LAUSR

As a novel metric to evaluate the quality of oncosurgical therapy, such as minimal invasive surgery, Return to Intended Oncologic Therapy (RIOT) has not been applied in patients with esophageal cancer (EC). This study aims to profile RIOT in locally advanced EC patients and to quantify its relationship with overall survival. We conducted a retrospective study on consecutive locally advanced EC (T3–4 and/or N1–3) patients who received esophagectomies followed by postoperative chemotherapy (PC) from April 2015 to August 2017. RIOT included whether the patient did or did not undergo intended PC and the time between surgery and the start of PC. Overall survival at each RIOT group was compared via log-rank test. Cox regression models were used to estimate the prognostic value of RIOT. Among 658 locally advanced EC patients (547 males and 111 females) with complete PC data, 433 received minimal invasive esophagectomies (MIE) and 225 received open esophagectomies (OE). The RIOT rates were 58.0% for MIE and 54.2% for OE (P = 0.358). The 1-year overall survival rate of patients receiving PC was higher than that of patients not receiving PC (88.2% vs 76.4%; P = 0.005). After adjustment of age, gender, surgery type and postoperative length of stay, patients with PC showed significantly better OS than those without PC (HR 0.60, 95% CI 0.41–0.87; P = 0.007). Total 253 patients (MIE 168, OE 85) presented verified dates of starting PC. Median RIOT time was 42 days (min-max, 13–162) for MIE and 43 days (16–169) for OE (P = 0.855). Among those 253 patients, 179 (70.8%) started RIOT within 8 weeks. After 8 weeks, every one week delay of RIOT related to a 17% increase on the risk of death (P = 0.014). Using a real world data, our study provided baseline profiles of RIOT in locally advanced EC patients who received esophagectomies and PC. Compared to OE, MIE did not show a significant advantage in RIOT rates and RIOT time. In spite of the short follow-up, successful RIOT is related to better OS. Prospective studies with longer follow-up are required for further application of RIOT in the evaluation of oncosurgical therapy. All authors have declared no conflicts of interest.