Neoadjuvant Chemoradiotherapy Versus Perioperative Chemotherapy ± Immunotherapy in Resectable Esophageal Adenocarcinoma.

Objective

We sought to compare pathologic responses, perioperative outcomes, and early survival between induction chemoradiotherapy and perioperative chemotherapy ± immunotherapy in patients with resectable esophageal adenocarcinoma.

Methods

Using our prospectively maintained institutional database, we retrospectively identified adult patients with resectable, locally advanced esophageal or gastroesophageal junction adenocarcinoma who underwent esophagectomy from January 2013 to October 2024 after induction chemoradiotherapy or perioperative chemotherapy ± immunotherapy. The primary outcome was rate of pathologic complete response. Secondary outcomes included rate of microscopically margin-negative resection, rate of pathologic node-negative status, and perioperative morbidity. Event-free survival and overall survival were estimated using Kaplan-Meier methods and were evaluated using Cox regression.

Results

Of 811 patients, 650 (80%) received chemoradiotherapy, and 161 (20%) received chemotherapy ± immunotherapy. Median follow-up was 29.9 months. Chemoradiotherapy was associated with higher rates of pathologic complete response (23% vs. 14%) and pathologic node-negative status (61% vs. 46%; both P<0.001 after adjustment). The rate of microscopically margin-negative resection was 96% in both groups. Two-year event-free survival was higher in the chemotherapy ± immunotherapy group (66% vs. 53%; adjusted hazard ratio, 0.53; P=0.007). Two-year overall survival was not statistically different between groups. On multivariable analysis, perioperative chemotherapy ± immunotherapy and pathologic complete response were independently associated with longer event-free survival.

Conclusions

Both induction strategies achieved excellent perioperative outcomes. Despite higher rates of pathologic complete response with chemoradiotherapy, perioperative chemotherapy ± immunotherapy was associated with longer early event-free survival. This association was attenuated in propensity score-matched and 90-day landmark analyses. Longer follow-up is warranted.