Intro: Concurrent chemoradiation (CCRT) is a proven treatment for early stage (T2/T3) laryngeal squamous cell carcinoma (LSCC) with good rates of organ preservation. However, rates of dysphagia and aspiration after CCRT range from 37 to 82%, respectively and contribute to morbidity in patients who are otherwise ‘disease free’. Known risk factors for poor laryngeal function following CCRT include radiation dose, T- and N-stage, smoking, and patient factors. However, involvement of critical structures within the larynx by tumor may contribute to worse post-treatment outcomes. As such, we examined the relationship between early-stage laryngeal cancer patients receiving CCRT and tumor subsite with post-treatment outcomes of dysphagia and aspiration.
Methods: A retrospective cohort study of T2 or T3 (AJCC 6th) LSCC patients was performed at our academic institution. All patients underwent CCRT. Primary tumor stage, nodal status, anatomic subsite involvement, vocal cord paralysis/paresis, age, and aspiration events were recorded as mentioned in notes. Degree of swallowing dysfunction was assessed secondarily using patient BMI and G-tube status. Primary outcomes were change in BMI, measured from treatment start to 6 months, and from 6 months after treatment out to 2.5 years. Secondary outcomes were total G-tube time after treatment, total trach time after treatment, and aspiration events in the first year. Longitudinal analysis was used to compare subsite involvement and BMI change. Secondary outcomes were assessed with univariate and multivariate analysis.
Results: We found 53 patients with LSCC treated with CCRT at WVU. Overall the majority of patients had a significant decrease in BMI after treatment with a gradual increase in the measured time period thereafter. We found that no LSCC subsite predicted a significant change in the initial drop in BMI, but involvement of the arytenoid was a significant predictor of a slower rate of return to baseline BMI (p<0.03). Sub-group analysis found that patients with N+ disease had significantly worse weight loss with treatment and, in contrast to N0 patients, never regained their original BMI. Secondary outcome analysis showed that pre-epiglottic and paraglottic space involvement predicted significantly longer G-tube time after treatment (p<0.05). Additionally, vocal cord paresis or paralysis resulted in significantly increased rates of aspiration pneumonia and tracheostomy time (p < 0.05).
Discussion: We present a retrospective review examining the association of laryngeal subsite involvement with post-treatment dysphagia and aspiration outcomes for T2/T3 LSCC. To date, this is the only known study examining this relationship. Amongst subsites examined, we found that arytenoid involvement can predict a slower return to pre-treatment BMI. Additionally, the presence of nodal disease may portend overall lower BMI following treatment. Finally, we found that patients with vocal cord paresis or paralysis had high rates of aspiration events. These subsites correlate to paraglottic invasion, suggesting that patients with paraglottic involvement undergoing CCRT should be counseled about pulmonary complications or potential functional laryngectomy. Additional studies should focus on prospective data and objective measures of dysphagia and aspiration to better identify early-stage laryngeal cancer patients who are at risk for serious post-treatment outcomes.