Mo1679 OUTCOMES OF ENDOSCOPIC RESECTION VERSUS LAPAROSCOPIC SURGERY FOR SUBMUCOSALLY INVASIVE COLORECTAL CANCERS: AN OBSERVATIONAL STUDY

Abstract

Endoscopic resection (ER) has a distinct advantage for removing submucosally invasive (T1) colorectal cancers. Endoscopic submucosal dissection (ESD) enables en bloc resection regardless of tumor size, and offers appropriate treatment allowing precise histological assessment. The indications for laparoscopic colectomy (LC) overlap with those for ER. The aim of this study was to reconsider the treatment strategy for T1 cancers by comparing outcomes between ER and LC. Patients with T1 colorectal cancers resected from May 2013 to December 2018 in one institution were reviewed. Synchronous multiple cancers in any organs were excluded. T1 cancers were classified as T1a (<1mm invasion) or T1b (≥1mm invasion). Patients followed appropriate clinical pathways. Outcome measures included background characteristics, short-term and long-term outcomes. Complications are classified by Clavien-Dindo grades. Eighty-one patients (female 30, male 51; mean age 68.5±11.3 years) underwent ER (ESD 43%, endoscopic mucosal resection (EMR) 38%, polypectomy 19%) and 87 patients (female 36, male 51; mean age 67.3±10.6 years) underwent LC (additional LC after ER 57%, laparotomy 1%). There were no differences in age or gender between ER and LC, but a significant difference was observed in the proportion of T1b (ER65% vs. LC90%, p<0.0001). En bloc resection rate of ER was 91% (74/81, piecemeal; ESD 2, EMR 5). In ER, Clavien-Dindo grade II complications occurred in 9.9% (8/81; post-bleeding 5, intraprocedural perforation 2, post-penetration 1), all treated endoscopically or nonoperatively. In LC, Clavien-Dindo grade II or greater complications occurred in 9.2% (8/87; anastomotic leakage 2, severe delirium 2, intestinal perforation 1, intestinal obstruction 1, intraabdominal abscess 1 and anastomotic hemorrhage 1), and 2 patients underwent surgical intervention. Median days to oral intake (ER 1, LC 2, p = 0.0188) and hospital stay (ER 3, LC 9, p <0.00001) were shorter in ER. Quality of life was decreased in one patient with anastomotic leakage. Median observation time (interquartile range, months) was 26.7 (14.5-47.6) in ER alone (n=39), 34.0 (22.0-46.9) in LC alone (n=37) and 48.1 (24.1-61.1) in LC after ER (n=50). Recurrence occurred in one patient who had severe venous invasion in the specimen removed by ER, and a hepatic metastasis was detected and resected 49 weeks later, with no recurrence 5 years after surgery. ER was less invasive treatment for T1 cancers with shorter hospital stay and less frequent serious complications compared with LC. Although the observation period was short, recurrence rarely occurred when standard management was provided. A treatment strategy to precede ER and add LC based on histological assessment is recommended.