Postoperative radiotherapy for carcinoma of the esophagus: a prospective, randomized controlled study

A prospective, randomized controlled study of radiotherapy after resection of esophageal carcinoma was carried out in 130 patients. Patients were stratified according to whether the resection was curative or palliative and were then randomized to receive postoperative radiotherapy or no additional treatment. Sixty patients underwent curative resection; 30 each were randomized into the radiotherapy group (CR + R) and the control group (CR). Seventy patients underwent palliative resection; 35 each were randomized into the radiotherapy group (PR + R) and the control group (PR).

No complications occurred while the patients were undergoing radiotherapy treatment. On follow-up, complications in the intrathoracic stomach occurred in 24 patients (37%) who underwent radiotherapy compared with four patients (6%) in the control group (p < 0.0001). Seventeen of these 24 patients in the radiotherapy group had gastric ulceration and there were five deaths as a result of bleeding. Local recurrence developed significantly less frequently in the PR + R group compared with the PR group (seven patients [20%] vs 16 patients [46%]; p = 0.04); no difference was observed between CR + R and CR groups (10% and 13%, respectively). Intrathoracic recurrence occurred in fewer patients in the radiotherapy groups (CR + R and PR + R) compared with the control groups (CR and PR) (four patients vs 15 patients; p = 0.01). In patients with residual tumor in the mediastinum after resection, two (7%) of 29 patients who underwent radiotherapy died of tracheobronchial obstruction, compared with nine (33%) of 27 patients in the control groups (p = 0.03). No difference in local recurrence was observed for extrathoracic or anastomotic recurrence. Distant metastasis developed in 12 patients (40%) in the CR + R group, nine patients (30%) in the CR group (p = 0.59), 24 patients (69%) in the PR + R group, and 18 patients (51%) in the PR group (p = 0.22). The time of onset of metastasis was 5.1 months for the PR + R group, which was shorter than the 8.5 months for the PR group (p = 0.05). The time of onset of metastasis was similar for the CR + R and CR groups (9.9 months and 11.0 months, respectively; p = 0.76). The overall median survival of patients after postoperative radiotherapy (CR + R and PR + R) was 8.7 months, which was shorter than the 15.2 months for the control groups (CR and PR) (p = 0.02).

The shorter survival of patients who underwent postoperative radiotherapy was the result of irradiation-related death and the early appearance of metastatic diseases. The role of postoperative radiotherapy is therefore limited to a specific group of patients with residual tumor in the mediastinum after operation, for whom radiotherapy can significantly reduce the incidence of local recurrence obstructing the tracheobronchial tree.