LAUSR

Objective The optimal treatment for resectable esophageal squamous cell carcinoma (ESCC) remains controversial. Surgery is the primary treatment but with poor results. Attempts to improve patient survival have been made by introducing chemotherapy, radiotherapy, or both. However, randomized comparisons for all these strategies are not always available. This network meta-analysis compared the overall survival of neoadjuvant and adjuvant therapy with surgery alone to identify the most effective approach. Methods We systematically searched electronic databases (PubMed, Embase, and Cochrane Library) for relevant studies published before April 2021. Only phase II and III randomized controlled trials comparing the following treatments were included: surgery alone, neoadjuvant chemotherapy (NCT), radiotherapy (NRT) or chemoradiotherapy (NCRT), adjuvant chemotherapy (ACT), radiotherapy (ART), or chemoradiotherapy (ACRT). The hazard ratios (HR) and 95% confidence intervals (CIs) of overall survival (OS) was identified as the measurement of effectiveness. A network meta-analysis was conducted to synthesize the evidence under the Bayesian framework, and the relative effects of all possible comparisons were made. The ranking analysis was performed to support the decision in clinical practice. Results A total of 19 relevant trials with 3,749 patients were identified. Compared with surgery alone, NCRT (HR 0.76, 95% CI 0.65–0.89) and NCT (HR 0.81, 95% CI 0.70–0.94) significantly improved OS, while other treatments, including NRT (HR 0.86, 95% CI 0.66–1.08), ACRT (HR 0.73, 95% CI 0.49–1.08), ACT (HR 0.96, 95% CI 0.75–1.21), and ART (HR 0.86, 95% CI 0.66–1.14), provided no significant survival advantage. None of the neoadjuvant and adjuvant treatments showed a statistically significant difference in OS to each other when compared in pairs. Conclusion For resectable esophageal squamous cell carcinoma, this network meta-analysis showed that NCRT may be the optimal strategy, NCT may be the second choice, while other multimodality treatments could not improve OS compared with surgery alone. It remains unclear whether ESCC will benefit from adding radiotherapy into the neoadjuvant treatment. Systematic Review Registration We registered this meta-analysis protocol at the prospective register of systematic reviews, PROSPERO https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=172745 (Identification code: CRD42020172745).