Impact of dosimetric factors on long-term percutaneous enteral gastrostomy (PEG) tube dependence in head and neck cancer patients after (chemo)radiotherapy. Results from a prospective randomized trial

To analyze dose-volume histogram (DVH)-derived data on the exposure of organs at risk with impact on long-term percutaneous enteral gastrostomy (PEG) tube dependence in head and neck cancer patients at 6 and 12 months after definitive or adjuvant (chemo)radiotherapy. Sixty-one patients were prospectively treated with (chemo)radiotherapy. Prophylactic or reactive gastrostomy tube placement was performed in 41 (67.2%) patients. Dose-volume histogram parameters were obtained for the swallowing apparatus. Median follow-up time was 25 (2-34) months. Overall survival was shorter in patients with inlying PEG tubes at 6 and 12 months (log rank p = 0.038 and p = 0.017) after therapy completion. The estimated median time of tube dependency was 6 (95% confidence interval: 2-14) months. After 6 months, 46.5% of patients were tube dependent. After 12 months, this estimated proportion fell to 31.5%. For both time points, the volume to the larynx (in %) receiving at least 50 Gy (larynx V50Gy) exceeding 53% was predictive for long-term tube feeding (6 months: p = 0.041 and 12 months: p = 0.042) being an independent predictor during multivariable analysis. There was no clinical feature influencing tube dependence after 12 months. Long-term gastrostomy dependence was found to be strongly associated with an exposure of laryngeal structures (specifically, V50Gy ≥ 53%) during radiotherapy. Consequently, the avoidance of supraglottic as well as glottic structures is warranted.