Neoadjuvant chemoradiation (nCRT) followed by surgery is the preferred treatment for locally advanced ESCC. But a recent trial suggested close observation might be a reasonable option in patients achieving clinical complete response (CR) to nCRT. For this strategy, accurate clinical assessment for predicting pathologic CR (pCR) is essential. In NCCN guidelines PET/CT is recommend as response assessment, whereas endoscopy is optional after nCRT.
In 234 patients who received nCRT (46–50.4Gy) plus surgery for locally advanced ESCC at Asan Medical Center from 2007 to 2014, the performance of endoscopy and PET/CT which were done 4-8 weeks after nCRT for predicting pCR was evaluated. Metabolic CR (mCR) was defined as complete resolution of FDG uptake within all lesions, making them indistinguishable from surrounding tissue, and endoscopic CR (eCR) as no residual mucosal lesions except for scar change.
pCR (ypT0N0) was achieved in 108 patients (46.2%), and ypT0N+ in 17 (7.3%). Among patients who underwent PET/CT (n = 231), mCR was obtained in 102 (44.2%), and non-mCR in 81 (35.1%), whereas metabolic response could not be assessed due to diffuse esophagitis in 48 (20.8%). The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of mCR for pCR was 56.6%, 46.4%, 58.8%, and 71.6%, respectively. Among patients who underwent endoscopy (n = 229), eCR was obtained in 42 (18.3%), and the sensitivity, specificity, PPV, and NPV of eCR for pCR only in primary tumor site (ypT0N+/-) was 29.5%, 94.4%, 85.7%, and 54.0%, respectively. When adding endoscopic response to metabolic response, the sensitivity, specificity, PPV, and NPV of clinical CR for pCR was 27.9%, 94.3%, 80.6%, and 60.5%, respectively, and the positive likelihood ratio for pCR was 4.9 (95% CI 2.2–10.6).
The addition of endoscopic evaluation to metabolic response after nCRT improved specificity and PPV for pCR compared to metabolic response alone, which could help in applying surveillance strategy without immediate surgery in patients achieving clinical CR after nCRT for ESCC.
Clinical trial identification
Legal entity responsible for the study
Asan Medical Center.
Asan Medical Center.
All authors have declared no conflicts of interest.