Preoperative predictors of priority outcomes of pelvic exenteration for locally recurrent rectal cancer: Results of the mixed-methods EviSurg study

Brown KGM, Ng KS, Solomon MJ, Vu J, Lo S, Sutton PA; EvigSurg Collaborative Group; Steffens D. Eur J Surg Oncol. 2026 Jan;52(1):111183. doi:10.1016/j.ejso.2025.111183. PMID: 41240801.TOP 10%

Abstract

Background: For patients with locally recurrent rectal cancer (LRRC), the potential benefit of exenteration (long-term disease control or cure) must be weighed against the morbidity and functional consequences of surgery. Consensus-derived priority outcomes of exenteration were recently established through a comprehensive series of studies, and represent shared priorities of patients, carers, and clinicians. This study aimed to identify preoperative predictors of priority outcomes following pelvic exenteration for LRRC, specifically R0 resection, overall survival, and quality of life (QOL).

Methods: Prospectively collected data for 300 patients who underwent pelvic exenteration for LRRC between 1994 and 2023 were extracted from an institutional database. Preoperative prediction models for R0 resection, overall survival, and global QOL at 12 months (priority outcomes) were developed using logistic regression (R0 resection), Cox regression (survival), and linear regression (QOL).

Results: R0 resection was achieved in 81 % of patients (244/300). Neoadjuvant radiotherapy (OR = 2.773, 95 % CI = 1.384-5.555, p = 0.004) and the need for pelvic bone excision (OR = 2.311, 95 % CI = 1.20-4.45, p = 0.012) were associated with improved R0 resection rates. The need for sciatic nerve resection was associated with lower R0 resection rates (OR = 0.365, 95 % CI = 0.177-0.753, p = 0.006). Age was independently associated with overall survival (HR = 1.022, 95 % CI = 1.006-1.039, p = 0.007). Baseline QOL score was independently associated with a decline in QOL between baseline and 12 months (β = -1.011, 95 % CI = -1.196 to -0.825, p < 0.001).

Conclusions: These findings enable evidence-based prediction of R0 resection, survival and QOL following exenteration for LRRC. This will support development of clinical prediction tools to improve patient selection, enhance preoperative counseling, and inform shared decision-making.

Keywords: Patient priorities; Pelvic exenteration; Priority outcomes; Recurrent rectal cancer.