Answers are grounded in this paper and cite their sources.
Background Microvascular invasion (MVI) is an adverse pathological feature of hepatocellular carcinoma (HCC) that is usually confirmed only after resection. We examined whether the preoperative Glasgow Prognostic Score (GPS) and modified Glasgow Prognostic Score (mGPS) were associated with pathological MVI in early-stage HCC. Methods This single-center retrospective cohort included 533 patients who underwent curative-intent hepatectomy between January 2020 and December 2025. Logistic regression evaluated GPS and mGPS categories with sequential covariate adjustment. Influence diagnostics and leave-one-out refitting assessed dependence on individual score-2 patients. Exploratory receiver operating characteristic analysis compared a clinical covariate model with models adding GPS or mGPS. Results Pathological MVI was present in 212 patients (39.8%), including 30 of 50 patients (60.0%) in the common GPS 2/mGPS 2 category. In fully adjusted models, GPS 2 (odds ratio [OR] 2.46, 95% confidence interval [CI] 1.18–5.12; P = 0.017) and mGPS 2 (OR 2.47, 95% CI 1.19–5.12; P = 0.015) were associated with MVI, whereas score-1 categories and adjusted trend tests were not significant. The clinical covariate model had an area under the curve (AUC) of 0.669; AUCs after adding GPS or mGPS were both 0.686 (both paired P = 0.099). Conclusions GPS 2 and mGPS 2 were associated with pathological MVI, but the associations were not graded across score levels and produced only small, non-significant improvements in discrimination. These associative findings do not establish causality or support GPS or mGPS as standalone MVI prediction tools; external validation is required.
No takes yet. Share an insight, caveat, or question.
Cao et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: