Neoadjuvant Chemoradiotherapy Versus Chemotherapy in Resectable Esophageal and Esophagogastric Junction(EGJ) Cancer: Propensity-Weighted and Matched Analyses of a Histology-Guided Cohort
摘要
Whether the higher pathologic complete response (pCR) rates seen with neoadjuvant chemoradiotherapy (NACTRT) translate into superior survival compared with neoadjuvant chemotherapy (NACT) in resectable esophageal and esophagogastric junction (EGJ) cancers remains unclear, particularly in the era of modern perioperative chemotherapy. This study evaluated the comparative effectiveness of NACTRT versus NACT within a real-world, histology-guided treatment framework. A retrospective cohort study was performed including consecutive patients with locally advanced esophageal or EGJ carcinoma who received neoadjuvant therapy (either NACT or NACTRT) followed by planned curative-intent esophagectomy between 2019 and 2025 at a tertiary cancer center. Since January 2023, our institution adopts a histology-based neoadjuvant policy: squamous cell carcinoma (SCC) of the esophagus is treated with NACTRT, whereas lower esophageal/EGJ adenocarcinoma is treated with NACT. The primary endpoint was pCR (ypT0N0). Secondary endpoints included recurrence-free survival (RFS) and overall survival (OS). Propensity score methods – inverse probability of treatment weighting (IPTW) and 1:1 propensity score matching (PSM) – were used to balance baseline covariates between treatment groups. Weighted logistic regression assessed associations with pCR. Weighted Cox proportional hazards models and restricted mean survival time (RMST) analyses evaluated RFS/OS, with Schoenfeld residuals testing proportional hazards (PH) assumptions. A sensitivity analysis excluding very delayed surgeries (> 90 days post-neoadjuvant therapy) was performed. Multivariable Cox regression identified independent prognostic factors for RFS and OS. Nomograms to predict 3- and 5-year RFS and OS were constructed based on key pathologic factors. SCC comprised 67.4% and adenocarcinoma 31% of the cohort; 86.1% received NACT and 13.9% NACTRT. Overall, pCR was achieved in 27.3% of patients. NACTRT yielded substantially higher pCR rates than NACT (64.0% vs. 22.4%, p < 0.001). After IPTW adjustment, NACTRT remained associated with significantly greater odds of pCR (weighted OR ≈3; p < 0.01). However, neither unadjusted nor IPTW-adjusted survival analyses demonstrated a significant difference between NACTRT and NACT. Median RFS was ~ 20 months for both groups. At 2 years, RFS was ~ 50% with NACT vs. ~ 75% with NACTRT (p = 0.98), and OS was ~ 74% vs. ~ 85% (p = 0.85) after weighting. PSM cohort analysis confirmed these findings, showing no significant RFS or OS benefit for NACTRT (log-rank p > 0.8). In this histology-guided neoadjuvant strategy, NACTRT achieved marked tumor response gains (higher pCR) without a clear survival separation compared to modern chemotherapy. These real-world outcomes indicate that while NACTRT offers excellent locoregional control, optimal systemic therapy remains paramount for long-term survival. Treatment should be individualized by histology, prioritizing perioperative chemotherapy for adenocarcinoma and NACTRT for SCC to maximize local control.