부산대학교 도서관

Background Nearly one third of esophageal cancer patients show a pathologically complete response in their resection specimens after neoadjuvant chemoradiotherapy (nCRT) according to CROSS regimen. This raises questions whether all patients benefit from surgery or if active surveillance can be applied to patients with a clinically complete response (cCR) after nCRT. This retrospective-multicenter propensity matched study compared outcomes of patients with a cCR after nCRT undergoing active surveillance or standard surgery. Methods Patients that refused surgery after nCRT between 2012–2017 from 4 hospitals were included. For the standard surgery group, patients from the preSANO trial were enrolled. A cCR was defined as endoscopies with multiple (bite-on-bite) biopsies, EUS-FNA and PET-CT showing no residual disease 6 and 12 weeks after completion of nCRT. Optimal propensity-score matching generated a matched cohort (1:2) matched for age, comorbidities, cT, cN, histology of the tumor and biopsy type. For comparison of severity of complications according to Clavien-Dindo (CD) classification, a separate optimal propensity-score matching cohort was generated (1:2) for all patients in the active surveillance group that underwent surgery. Primary outcome was overall survival, secondary outcomes were rate of radically resected tumors, distant dissemination rate and rate of postoperative complications according to the CD-classification. Results 75 patients were identified of whom 50 patients underwent standard surgery and 25 patients underwent active surveillance. 13 of 25 patients in the active surveillance group underwent surgery for locoregional recurrent disease. Median follow-up was 23.7 months for the standard surgery group and 18.8 months for the active surveillance group. There was no statistically significant difference between the groups in overall survival (HR = 0.48, 95%C.I. 0.10–2.2, P  = 0.96). In both groups, all tumors were radically resected. There were no statistically significant differences in distant dissemination rate between the active surveillance and standard surgery group (16.0% versus 22.0%, P = 0.76) or in severity of complications (CD ≥ 3;46.2% versus 23.1%, P = 0.16). Conclusion There was no statistically significant difference in overall survival, distant dissemination rate and severity of complications between patients undergoing standard surgery or active surveillance after nCRT. However, since sample sizes were small, especially for the severity of complications, these results should be interpreted with caution. Disclosure All authors have declared no conflicts of interest. [ABSTRACT FROM AUTHOR]