Endoscopic Ultrasound After Neoadjuvant Chemotherapy in Locally Advanced Esophageal Cancer: Can it Predict Survival?

Abstract

Endoscopic ultrasound (EUS) plays a central role in the initial staging and treatment of esophageal cancer (EC). However, EUS for re-staging pts after neoadjuvant therapy is controversial. Few studies have suggested that EUS post-neoadjuvant therapy can accurately predict tumor response and overall survival in pts with locally advanced disease. Aim: Evaluate if EUS re-staging after neoadjuvant chemotherapy can predict long-term survival in patients with locally advanced operable EC. Methods: Data from an ongoing study on neoadjuvant chemotherapy (Taxol, FUDR, Leucovorin and Cisplatin) for locally advanced EC (1999-2004) was collected. Radial EUS performed pre- and post-chemotherapy. Maximum tumor thickness was measured on EUS pre- and post-chemotherapy and EUS stage compared to surgical pathology (AJCC TNM stage). More than 50% wall thickness reduction was classified as a response by EUS. Results: 38 pts. were included (33M, 5F), mean age 62 yrs. There were 35 adenocarcinomas of the gastroesophageal junction and three squamous carcinomas. The median follow-up time among 26 pts still alive was 19 months (range: 3-75 months). The 1-, 3- and 5-yrs overall survival was respectively: 88%, 63% and 45%. The 1- and 3-yrs survival of pts who had more than 50% reduction in tumor thickness after chemotherapy was superior to those with less than 50%: 1yr survival (92% vs. 85%) and 3-yrs (70% vs. 59%). Log-rank test of Kaplan-Meier survival curves was not statistically significant p=0.39. The Log-rank test showed that pts who lowered their T or N stage on EUS post-chemotherapy had statistically significant better overall survival than those without downstage (L-R test, p=0.004). At 3-yrs the survival in the downstage group was 88% compared to 40% in those that did not downstage. N1 disease on surgery was associated with worse survival (L-R test, p=0.012). At 1-year the survival in the N1 group compared to N0 group was 79% vs. 94%, respectively and at 3-yrs: 47% vs. 83%. Conclusion: Patients with tumor response (>50% reduction of wall thickness) to neoadjuvant chemotherapy verified by EUS appears to have an improved survival but it did not reach statistical significance in our study at this point. Tumor downstaging after neoadjuvant chemotherapy verified by EUS and absence of nodes on pathological staging are associated with improved survival. The role of EUS appears promising in predicting patient's survival based on tumor response after neoadjuvant chemotherapy. Endoscopic ultrasound (EUS) plays a central role in the initial staging and treatment of esophageal cancer (EC). However, EUS for re-staging pts after neoadjuvant therapy is controversial. Few studies have suggested that EUS post-neoadjuvant therapy can accurately predict tumor response and overall survival in pts with locally advanced disease. Aim: Evaluate if EUS re-staging after neoadjuvant chemotherapy can predict long-term survival in patients with locally advanced operable EC. Methods: Data from an ongoing study on neoadjuvant chemotherapy (Taxol, FUDR, Leucovorin and Cisplatin) for locally advanced EC (1999-2004) was collected. Radial EUS performed pre- and post-chemotherapy. Maximum tumor thickness was measured on EUS pre- and post-chemotherapy and EUS stage compared to surgical pathology (AJCC TNM stage). More than 50% wall thickness reduction was classified as a response by EUS. Results: 38 pts. were included (33M, 5F), mean age 62 yrs. There were 35 adenocarcinomas of the gastroesophageal junction and three squamous carcinomas. The median follow-up time among 26 pts still alive was 19 months (range: 3-75 months). The 1-, 3- and 5-yrs overall survival was respectively: 88%, 63% and 45%. The 1- and 3-yrs survival of pts who had more than 50% reduction in tumor thickness after chemotherapy was superior to those with less than 50%: 1yr survival (92% vs. 85%) and 3-yrs (70% vs. 59%). Log-rank test of Kaplan-Meier survival curves was not statistically significant p=0.39. The Log-rank test showed that pts who lowered their T or N stage on EUS post-chemotherapy had statistically significant better overall survival than those without downstage (L-R test, p=0.004). At 3-yrs the survival in the downstage group was 88% compared to 40% in those that did not downstage. N1 disease on surgery was associated with worse survival (L-R test, p=0.012). At 1-year the survival in the N1 group compared to N0 group was 79% vs. 94%, respectively and at 3-yrs: 47% vs. 83%. Conclusion: Patients with tumor response (>50% reduction of wall thickness) to neoadjuvant chemotherapy verified by EUS appears to have an improved survival but it did not reach statistical significance in our study at this point. Tumor downstaging after neoadjuvant chemotherapy verified by EUS and absence of nodes on pathological staging are associated with improved survival. The role of EUS appears promising in predicting patient's survival based on tumor response after neoadjuvant chemotherapy.