Nodal status and advanced clinical stage as determinants of response and survival after induction chemoradiotherapy in esophageal and gastroesophageal junction cancer: a single-centre retrospective cohort study

Background: Induction (neoadjuvant) chemoradiotherapy based on the CROSS regimen (weekly carboplatin/paclitaxel with concurrent radiation) is a standard of care for resectable esophageal and gastroesophageal junction (GEJ) cancer. However, the pivotal trials enrolled predominantly cT2–3 tumors with limited nodal involvement, did not stratify by nodal status, and largely excluded cT4, cN2–3, and oligometastatic (cM1) disease. Because these higher-risk patients are common in routine practice yet under-represented in trials, real-world data are needed to clarify how baseline nodal and clinical-stage affect outcomes after CROSS-based treatment. We evaluated the association between baseline clinical-stage and outcomes following CROSS-based treatment in a real-world cohort.

Methods: We performed a single-centre retrospective cohort study of consecutive patients treated with CROSS-based induction chemoradiotherapy [weekly carboplatin area under the curve (AUC) =2 and paclitaxel 50 mg/m² with concurrent 41.4 Gy] for esophageal/GEJ cancer between 2012 and 2022 at a Canadian regional cancer centre. Patients were identified from institutional pathology and systemic-therapy databases, cross-referenced with electronic records, and stratified a priori by initial clinical staging: group 1 (cT1–3N0M0), group 2 (cT1–3N1M0), and group 3 (cT4 and/or cN2–3 and/or oligometastatic cM1), the latter largely excluded from CROSS-based trials. Overall survival (OS) and progression-free survival (PFS), both measured from diagnosis, were primary outcomes; resection rate and pathologic complete response (pCR) were secondary. Survival was estimated using Kaplan-Meier methods and compared with log-rank testing; multivariable Cox models adjusted for age, sex, and Eastern Cooperative Oncology Group (ECOG) status, with two-sided statistical significance defined as P<0.05.

Results: Among 138 patients (median age, 67 years; 81% male; 11% ECOG 2–3), 49% were group 1, 39% group 2, and 12% group 3, with no significant differences in age, sex, ECOG performance status, tumor location, histology, grade, or human epidermal growth factor receptor 2 (HER2) status. Overall, 86 deaths and 102 progression events occurred. Resection rates decreased with advancing clinical burden (70.6%, 57.4%, and 25.0%, respectively; P=0.004). Among resected patients, pCR occurred in 18.8%, 12.9%, and 0% (P=0.77). Median PFS was 34, 13, and 6 months, respectively (global log-rank P=0.001). Median OS was 46, 19, and 17 months (P=0.09). In adjusted models, PFS was significantly worse in group 2 [hazard ratio (HR) =1.86; 95% confidence interval (CI): 1.20–2.89] and group 3 (HR =2.84; 95% CI: 1.54–5.25), whereas the OS difference for group 3 was imprecise (HR =1.85; 95% CI: 0.91–3.78; P=0.09).

Conclusions: Greater baseline clinical disease burden was associated with progressively lower resection and pCR rates and significantly shorter PFS following CROSS-based induction chemoradiotherapy, with outcomes for cN1 (group 2) patients approaching those of the advanced-stage group. Given the retrospective design and the small advanced-stage cohort, these findings are hypothesis-generating but are consistent with a need for intensified, systemic-dominant perioperative strategies for clinically node-positive, advanced locoregional, or oligometastatic esophageal/GEJ cancer.