A retrospective analysis of cisplatin dosing strategies when used with radiation on outcomes in head and neck squamous cell carcinoma of the oropharynx.

Abstract

6079 Background: Cis is the gold standard radiosensitizer for CRT to treat head and neck cancer. Prospective trials have required that cis be administered early in the week (Mon/Tues) to optimize radiosensitization without evidence to support this practice. This retrospective analysis considers the impact of cis day of week (DOW) administration and other variables on OPSCC pt outcomes. Methods: We reviewed OPSCC cases treated with primary CRT at our center. Pts treated with non-cis or induction chemotherapy were excluded. Data collected includes age, DOW (Mon/Tues vs Wed/Thurs/Fri), smoking status, total dose (TD) of cis (≤200mg/m2 vs > 200mg/m2), single dose (SiD) [100mg/m2 x1 day] vs split dose (SpD) [50mg/m2 x2 days] administration, T stage (0-2b vs 2c-3), N stage (0-2b vs 2c-3), overall survival (OS) (from start of RT), local/regional/distant failure, KPS and HPV/p16 status. Univariate Cox proportional hazards regression was used to analyze OS and multivariate Cox proportional hazard model investigated the relationships between OS and cis dosing variables while controlling for other factors. Results: 745 pts with OPSCC were treated with CRT from 7/31/2001-2/7/2014. 459 used cis based regimens and were included. Median age at start of RT was 55. 311 pts (67.8%) received SpD, 124 (27%) received SiD, 8 (1.7%) received weekly cis and 16 (3.5) received mixed cycles. 269 (58.6%) received ≤200mg/m2. 232 (50.5%) were HPV/p16 positive, 40 (8.7%) were negative and 187 (40.7%) were unknown. Median f/u was 7.9yrs (1.8m -18.9yrs). There were 92 (20%) deaths, and 75 (16.4%) recurrences (local/regional and distant) with 44 (9.6%) competing events. In univariate analysis, age, N stage, T stage, KPS and HPV/p16 status were significantly associated with OS, while DOW, TD, and SiD/SpD were not. In multivariate analysis (MVA), none of the associations between cis dosing and OS were significant (although MVA was limited by low number of events and total variables included). Conclusions: This retrospective analysis suggests that the DOW cis is given has no impact upon CRT outcomes for OPSCC pts. SpD cis represents an alternative administration approach.