Prognostic Significance of Referral Patterns and Time Trends in Head and Neck Cancer Management

Abstract

<h3>Purpose/Objectives</h3> Early diagnosis and treatment of head and neck squamous cell carcinoma (HNSCC) is paramount to improving patient prognosis. The purpose of this study was to analyze the association of patient demographic, clinical, and management variables with time to treatment initiation (TTI) and one-year survival. Additionally, we evaluated the effect of potential areas of delay, such as the COVID-19 pandemic, in association with TTI and one-year survival. <h3>Materials/Methods</h3> Medical records of patients who were diagnosed with HNSCC from 2018 to 2020 at a tertiary academic medical center were reviewed to extract demographic and clinical information. Univariate and multivariate linear and cox-regression were used to evaluate TTI and one year survival with our study variables. Kaplan Meier (KM) curves were created to evaluate the difference in hazard over time between patients diagnosed in 2019 and 2020. <h3>Results</h3> A total of 233 patients met eligibility criteria. The median time between symptom onset and presentation to a medical provider was 1 month (Interquartile range [IQR]: 0.5-3). 118 patients (50.9%) were referred to one of our academic head and neck specialists by their PCP, 37 (15.9%) by a dental provider, 32 (13.8%) through the emergency department, 26 (11.2%) by a local ENT, and 20 (8.6%) by other means. Median time for oncologic work-up was 25 days (IQR: 16-38). Median TTI was 37.5 days (IQR: 27-50.75). 21% of patients reported delays in treatment, with the top three reasons cited as concurrent medical problems (34.69%), patient deferrals (16.33%), and delayed dental exams (8.16%). Univariate analysis for TTI found a significant increase in TTI in female patients (p=0.041) and those undergoing salvage therapy (p=0.045), and chemoradiation (p=0.008). Univariate cox-regressions predicting one-year mortality found direct admission status (p=0.005), the use of pre-treatment PEG tubes (p=0.006) and tracheostomies (p=0.025) significantly increased one-year mortality. This association was not significant in a multivariate models. Other demographic and clinical variables, including smoking history, median income by zip code, distance from hospital, tumor subsite, and others, did not significantly impact TTI or one-year mortality. Hazard ratios for one-year mortality and Kaplan-Meier estimates for one-year survival were not significant when adjusted for year of diagnosis (p>0.05) and TTI quartile (p>0.05). <h3>Conclusion</h3> Our analysis determined that many patient and management variables may not significantly impact TTI and one-year survival. Patients diagnosed in 2020 showed no difference in overall survival compared to prior years, despite COVID-19. The scope for improved HNSCC management may lie in continuing to optimize care in a timely manner for those with characteristics of advanced disease, many of which were found to be predictive of one-year mortality.