Back to Annual Meeting Program

The exact role of neoadjuvant therapy (neoT) including its prognostic influence in esophageal cancer is still under debate. Pooled data (metaanalysis) suggest a prognostic advantage of neoT but definitve data are lacking. We analyzed our institutional experience with resected esophageal cancer including the effect of neodjuvant therapy on long-term outcome.
Methods: We evaluated overall survival in 304 patients undergoing esophageal resection between 1988 and 2010 (patients with hospital mortality excluded). 53% had squamous cell (SCC) and 46% adenocarcinoma (AC). Indications for neoT were in general T-stage >2 and/or positively staged lymph nodes. Tumors were in the lower third in 64%. 66% of the patients underwent neoT (60% chemoradiation 36 Gy+FU+Cisplatin; 6% chemotherapy alone). The proportion of neoT increased from 16% in the first third to 78% in the last third of the study period. In pathological analysis the median number of examined nodes was 17; 43% were node positive. Survival was analyzed by the Kaplan-Meier- and Cox-models.
Results: The proportion of patients with AC increased from 22% (first third) to 61% (last third of the study period). After neoT 81% of the patients showed partial or total response. Patients without neoT had more frequently positive margins (13% vs 4% after neoT; p<0.01). Postoperative nodal disease was independent on neoT (40% after neoT vs 50% without neoT) although patients in the neoT group had more frequently positive nodes in pretherapeutical staging (71% vs 39% in patients without neoT; p<0.01). Overall 5-year survival (5y-Surv) was 36% and improved clearly during the study period (5y-Surv 14% until 1994; 35% 1994-2001; 49% since 2002; p<0.001), parallel to an increased use of neoT. This significant improvement in survival over time was also seen in the subgroups of patients with SCC (p<0.01) and AC (p<0.001). 5y-Surv in patients with response (any/total) was 52%/60%, but only 19% in patients without response/without neoT (p<0.001). In further univariate analysis positive nodes (p<0.001), positive margins (p<0.001) and LNR>0.1 (p<0.001) significantly worsened prognosis. In multivariate analysis a LNR>0.1 (p<0.001; RR 11), no response to neoT (p<0.01; RR 1.6) and SCC (p<0.02; RR 1.5) were independent negative prognostic factors. Compared to SCC patients with AC had higher rates of positive margins (10% vs 4%) and LNR > 0.1 (43% vs 16%).
Conclusions: Tumorbiological parameters (histological type, LN-ratio) influenced prognosis after resection of esophageal cancer. Response to neoadjuvant therapy independently improved the outcome and contributed to the clearly better outcome achieved in the later study period.