Rubbia-Brandt vs Blazer scores for survival prediction in colorectal liver metastases after chemotherapy

<ns3:p> Background Pathological tumor regression after neoadjuvant chemotherapy is a major prognostic factor in colorectal cancer liver metastases (CRCLM). However, the comparative predictive value of existing histological response scoring systems remains unclear. Methods We retrospectively reviewed 70 patients with stage IV CRCLM who underwent liver resection after neoadjuvant chemotherapy between 2015 and 2021. Pathological response was assessed using two scoring systems: the five-tier Tumor Regression Grade (TRG) by Rubbia-Brandt and the three-tier system proposed by Blazer. Survival analyses were performed using Kaplan–Meier curves, log-rank tests, and Cox regression. Prognostic performance was evaluated through likelihood ratio χ <ns3:sup>2</ns3:sup> tests, trend tests, and area under the ROC curve (AUC) with 95% confidence intervals (CI). Results The median follow-up was 32 months. Median overall survival was 40.1 months in patients with TRG 1–2, 32.1 months for TRG 3, and 18.5 months for TRG 4–5 (p = 0.03). The Rubbia-Brandt TRG score showed good prognostic discrimination with an AUC of 0.80 (95% CI: 0.69–0.90). In contrast, the Blazer score showed no statistically significant survival difference among categories (p = 0.26) and yielded a lower AUC of 0.60 (95% CI: 0.48–0.73). Both the likelihood ratio χ <ns3:sup>2</ns3:sup> (7.41 vs. 2.68) and trend test (p = 0.03 vs. p = 0.09) favored the Rubbia-Brandt system. Conclusions In this series, the Rubbia-Brandt TRG score demonstrated superior prognostic performance compared to the Blazer score for assessing survival after liver resection for CRCLM. These findings support the integration of pathological response scoring into postoperative risk stratification, in combination with classical staging elements. Further validation in larger prospective cohorts is warranted. </ns3:p>