Impact of the Charlson Comorbidity Index on the treatment strategy and survival in elderly patients after non-curative endoscopic submucosal dissection for esophageal squamous cell carcinoma: a multicenter retrospective study

Background In elderly patients with superficial esophageal squamous cell carcinoma (ESCC), the optimal treatment strategy after non-curative endoscopic submucosal dissection (ESD) remains unclear. We aimed to evaluate the validity of additional treatments after non-curative ESD and post-ESD survival predictors in elderly patients with ESCC. Methods Elderly patients (age > 75 years) treated with ESD for ESCC between January 2010 and July 2014 at six tertiary referral hospitals in Japan were retrospectively investigated and stratified according to lymph node metastasis risk, based on histological findings (high-risk factors: positive lymphovascular invasion, submucosal invasion, and positive/indeterminate vertical margin) and post-ESD treatment strategy: group A (287 patients; low risk), group B (41 patients; high risk, without additional treatment), and group C (32 patients; high risk, with additional treatment). We evaluated 3- and 5-year overall survival and disease-specific survival, and prognostic factors for post-ESD survival. Results At a median follow-up of 38, 40, and 49 months, respectively, there was 1 esophageal cancer-related death in group A, 1 in group B, and none in group C, whereas 22, 9, and 3 patients in groups A, B, and C died of other diseases. The groups differed significantly in overall survival (92.4%; 87.6%; 93.4%, p = 0.022), although not in disease-specific survival (99.4%; 96.3%; 100%, p = 0.217). On multivariate analysis, Charlson Comorbidity Index (CCI) >= 2 was the only independent risk factor for post-ESD death (hazard ratio 7.92; 95% confidence interval 3.42-18.3; p < 0.001). Conclusions A follow-up strategy without additional treatment after ESD for ESCC may be acceptable in high-risk elderly patients, especially for CCI >= 2.