Abstract
PURPOSE: The optimal mode of neoadjuvant treatment for esophageal squamous cell carcinoma (ESCC) has not been well characterized. Our study compared neoadjuvant chemotherapy (NCT) with neoadjuvant chemoradiotherapy (NCRT) for patients with ESCC. METHODS: Data from ESCC patients receiving NCRT or NCT combined with esophagectomy between 2010 and 2018 from the National Cancer Center in China were retrospectively collected. Long-term survival, pathological response, and perioperative mortality and morbidity were compared between the NCRT and NCT groups. A Cox proportional hazards model and propensity score matching (PSM) were used to minimize bias due to potential confounding. RESULTS: Out of 327 eligible patients with ESCC in our study, 90 patients were identified in each group by PSM. The complete pathologic response (pCR) rate in the NCRT group was markedly higher than that in the NCT group (before PSM: 35.1% vs. 6.0%; after PSM: 38.9% vs. 5.6%; both P < 0.001). The rates of 30-day or 90-day mortality were comparable between the two groups, but the NCRT group had a longer postoperative hospital stay (P < 0.001 before PSM and P = 0.012 after PSM) and more postoperative complications (P < 0.001 before PSM and P = 0.014 after PSM), especially, anastomotic leaks (P = 0.001 before PSM and P = 0.013 after PSM). No significant differences in 5-year overall survival (OS) (P = 0.439) or 5-year relapse-free survival (RFS) (P = 0.611) were noted between unmatched groups, but the trend favored NCRT in the propensity score-matched group (77.3% vs. 61.3%; hazard ratio [HR] 1.57; 95% confidence interval [CI] 0.86-2.87; P = 0.141 for OS, and 77.8% vs. 60.5%; HR 1.72; 95% CI 0.95-3.11; P = 0.073 for RFS). Multivariate analysis showed that only ypT and ypN stages were independent predictors of OS before and after PSM (both P < 0.05). CONCLUSION: There was no difference in survival between the NCT and NCRT groups, although a trend favored NCRT related to the significantly higher pCR rates. Prospective head-to-head clinical trials to compare these two types of neoadjuvant therapies in ESCC are warranted.