LYMPHADENECTOMY AND SURVIVAL AFTER NEOADJUVANT CHEMORADIATION FOR ESOPHAGEAL CANCER: IS MORE BETTER?
Susanna W. de Geus*1, sameer hirji2, Kei Suzuki1, Teviah Sachs1, Sing Chau Ng1, Scott Swanson2, Virginia Litle1, Jennifer F. Tseng1
1Boston Medical Center, Bostom, MA; 2Brigham and Women's, Bostom, MA
Background: The National Comprehensive Cancer Network (NCCN) Guidelines for Esophageal Cancer recommend resection of at least 15 lymph nodes for patients treated with upfront surgery. In contrast, recent post hoc analysis of a European randomized controlled (CROSS) trial suggests that after neoadjuvant chemoradiation (nCRT) the association between the number of lymph nodes resected and survival disappears. In addition, the optimal number of lymph nodes needed to be examined after nCRT remains unclear. Therefore, the purpose of this study was to assess the impact of number of lymph nodes examined on survival in patients with esophageal adenocarcinoma (EAC) who underwent nCRT.
Methods: The National Cancer Database (NCDB) was queried for patients who underwent surgery for EAC. Lymph node downstaging was defined as cN > (y)pN. In patients who underwent nCRT, propensity scores were created predicting the odds of undergoing resection of ≥ 15 nodes. Patients were matched on propensity score. Overall survival analyses were performed using the Kaplan-Meier method. Sensitivity analyses were performed using various nodal cutoffs.
Results: In total, 1,313 (16.9%) patients were identified. nCRT was associated with lymph node downstaging (37.8% vs. 4.7%; p<0.001) compared to upfront surgery. nCRT was also predictive for resection of < 15 lymph nodes (vs. ≥ 15: Adjusted Odds Ratio [AOR], 1.24; p=0.002) on multivariable analyses. After nCRT, resection of at least 15 nodes was performed in only 2,085 (46.7%) patients. In addition, examination of ≥ 15 lymph nodes (vs. < 15: AOR, 1.965; p<0.001) was predictive for having ≥ 2 positive nodes, even after nCRT. In patients who underwent nCRT, resection of ≥ 15 lymph nodes was associated with significant survival benefit (median survival, 38 vs. 31 months; p<0.001) after matching. On sensitivity analyses, resection of ≥ 20 nodes (vs. 15-20: median survival, 42 vs. 35 months; p=0.0048) and ≥ 25 nodes (vs. 20-25: median survival, 47 vs. 40 months; p=0.0976) were predictive of favorable survival. However, resection of ≥ 30 nodes (vs. 25-30: median survival, 46 vs. 47 months; p=0.2115) was not.
Conclusions: After nCRT, resection of 15 - 30 lymph nodes was associated with longer median survival. Removal of thirty nodes may provide adequate nodal staging after nCRT. Prospective trials are warranted to determine the optimal nodal yield after nCRT.
Back to 2019 Abstracts
