Impact of relative hospital volume on locoregional recurrence and survival in patients with complex colon cancer

The association between hospital volume and long-term outcomes in colon cancer remains debated, but may be particularly relevant for complex colon cancer (CCC), such as cT4 tumors and those with infectious tumor-related complications. This study assessed the impact of relative hospital volume on locoregional recurrence (LRR) and overall survival (OS) in patients with CCC. In this retrospective cohort study, patients from 48 hospitals in the Netherlands who underwent resection for non-metastatic colon cancer between January 2014 and December 2015 were selected. Hospitals were classified into quartiles based on percentage of CCC resections of their annual volume (low Q1, medium Q2–Q3, high Q4). Primary outcome was 5-year LRR. Secondary outcome was 5-year OS. Among 8164 patients, 977 had CCC (12.0%). Low-volume comprised < 11.7% CCC resections, medium-volume 11.7%–18.3% resections, and high-volume > 18.3% CCC resections. Five-year LRR rate after primary resection of CCC was significantly lower in high-volume hospitals: high-volume 18.5% versus medium-volume 27.9% and low-volume 30.7% (Gray’s test p = 0.006). Treatment in high-volume hospitals was independently associated with a reduced risk of LRR compared with low-volume hospitals (HR 0.60; 95%CI 0.47–0.78). In a sensitivity analysis using a more restrictive definition of LRR, this association was attenuated (HR 0.87; 95%CI 0.77–0.98). Five-year OS was highest in high-volume hospitals (67.1%), but not significantly different from low-volume (56.0%) and medium-volume hospitals (58.2%) (log-rank p = 0.119). Treatment at hospitals with a high relative volume of CCC resections was associated with a lower risk of LRR, while OS did not differ significantly across hospital volumes. These findings suggest that dedicated tertiary care for complex cases, rather than increasing overall colon cancer volume, may play a role in lowering the risk of locoregional recurrence in CCC.