<h3>Purpose/Objective(s)</h3> We investigated whether increasing the RT dose from 60 Gy to 67-68 Gy, which is comparable to the dose for squamous cell carcinoma of the head and neck, would improve the treatment outcome in cT3-4NxM0 patients enrolled for definitive chemoradiotherapy (CRT) for locally advanced cervical esophageal cancer based on late adverse events and recurrence patterns. <h3>Materials/Methods</h3> From 2004 to 2020, 83 patients with cT3-4bNxM0 squamous cell carcinoma of the cervical esophagus (including 22 cases of Ce, 8 cases of CePh, and 53 cases of CeUt, but not UtCe) were enrolled in a radical CRT (1.8-2.1 Gy/F, total dose >60 Gy, and concurrent CDDP80mg/m<sup>2</sup>x1 and 5-fluorouracil 800 mg/m<sup>2</sup>x5, 2 courses) protocol at our institution. Pretreatment imaging, including MRI and ultrasound tracheal endoscopy, showed that 71% of 59 patients were cT4 (89% tracheal invasion) and the rest were cT3. cN1 or higher was found in 73 patients (88%) and cN0 in 10 patients (12%). Thirty-six patients were in the cohort group for VMAT of 67-68 Gy/33 F/7 wks with D50 indication for PTV, and the rest were in the non-cohort group. In the remaining 47 patients in the non-cohort (3DCRT) group, 30 patients were completed at 60 Gy (or less than 60 Gy for surgery), and only 14 patients underwent 3DCRT of 62 Gy or more, which was presumably added due to poor response. A total of four patients were judged to be candidates for salvage surgery at 50 Gy intermediate assessment: one in the cohort group and three in the non-cohort group. The median observation period for survivors was 87.4 months (95% CI: 65.0-109.7). <h3>Results</h3> OS was 30.3% (SE 5.4%) at 5 years in all cT3-4NxM083 patients. cT3/T4, cN0/N1, age, gender ratio, and tumor occupying subsite CePh/CeUt types in cohort vs. non-cohort were not significantly skewed. There was no significant difference in OS between the cohort group and the non-cohort group in the ITT comparison of all patients, including the 4 patients who were converted to surgery (Breslow, p=0.84). Multivariate analysis showed that the two factors significant for OS were cT3 vs. cT4 (p=0.035, HR=2.12) and tumor occupation site Ce/CePh vs. CeUt (p=0.047, HR=1.91). There was no significant difference in local recurrence and regional recurrence, 41.5% and 12.2% in the cohort group and 39.5% and 7.9% in the non-cohort group, respectively. The main late adverse events Grade 3 or higher were esophageal stenosis in 6 cases and bilateral recurrent nerve palsy in 1 case, but there was no significant difference between the cohort group (3 cases) and the non-cohort group (4 cases), and it could not be said to occur at doses higher than 60 Gy by dose. <h3>Conclusion</h3> It has long been pointed out that cervical esophageal cancer may require higher doses than thoracic esophageal cancer. However, in advanced cancers of cT3 or higher, treatment planning to obtain an appropriate dose distribution of 60 Gy or higher in VMAT did not directly lead to improved OS.
Read full abstract