Nonoperative Management versus Surgery in Patients with Complete Endoscopic Response to Neoadjuvant Therapy for Rectal Cancer

Abstract

There is increasing evidence that surgery can be avoided in patients with complete endoscopic response (CER) to neoadjuvant therapy (NAT) for rectal adenocarcinoma. We compare nonoperative management to surgery in such a cohort of patients from an integrated healthcare system with the hypothesis that surgery can be avoided without compromising outcomes. We retrospectively reviewed records of surgery eligible patients completing neoadjuvant chemoradiation (nCRT) for nonmetastatic rectal adenocarcinoma at our institution from January 2015 through January 2019. Treatment consisted of external beam radiotherapy to 50-54 Gy with concurrent 5FU-based chemotherapy. Additional systemic treatment was administered at the discretion of the treatment team. CER was defined as negative digital rectal exam (DRE) and negative endoscopy (flat mucosa or scar without stricture or ulceration) at the end of NAT. Patients with CER and refusing surgery underwent strict surveillance with DRE and endoscopy. MRI was frequently incorporated into surveillance. The primary endpoint was 36-month non-regrowth recurrence rate. Non-regrowth recurrence was defined as any non-luminal intrapelvic or distant metastatic recurrence. Within the study period, 464 patients completed nCRT. Median follow-up for the entire cohort was 36 months. CER to NAT was achieved in 95 patients (20%), 42 of whom had surgery, and 53 of whom were observed. Nonoperative patients were older (median age 64 vs 57 years), were less likely to have nodal involvement (45% vs 79%), had lower tumors (median distance from the anal verge 4 vs 8 cm), and were less likely to receive additional systemic therapy (59% vs 88%). Abdominoperineal resection (APR) was anticipated in 26 (49%) nonoperative patients. In the surgery group, 21 patients (50%) had pathologic complete response. In the nonoperative group, 6 patients (11%) had local regrowth (36-month actuarial rate 15%), and all patients were locally salvageable at the time of regrowth. The 36-month rate of non-regrowth recurrence was similar between groups (surgery 8% vs observation 15%, p = 0.38). There were 5 deaths in nonoperative patients (3 non-rectal cancer deaths, 2 rectal cancer deaths). There were no deaths in surgery patients. Overall survival at 36 months favored the surgical group (100% vs 88%, p = 0.03), however rectal cancer-specific survival at 36-months was similar between groups (100% vs 95%, p = 0.16). Observation is a feasible strategy that may improve quality of life without compromising disease outcomes in carefully selected patients with CER to NAT for rectal adenocarcinoma, as most patients with local regrowth are able to be salvaged. Further studies are needed to determine the effect of nonoperative management on non-salvageable recurrences in this group of patients.