Impact of Treatment Timing on Disease Outcomes in Patients Treated With Definitive Concurrent Chemoradiation for Head and Neck Cancer

Abstract

The ideal timing for the initiation of chemotherapy and radiation therapy (RT) in the use of definitive chemoradiation (CRT) for patients with head and neck cancer is not well established. We sought to evaluate the impact of the timing on the initiation of these two modalities on clinical outcomes.Patients with squamous cell carcinoma of the head and neck who were treated using definitive chemoradiation from 2012-2018 were identified. Patients undergoing re-irradiation, post-operative CRT, had recurrent or second primaries, or ECOG 3-4 were excluded. Outcomes including locoregional control (LRC), progression-free survival (PFS), and overall survival (OS) were estimated and compared between subgroups of the cohort based on the timing in which chemotherapy and RT were initiated: chemotherapy first, same day start, within 24 hours, start on Monday/Tuesday/Wednesday (MTW). Univariable Cox proportional hazards models were generated using baseline patient/treatment characteristics including age, gender, ECOG performance status, Charlson Comorbidity Score, stage and type of cisplatin (bolus versus weekly cisplatin) in addition to chemotherapy/RT timing. Multivariable models were generated using the variables found significant (P < 0.25) on univariable models to assess their impact on LRC, PFS and OS.A total of 131 patients were included for analysis consisting of oropharynx (64%), larynx (22.9%), nasopharynx (6.9%), hypopharynx (3.1%), oral cavity (1.5%), and unknown primary (1.5%). Chemotherapy was administered as bolus cisplatin every 3 weeks in 40% of patients and weekly cisplatin in 60% with a median cumulative dose of 240 mg/m2. In the multivariable analysis (MVA), starting chemotherapy before RT was associated with improved LRC (HR 0.33, 95% CI 0.11-0.99, P = 0.048); type of concurrent cisplatin was included in the model but not statistically significant. 3-year LRC for patients starting chemotherapy first was 90.9% compared to 78.2% in those starting RT first. In univariable models, ECOG and bolus cisplatin were both associated with OS, while ECOG alone was associated with PFS. No predictors were found significant on multivariable models for PFS or OS. The median PFS was not reached and 3-year PFS was 69.1% (95% CI: 60.6, 77.6). Similarly, median OS has not been attained, with a 3-year OS of 77.8% (95% CI: 70.0, 85.7).Starting chemotherapy prior to RT improves LRC, but did not impact PFS or OS. Starting on the same day, within 24 hours, or on MTW were not associated with clinical outcomes. Further prospective studies are warranted to confirm the ideal timing in the initiation of chemotherapy and radiation therapy in the use of definitive concurrent chemoradiation for patients with head and neck cancer.