P-0008THE BENEFITS OF NEO-ADJUVANT CHEMORADIOTHERAPY FOR ESOPHAGEAL CANCER, ARE THEY FOR ALL?

Background: Neoadjuvant chemoradiotherapy (NCRT) followed by esophagectomy is the most common approach for patients (pts) with locally advanced esophageal cancer (EC). NCRT has the potential to downstage EC, increase complete resection (R0) and to induce a complete pathologic remission (cPR) which has been demonstrated to be a predictor of survival. However a considerable number of pts are non-responders to NCRT. The aim of this study was to analyze clinical and pathologic response to NCRT and correlate it with survival. Methods: We retrospectively reviewed clinical files of 54 pts diagnosed with EC between 2008 and 2012. Of these we included 25 pts with either squamous cell carcinoma (SCC) or adenocarcinoma (ADC) that were selected for NCRT. Restaging after NCRT was performed according to the RECIST criteria v1.1. The pathologic response (PR) was classified according to the regression grade criteria of the College of American Pathologists (G0= complete PR, G1-2= partial PR, G3= poor PR). The statistical analysis was performed in SPSS v20. Results: Twenty-five pts were included, of which 23 were males and the overall median age was 56 years. The histologic diagnosis was SCC in 80% of pts and ADC in the remaining 20%. The anatomic distribution of EC was thoracic in 80% of cases and 20% at the esophagogastric junction. At the initial staging there were 3 pts with T2 (12%), 14 with T3 (56%), 8 with T4 (32%) and 21 with N+ (84%). Among the 25 pts that started NCRT a total of 8 were not submitted to surgery because of clinical progression (decreasing performance status, unresectability or metastatic disease). Of those 17 pts surgically resected 7 (41,2%) had a cPR, 9 (52,9%) had a pPR and 1 had no response. There was a complete resection (RO) in 16 (94.7%) pts. The median follow-up was 790 (108-1841) days. The median overall survival for pts submitted to surgery after NCRT was significantly higher comparing with those without tumor resection (492 vs.138 days, p = 0,000). Although we did find different median overall survival for pts with cPR vs. pPR (1058 vs. 429 days) we didn’t find statistical significance for this result (p = 0.165). When comparing pathologic remission grades G0, G1 and G2 we found a significantly different overall survivals among the three groups (1058, 492 and 259 days, respectively, p = 0.041). Conclusion: This study demonstrated that NCRT was effective at inducing a complete pathologic response and downstaging locally advanced EC allowing surgery even in pts initially unresectable, which increased their overall survival. Though surgery after NCRT increased survival, there is still an important group of pts who did not respond to NCRT. Further studies aimed at identifying response predictors are needed.