Which neoadjuvant treatment is best for locally advanced esophageal cancer? a systematic review and network meta-analysis.

Abstract

e16121 Background: In recent years, perioperative treatment strategies for locally advanced esophageal cancer (LAEC) have become more diverse, including neoadjuvant chemotherapy (NCT), neoadjuvant plus adjuvant chemotherapy (NCT+ACT), neoadjuvant chemoradiotherapy (NCRT), neoadjuvant immunochemotherapy (NICT), neoadjuvant immunoradiotherapy (NIRT), neoadjuvant immmunochemoradiotherapy (NICRT), and targeted therapy. However, it remains unclear which treatment approach offers the most benefit to patients. Methods: A systematic literature search was conducted in PubMed, Web of Science, Embase, Cochrane, ClinicalTrials.gov, and American Society of Clinical Oncology (ASCO) meeting proceedings. The search focused on phase II/III randomized controlled trials (RCTs) investigating neoadjuvant treatments for EC. Key outcomes including pathological complete response (pCR), R0 resection rate, recurrence, overall survival (OS), and 3 or higher treatment-related adverse events (TRAEs)，were analyzed. This systematic review was adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Network meta-analysis was conducted using a Bayesian frequentist approach. Indirect comparisons were presented in league tables, with hazard ratios (HRs) and 95% confidence intervals (CIs) for OS, and relative risks (RRs) and 95% CIs for others. A subgroup analysis was performed for ESCC. The protocol is registered with PROSPERO (registration number: CRD42025633839). Results: A total of 12 phase III RCTs and 5 phase II RCTs involving 4,249 patients were included. For pCR, NCRT, NICT, NCT+ACT, and NICRT demonstrated superior benefits compared to NCT, (RR: 0.23; 95%CI, 0.09-0.48 for NCRT, 0.36; 0.19-0.64 for NICT, 0.16; 0.03-0.66 for NCT+ACT, 0.13; 0.04-0.38 for NICRT). Additionally, NICRT outperformed NICT(0.38; 0.11-0.99). NICT showed a higher R0 resection rate than other treatments, supported by SUCRA analysis. Regarding recurrence, all treatments were better than surgery alone. NCRT combined with targeted therapy reduced recurrence risk compared to NCRT (RR: 0.55; 95%CI, 0.3-1), NICT (RR: 0.41; 0.18-0.94) and NCT (0.47; 0.22-0.97). NCRT, NCT+ACT and NICRT showed an increased risk of grade 3 or higher TRAEs than NCT (2.6; 1.14-7.53, 3; 1.01-12-16, 3.79; 1.35-13.85), with NCRT showing a higher risk than NICT (2.16; 1.04-5.21). For OS, SUCRA showed that all treatments were superior to surgery alone. Additionally, subgroup analysis for ESCC confirmed consistent results in terms of pCR and R0 resection rate. Conclusions: This network meta-analysis suggests that NICRT may be the the optimal treatment for LAEC patients. While the addition of radiotherapy increases the risk of grade 3 or higher TRAEs, incorporating targeted therapy further lowers the recurrence risk. Longer follow-up data and results from Phase III RCTs are expected.