Purpose/Objective(s): Merkel Cell Carcinoma (MCC) is a neuroendocrine carcinoma of the skin commonly occurring in the head and neck region. The purpose of this study is to evaluate role of total radiation dose delivered in MCC patients receiving surgery with adjuvant radiation.
Materials/Methods: 14,414 MCC cases (histology code 8247) diagnosed between 1998 and 2011 were identified from the National Cancer Database (NCDB). After excluding non-HN cases, patients receiving surgery or radiation alone, patients with distant metastases, palliative treatment and missing information on follow-up data, 2,249 MCC patients receiving surgery with adjuvant radiation were eligible for analysis. Radiation treatment related exclusions included: non-external beam treatment modalities (N=26), missing regional dose (N=433) and patients receiving radiation dose of <30 Gy and >70 Gy (N=165). A final analysis population comprised of 1,625 patients. Total dose delivered was calculated as sum of regional and boost dose and was divided into the following categories: 30-<50 Gy, 50-55 Gy (referent) and >55-70 Gy. Total dose delivered was correlated with overall survival (OS) and 3-year survival rates were estimated using the Kaplan-Meier method. Crude and adjusted hazard ratios (HR) with 95% confidence intervals (CI) were computed using Cox regression modeling. In bivariate and multivariate models following patient characteristics were analyzed: age (<65 and 65+ years), gender, tumor size (≤2, >2-≤3, >3-≤4, >4 cm), positive lymph nodes (0, 1, 2 or more), surgical margins (negative vs. positive margins), lymph node surgery, surgery type (local, gross and wide), Charlson-Deyo Score/CDS (0, 1 and 2) and adjuvant chemotherapy (no vs. yes). A probability value of < 0.05 was considered statistically significant.
Results: Total radiation dose delivered correlated with age (P=0.027), tumor size (P=0.002), positive lymph nodes (P<0.0001), surgical margins (P<0.0001), lymph node surgery (P<0.0001), CDS score (P=0.027), and surgery type (P=0.001). Gender (P=0.847) and adjuvant chemotherapy (P=0.30) distribution was not different by total dose delivered. The median follow-up was 33.5 and 48.2 months in whole cohort and surviving patients, respectively. 808 deaths were reported with a 3-year OS of 62.4%. The 3-year OS rate by total dose delivered was 48.9%, 70.3 % and 58.7% for 30-<50 Gy, 50-55 Gy and >55-70 Gy, respectively (P<0.001). Compared to 50-55 Gy, the crude HR for 30-<50 Gy and >55-70 Gy groups were 1.74 (CI 1.42-2.13) and 1.39 (CI 1.20-1.62), respectively. In a bivariate analysis all patient factors were added individually in the model and the adjusted HRs generated from these analyses did not significantly alter the crude HR (All P<0.05). All factors except lymph nodes status (due to >15% missing data) were analyzed in a multivariate model. On multivariate model after adjusting for age, gender, tumor size, surgical margins, lymph node, surgery type, Charlson-Deyo Score/CDS and adjuvant chemotherapy the adjusted HR for 30-<50 Gy and >55-70 Gy groups were 1.45 (CI 1.08-1.93) and 1.30 (CI 1.06-1.59), respectively.
Conclusions: In this cohort of MCC patients receiving surgery with adjuvant radiation, total treatment dose delivered between 50-55Gy was associated with the best overall survival. The result was independent of patient demographic, tumor and treatment factors.