Temporal Change and Risk Factor of Esophageal Stenosis after Salvage Radiotherapy for Superficial Esophageal Cancer with Non-Curative Endoscopic Submucosal Dissection

Abstract

Recent studies revealed the favorable outcomes of chemoradiotherapy for superficial esophageal cancer, so much attention has been paid to radiotherapy (RT) for an additional treatment after non-curative endoscopic submucosal dissection (ESD) as an organ preservation strategy. Although one concern regarding toxicity in this treatment strategy is esophageal stenosis, there are no reports about its detail. The aim of this study is to investigate esophageal stenosis after salvage RT for superficial esophageal cancer with non-curative ESD. We retrospectively reviewed 50 superficial esophageal cancer patients received salvage RT after non-curative ESD at one institution between 2011 and 2018. Tumor location was cervical/upper thoracic/middle thoracic/lower thoracic in 5/4/29/12 patients, respectively. Pathological invasion depth was mucosal/submucosal/muscularis propria in 20/29/1 patients, respectively. Median tumor length was 3 cm (range: 0.5-10 cm) and the circumference of tumor was more than three quarters in 10 patients. Prevention of stenosis such as oral prednisolone or intralesional steroid injection was performed in 21 patients. Median total dose was 60 Gy/30 fractions (range: 40-66 Gy). Forty-five patients received concurrent chemotherapy. Comparison of date was analyzed by Fisher’s exact test. P-values <0.05 were considered to indicate statistical significance. Median follow-up time was 43 months (range: 12-91) for all patients. Grade 2 esophageal stenosis was observed in 16 patients and grade 3 was observed in 3 patients. Grade 3 esophageal stenosis was significantly higher in cervical esophageal cancer (p = 0.02). The frequency of grade 2 or worse esophageal stenosis decreased over time (before RT, 6 months after RT, 1year, 2years: 15, 13, 10, 6 patients). Endoscopic balloon dilation (EBD) was performed in 10 patients before RT and in 14 patients after RT. Nine of 10 patients who received EBD before RT needed EBD after RT. Only one patient needed EBD regularly one year after RT. All grade 3 esophageal stenosis improved grade 2 or less by EBD. Tumor circumference ≥ three quarters, tumor length ≥ 4 cm, implementation of stenosis prevention, and diabetes were significantly associated with EBD (p<0.05). Total dose and pathological invasion depth were not associated in EBD. Esophageal stenosis after salvage RT for esophageal cancer with non-curative ESD improved over time by natural course or EBD, and only few cases required long-term EBD. Our results suggest esophageal stenosis was manageable and acceptable. However, cervical esophageal cancer is a risk factor of severe esophageal stenosis, so we consider that it is important to carefully determine the indication for ESD in those cases.