The impact of ERAS strategy implementation on long-term treatment outcomes in patients with stage III and high-risk stage II colorectal cancer

Abstract

Objectives: Patients with stage III and high-risk stage II colorectal cancer (CRC) require both curative surgery and adjuvant chemotherapy (ACT). Though ERAS strategy influences the optimal timing and total amount of patients undergoing ACT, evidence clarifying the impact of ERAS on oncologic outcomes is lacking. Methods: Between 2008 and 2014, 330 patients with primary resectable stage II-III CRC were enrolled in the ERAS arm (N=130) and traditional care (TC) arm (N=200). Results: Overall postoperative grade II-IV complications (PCs) rate was 7.8% and 21.6% for ERAS and TC arm (p<0.01), grade V – 1 (0.8%) vs. 7 (3.5%) (p<0.05). All patients who did not have PCs showed significant OS benefit – 85.8±4.1% and 81.7±4.5% vs. 53.3±17% and 45.7±9.6% (p<0.05). In the ERAS arm, 87 patients (67%) started ACT in optimal terms: within 4 weeks (w) - 66 (51%), 8w – 21 (16%). 29 (22%) failed to undergo ACT and 14 (11%) had ≥8 week delay. In the TC arm, 104 (52%) failed, 30 (15%) had no delay, 30 (15%) – 4 week delay, and 36 (18%) – ≥8 week delay. Multivariate analysis in the ERAS arm had confirmed PCs as an independent prognostic factor for delay of ACT ≥ 4 weeks (p=0.029, HR=0.399), but not failure; in TC arm – both delay and failure. In the ERAS arm, 3-year OS had no significant difference for those who underwent ACT within 4 and 8 weeks – 72.3±12.2% and 58.3±10.2% (p=0.320). However, patients with no-use and ≥8 week delay had significantly worse OS and DFS rates (29.6±32.2% and 30.5±26.8%) compared with ACT start within 4 weeks (p=0.023). No significant difference in the TC arm was observed in OS for those who started ACT within 4-8 weeks and with ≥8 week delay (64.2±19%, 52.5±20%, 46.1±9%), however, a time-dependent trend towards OS and DFS improvement was noticeable. No use of ACT was strongly associated with poor OS (18.2±16%). No significant difference in DFS was observed in both groups, except for no-use of ACT. Cox regression analysis defined ≥8 week delay and no-use of ACT as independent risk factor for OS and DFS deterioration (p<0.001, HR=1.639) in ERAS arm and PCs (p=0.021, HR=0.615) – in TC arm. Conclusion: ERAS strategy had a significant impact on long-term treatment outcomes for stage III and high-risk stage II CRC. Disclosure of interest: None declared.