Neoadjuvant Therapy for Advanced Esophageal Cancer: Still No Optimal Surgery

Abstract

Esophageal cancer are usually advanced and preoperative treatment has become the standard-of-care. The Dutch CROSS trial with Carboplatin, Paclitaxel with 41.4Gy of external radiotherapy resulted in a very impressive high pCR rate and superior survival compared to surgery alone, especially for squamous cell cancer (ESCC). However outside the randomized controlled trial, whether the results are reproducible or not, especially in Asian countries, is debatable. We analyzed our patients who had the CROSS regimen from 2012-2017, dividing patients into two groups: those within the selection criteria in the CROSS trial and those beyond (age >75 yrs old, tumor length >8cm, and clinical M1 stage of nodal involvement (AJCC 6th edition). Eighty-eight patients were included. There were 46 patients in the “CROSS eligible” group and 42 in the “CROSS ineligible” group. By intention-to-treat, the median survival was 24.2 months vs. 12.7 months, respectively (p=0.047). The results were much inferior compared to that published in the original CROSS trial. Multivariable analysis showed tumor length and resectability as independent variables affecting survival. The excellent results from CROSS were not reproducible, patient selection could be a key element accountable for the difference. In addition, we retrospectively studied outcome comparing the CROSS regimen with a historical set of patients who had undergone a regimen with Cisplatin and 5-Fluorouracil with 40 - 46 Gy of radiotherapy (PFRT). A propensity score-matched cohort was retrieved using patients from 2002-2012. One hundred patients were included in each group. R0 resection rates were 86.8% and 88.2% in CROSS and PFRT group respectively (p=0.06). pCR rates were 25% and 35.5% (p=0.17). Median survival showed a trend favoring the PFRT group, median survival were 16.7 months vs. 32.7 months (p=0.08). CROSS is perhaps the most widely used chemoradiation regimen for ESCC. The results are suboptimal. Best treatment strategy for locally advanced ESCC remains elusive.