We decided to compare oncological outcome of neoadjuvant chemotherapy followed by surgery (NAC) with adjuvant chemotherapy following surgery (ADJ), and NAC with definitive chemoradiotherapy for locally advanced esophageal cancer patients.
Consecutive patients histologically diagnosed with ESCC and planned to undergo NAC, ADJ and dCRT as initial treatment were eligible for this retrospective study. In ADJ, NAC and dCRT group, overall survival and non complete resection rate were investigated. We analyzed in intent-to-treat analysis.
98 patients included in NAC group, 90 were included in ADJ group, and 47 were included in dCRT group. No significant differences were observed in clinicopathological factors among NAC group (cStage IB/II/III/IV = 10/28/52/8) and ADJ group (cStage IB/II/III/IV = 6/22/57/5) and dCRT group (cStage IB/II/III/IV = 3/15/20/7). In intention-to-treat analysis, there was no significant difference in 3yOS rate by NAC group and ADJ group (NAC group, 66%; ADJ group, 67%; p = 0.301). However, non complete resection rate was significantly higher for NAC group than for ADJ group (ADJ group, 3.3%; NAC group, 16.7%; p = 0.009). In cStage III, non complete resection rate of ADJ group was 3.5% and NAC group was 25%. In the patients with cStage III, NAC group had a tendency for 3yOS rate to be low rather than ADJ group (NAC group, 51.1%; ADJ group, 66.3%; p = 0.09). In intention-to-treat analysis, there was no significant difference in relapse free survival between NAC group and dCRT group. In contrast, the rate of 3yOS tend to be higher in the NAC group than the dCRT group (NAC, 66.0%; dCRT, 49.8%; p = 0.166). After initial treatment failure, significantly more patients could undergo local treatment in NAC group than dCRT group (NAC, 74%; dCRT, 40%; P = 0.003).
In ESCC patients who could tolerate transthoracic esophagectomy, no significant difference was observed in survival between the NAC and dCRT groups. However, NAC group was extended short-term survival (3yOS), which might have been encouraged by utilizing local treatment after initial treatment failure.
Clinical trial identification
All authors have declared no conflicts of interest.