Radiotherapy is often used for head and neck cancers patients with high-risk features after definitive resection. Current guidelines generally recommend postoperative radiation therapy (PORT) for pathological T3 or T4 primary, pathological N2 or N3, nodal disease in neck levels IV or V, or other minor risk factors such as vascular embolism or perineural invasion \cite{Rosenthal2017}. Concurrent chemotherapy is further recommended for positive surgical margins or extracapsular spread \cite{Bernier2005}
The National Comprehensive Cancer Network guidelines for head and neck states that the preferred interval between resection and postoperative radiotherapy is < 6 weeks \cite{22017}. This recommendation is based on data from a clinical trial conducted in the 1980s, which found treatment delay greater than 6 weeks to be associated with an increase in cancer recurrence \cite{8482629}. However, conflicting reports based on retrospective data from Memorial Sloan Kettering \cite{2325418} and France \cite{11163507} have argued delay in starting radiotherapy does not by itself have negative impact if doses of >60 Gy is given. A systemic review in 2003 suggested radiotherapy initiation later than 6 weeks after surgery was associated with increased loco-regional recurrence rates, albeit most of the studies were retrospective in nature \cite{Huang_2003}. A very recent study based on the National Cancer Database reviewed nearly 42,000 patients with head and neck cancer who were treated with PORT. After adjusting for covariates, they found a detriment in overall survival for patients who inititated PORT >6 weeks compared with those <6 weeks \cite{Graboyes_2017}.  In contemporary practice, this issue is further complicated by the use of concurrent chemotherapy, since previous studies were all based on adjuvant therapy with radiation alone. 
Based on the above studies, the current consensus is to start radiotherapy within 6 weeks postoperatively. Previous studies have chosen to dichotomize the interval due to statistical considerations, thus grouping patients with minor delays (such as 7 weeks) together with severe delays (such as >10 weeks). There is also a lack of data to support 6 weeks as the optimal cut-point. We aim to examine the hypothesis that patients with shorter or minimally longer time-to-radiotherapy intervals have similar outcomes with standard intervals. We will utilize the Taiwan Cancer Registry Database, which captures the majority of cancer patients diagnosed and treated in Taiwan. We are able to conduct this analysis because data on a very large number of patients is available. Although most patients will receive adjuvant radiotherapy at 6 weeks after resection, we expect a substantial proportion will start adjuvant radiotherapy at 5 weeks or 7 weeks postoperatively. This allows us to compare prognostic outcomes, and establish whether a minor delay in starting adjuvant radiotherapy does indeed impact the local control or disease-free survival in head and neck cancer patients.