EP103 - Local Excision of Early Stage Rectal Cancer: Long-Term Outcomes Following a 40-year Experience

Purpose/Background: Local excision of early stage rectal cancer, while boasting superior functional outcomes and lower morbidity, has been associated with higher rates of local recurrence than radical resection. However, this literature predates the advent of modern pelvic imaging that optimizes patient selection and advanced surgical techniques. We aim to examine recurrence and survival rates among patients with stage I rectal cancers undergoing local excision over the last 40 years.

Methods/Interventions: Patients with early stage (pT1-T2Nx) primary rectal adenocarcinoma who underwent local excision with curative intent at our institution between 1980 and 2020 were included. Exclusion criteria included emergent operations, inflammatory bowel disease, hereditary syndromes, neoadjuvant therapy, and proctectomy immediately following local excision. Patients were divided into four decades for subgroup analysis; 1980-1990, 1991-2000, 2001-2010 and 2011-2020. The use and type of preoperative pelvic imaging and operative technique were also compared across each time period. Primary endpoints were survival and oncologic outcomes. Multivariate cox regression analysis was utilized to determine the significance of preoperative pelvic imaging, pathologic stage, and type of surgery on outcomes.

Results/Outcomes: Of 587 patients, 339 fulfilled inclusion criteria, 39 in the first, 104 in the second, 97 in the third and 99 in the fourth decade. There was no difference in patient characteristics, but ASA scores became progressively higher with time. In the first decade, patient selection was largely clinical. Endoluminal ultrasound became more common in the second decade. Modern pelvic imaging (CT, MRI, PET-CT) became more common in the third and fourth decades (p < 0.001). Utilization of transanal endoscopic surgery increased in the fourth decade. A greater proportion of tumors in the last decade were pT1 (p=0.029). At a mean follow-up of 9.2 years, although no difference was observed in overall survival (p=0.160), nor disease free survival (p=0.160) between decades, there was a significantly lower observed rate of local recurrence in the last two decades (p=0.002). This difference remained statistically significant among pT1 tumors (p=0.006) though not pT2 tumors (p=0.202). Local recurrence rates were significantly lower among patients who underwent preoperative advanced pelvic imaging (p=0.03). Again, this difference was observed among pT1 tumors (p=0.013), though not pT2 (p=0.151). Surgical technique did not influence local recurrence rates (p=0.361). Multivariate cox regression analysis confirmed preoperative advanced pelvic imaging (p=0.009) and T1 pathologic stage (p < 0.001) were associated with significantly lower rates of local recurrence, though this trend was independent of surgical technique (p=0.850).

Conclusion/Discussion: Over the last four decades, rates of local recurrence have decreased significantly among patients undergoing local excision of early stage rectal adenocarcinoma. During this period, we observed increased adoption of preoperative advanced pelvic imaging which is also independently associated with decreased local recurrence. Thus, routine preoperative pelvic imaging has augmented patient selection to identify those most suitable for local excision and resulted in improvement in local disease control in the long term, specifically in T1 tumors. These trends appear to be independent of advances in surgical technique.

Patient Characteristics by Subgroup Decade
Local Recurrence by Pathologic Stage