Predicting Progression-free Survival And Metastasis In 149 Patients With Esophageal Cancer After Esophagectomy

Background. This investigation was to study the prognostic factors that determined progression-free survival (PFS) and metastatic sites (local or distant) for esophageal squamous cell carcinoma (ESCC) after surgery.Methods. A total of 149 patients who received surgery treatment with stage T2-3N0M0 and T1-3N1M0 were retrospectively studied. Patients were divided into four group:surgery group, surgery-radiotherapy group, surgery-chemotherapy group, surgery-radiochemotherapy group. Chi-square test and fisher exact were used to analyze the prognostic factors with regard to metastatic sites. PFS was calculated using the Kaplan-Meier method and differences were analyzed with the log-rank test.Results. Differences related to metastatic sites did not show statistical significance on gender (P=0.133), age (P=0.737), length (P=0.773), depth of invasion (P=0.906), histological grade (P=0.069), and treatment (P=0.067). However, it showed a significant difference on lymph node (P=0.006). Differences related to PFS also only showed a significant difference on lymph node. PFS for patients with lymph nodes 0 and≥1 was 14 mo (9-19) versus 12 mo (11-13) (P=0.028), but it did not show statistical significance on gender (P=0.386), age (P=0.541), length (P= 0.736), depth of invasion (P=0.825), histological grade (P=0.340), and treatment (P= 0.113). What kind of treatment was taken had no significant differences in metastatic sites (P=0.067) and PFS (P=0.113). However further research found out that for patients with no lymph node, there was no difference in PFS among different treatment groups (P=0.729), but for patients with positive lymph nodes≥1, postoperative radiochemothrapy group had a longer PFS compared to other three groups (P= 0.011). For patients with local metastasis, Median PFS was 12 mo (95%CI:11-13) in surgery group,19 mo (95%CI:4-34) in surgery-radiotherapy group,11 mo (95%CI: 8-14) in surgery-chemotherapy group, and 23 mo (95%CI:19-27) in surgery-radiochemotherapy group; Median PFS was 22 mo (95%CI:17-27) in patients with postoperative radiotherapy (surgery-radiotherapy group+surgery-radiochemotherapy group),12 mo (95%CI:11-13) in patients without postoperative radiotherapy treatment (surgery group+surgery-chemotherapy group) (P=0.001); For patients with distant metastasis, Median PFS was 13 mo (95%CI:11-15) in surgery group,9 mo (95%CI:7-11) in surgery-radiotherapy group, 19 mo (95%CI:10-28) in surgery-chemotherapy group, and 12 mo (95%CI:9-15) in surgery-radiochemotherapy group; Median PFS was 16 mo (95%CI:10-22) in patients with postoperative chemotherapy (surgery-chemotherapy group+surgery-radiochemotherapy group),11 mo (95%CI:10-13) in patients without postoperative chemotherapy treatment (surgery group+surgery-radiotherapy group) (P=0.002).Conclusions. The number of involved lymph nodes was an independent predictor of metastatic sites and PFS. For patients with no lymph node, adjuvant postoperative treatment did not prolong the PFS; for patients with lymph node metastasis, single-adjuvant postoperative treatment did not prolong the PFS, but double-adjuvant postoperative treatment obviously prolong the PFS. For patients with local metastasis, it had longer PFS in patients with postoperative radiotherapy treatment (surgery-radiotherapy group+surgery-radiochemotherapy group) than patients without postoperative radiotherapy treatment (surgery group+surgery-chemotherapy group); for patients with distant metastasis, it had longer PFS in patients with postoperative chemotherapy treatment (surgery-chemotherapy group+ surgery-radiochemotherapy group) than patients without postoperative chemotherapy treatment (surgery group+surgery-radiotherapy group). So for patients without postoperative chemotherapy, the possibilities of distant metastasis are more at risk; in clinical reviews, we should shorten the inspection cycle of the abdominal computed tomography. For patients with postoperative chemotherapy, the possibilities of distant metastasis are less at risk, it is not necessary to shorten the inspection cycle of abdominal computed tomography.