Objective  To investigate the failure mode of patients with esophageal squamous cell carcinoma after receiving different irradiation methods, especially the failure mode of patients in complete remission (CR) after treatment.

  Methods  Patients with esophageal squamous cell carcinoma who met the enrollment criteria were retrospectively analyzed. Overall survival and recurrence rates of all patients and recurrence rates and recurrence location of patients who received elective nodal irradiation (ENI) and those who received involved-field irradiation (IFI) were analyzed. Next, differences in recurrence rates of patients with different therapeutic effects and differences in recurrence sites of patients with CR after treatment were analyzed between the ENI and IFI groups. Furthermore, a multifactor analysis of independent factors affecting the recurrence of all patients was performed. Finally, propensity score matching (PSM) analysis was performed to adjust the composition ratio of ENI and IFI patients. The differences in recurrence between the ENI and IFI groups after PSM were further compared, and independent factors affecting the incidence of patients after PSM were analyzed. All statistical analyses were performed using the SPSS19.0 statistical software.

  Results  The 1-, 3-, and 5-year recurrence rates were 27.6%, 52.2%, and 59.4%, respectively, and the median recurrence time was 11.6 months. The recurrence rates in the IFI and ENI groups were 47.3% and 49.2%, respectively, with no statistically significant difference (χ²=0.182, P=0.670). The difference in recurrence sites between the IFI and ENI groups was statistically significant (χ²=10.372, P=0.006). Moreover, the difference in immediate effects between the IFI and ENI groups was statistically significant (χ²=13.385, P=0.001). However, there was no significant difference in the overall recurrence rate (31.7% vs. 45.2%) (χ²=3.167, P=0.075) and recurrence site (χ²=3.457, P=0.178) in patients with CR in the ENI and IFI groups. After PSM analysis, each group included 172 patients. There was no statistically significant difference in the overall recurrence rate between the two groups (χ²=0.570, P=0.450), but the difference in the recurrence location was statistically significant (χ²=10.513, P=0.005). There was a statistically significant difference in the short-term effect between the two groups of patients (χ²=7.226, P=0.027). Compared with CR patients in the ENI group after PSM, the recurrence rate of CR patients in the IFI group (53.3% vs. 29.5%) was significantly increased (χ²=6.145, P=0.013), and the difference in recurrence sites between the two groups was also significant (χ²=7.467, P=0.024). Short-term efficacy and the presence of esophageal cancer lesions were independent factors affecting recurrence in patients after PSM analysis (P=0.014, 0.018).

  Conclusions  There was no statistically significant difference in the total treatment failure rate between the ENI and IFI groups; however, there was a significant difference in the recurrence site. Compared with CR patients after treatment in the IFI group, the field recurrence rate of patients in the ENI group was significantly reduced, but the recurrence rate in the field was higher. These conclusions need to be confirmed by carefully designed, larger, multicenter prospective trials.