Factors Predictive of Primary Therapeutic Modality in Early-Stage Oropharyngeal Squamous Cell Carcinoma

Abstract

<h3>Purpose/Objective(s)</h3> Traditionally, definitive radiotherapy (RT) was the preferred treatment for oropharyngeal squamous cell carcinoma (OPSCC). Changes in both surgical technology (the advent of transoral robotic surgery) and the pathogenesis of the disease (increase in patients with HPV-associated disease) have altered this paradigm. We aimed to assess factors influencing the use of definitive surgery versus RT among patients with early stage OPSCC. <h3>Materials/Methods</h3> We used the Surveillance, Epidemiology, and End Results (SEER) Head and Neck with HPV Status Database along with the SEER Census Tract-level Socioeconomic Status (SES) and Rurality Database. From 40,787 patients treated from 2010 – 2016, we identified 9,292 who met inclusion criteria (T0-3, N0-2, M0 disease with known status for all variables included in our model). Patients undergoing pharyngectomy were deemed to have definitive surgery. Those receiving RT in the absence of a pharyngectomy were deemed to have definitive RT. We performed a multivariable logistic regression for factors correlated with definitive RT. Our model initially included age and year of diagnosis, gender, race, primary site, T- and N-classification, HPV status, marital status, rural/urban status, and socioeconomic status (SES) as indicated by Yost index. We used a backwards stepwise selection algorithm until no variable had a p value > 0.10; the final model included all variables except gender, race, and year of diagnosis. <h3>Results</h3> 6,008 patients (64.7%) received definitive RT and 3,284 (35.3%) received definitive surgery. Of those receiving definitive surgery, 2,510 (76.4%) also received RT. On multivariable logistic regression, higher T- or N-classification and non-tonsillar fossa primary site significantly increased the likelihood of definitive RT. Patients with HPV-positive disease were less likely to receive definitive RT (odds ratio (OR) = 0.87, 95% CI 0.77 – 0.98, p = 0.021). Patients in entirely rural census tracts were less likely to receive definitive RT (OR = 0.81, 95% CI 0.67 – 0.99, p = 0.042). Those living in census tracts in the lowest SES tertile were more likely to receive definitive RT (OR = 1.17, 95% CI 1.02 – 1.33, p = 0.014). <h3>Conclusion</h3> We identified sociodemographic and clinical factors associated with primary treatment for early stage OPSCC. Tumor size and primary site, appear to influence treatment allocation; however, non-clinical factors may also play a role. Patients living in rural areas are more likely to receive definitive surgery; patients in lower SES areas are more likely to receive definitive RT. Further investigation into these differences is warranted to ensure equal access to all therapeutic options for patients with early stage OPSCC.