Outcomes Following Radiation for Cutaneous Squamous Cell Carcinoma (cSCC)

Abstract

Most cases of cutaneous squamous cell carcinoma (cSCC) are cured with surgical excision; however, patients with high-risk features or unresectable tumors are often treated with radiation therapy (RT). Limited data are available to guide adjuvant treatment and estimate risk and patterns of recurrence following RT. We performed a retrospective analysis of cSCC patients treated with RT at 4 affiliated institutions between 1/2007 and 6/2017. Immunosuppression (IS) was defined as patients with organ transplant, heme malignancy/lymphoma, HIV, daily use of >20 mg prednisone (or equivalent) or immunomodulators. Biological Effective Dose (BED) was calculated using alpha/beta = 10. Predictors of recurrence were determined using Chi-squared testing and logistic regression. We identified 105 patients. Median age was 75 years (range 43-102 years); 82% were male and 13% were IS. The most common primary site was scalp (46%) followed by ear (20%), lip (12%) and cheek (12%). Seven (6%) had regional node metastasis. Initial treatment included surgery in 91 (87%) of patients, of which 47 underwent Mohs. Among these 91 patients, 51 (56%) had positive margins, and 47 (45%) had perineural invasion (PNI). Median prescribed RT dose was 60 Gy among 51 patients treated with photons and 58 Gy in 54 patients treated with electrons. Concurrent chemotherapy was administered in 5 (5%) patients, 4 (80%) of whom were treated in the adjuvant setting. With median follow-up of 7 months (range 1-43 months), there were 28 (27%) recurrences. Most of the recurrences were regional (n=17, 16%). There were 9 (9%) in-field, 4 (4%) marginal, and no distant recurrences. On univariate analyses, factors associated with any recurrence were IS (p=0.03) and RT parameters: photon RT (p=0.02), lower dose (p=0.046), decreased number of fractions (p=0.049), and lower BED (p=0.041). Depth of invasion was also marginally inversely associated with risk of recurrence (p=0.045), possibly because patients with increased depth received higher BED treatment (p=0.01). The association between recurrence and IS was driven by in-field failures (7% in non-IS versus 16% in IS) and regional recurrences (13% in non-IS versus 32% in IS). Among the 14 patients who did not receive surgery and were treated with definitive RT, there was 1 in-field failure (7%) in a patient that received hypofractionated RT to 36 Gy in 6 fractions. There were also 1-marginal (7%) and 2 regional (14%) recurrences among the definitive RT patients. We report among the largest experience of outcomes following radiation for cSCC. We find favorable rates of in-field control, even among patients treated with definitive radiation. IS was associated with higher rates of regional and in-field recurrence, suggesting a need for further evaluation of elective nodal treatment and intensification strategies in this high-risk patient population.