-risk features, our cohort of patients did not show a

2.3 to 6.1 for regional recurrence, and 4.1 to 6.7 for disease- specific death. 4,5,13,14 For advanced head and neck CSCCs, results are less consistent, although 3 series do demonstrate higher risks of nodal disease, recurrence, and lower survival in patients with poorly differentiated tumors. 19-21 Even though we did not see a univariate association between his- tologic grade and recurrence, our multivariate analysis demonstrated that patients with poorly differentiated tumors were 2.34 times as likely to develop recurrence. Similarly, a history of recurrence has long been identified as a risk factor for future recurrence. In Goepfert’s 1984 study of patients with head and neck CSCCs, patients with previously treated tumors were twice as likely to have a recurrence. 18 History of recurrence is not described as a risk factor in most dermatologic series, where the majority of tumors are treated at their primary presentation. However, a meta-analysis of 71 studies on CSCC published in 1992 did identify history of recurrence as a risk factor for future recurrence. 15 In more recent series looking at CSCCs of the head and neck, history of recurrence has been more reliably identified as a risk factor for future recurrence and poor sur- vival. 7,8,19,22 A study by Sweeny et al published in 2014 contained a similar proportion of recurrent tumors (51%) as our series and found that history of recurrence was a strong risk factor for decreased overall survival (49% vs 69%, P =

Made with FlippingBook Annual report