Risk factors and a prediction nomogram for intracranial atherosclerotic stenosis in patients with cancer-related...

To develop and validate a nomogram for predicting intracranial atherosclerotic stenosis (ICAS) severity in patients with cancer-related cerebral infarction (CRCI).

This retrospective study included 520 patients with active cancer and acute large-artery infarction. Based on CT angiography (CTA), they were classified into three ICAS severity groups (no/minimal, mild, and moderate to severe stenosis). The cohort was randomly divided into training and internal validation sets (7:3). Ordinal logistic regression was used to identify independent predictors, which were then incorporated into a nomogram. Model performance was assessed for discrimination, calibration, and clinical utility, followed by internal validation.

Multivariate ordinal logistic regression revealed five factors independently associated with higher ICAS severity: an elevated Tumor Node Metastasis (TNM) stage (III/IV) (OR = 4.907), increased neutrophil-to-lymphocyte ratio (NLR) (OR = 1.739), higher serum levels of lipoprotein-associated phospholipase A2 (Lp-PLA2) (OR = 1.029), a D-dimer concentration exceeding 0.5 mg/L (OR = 4.817), and increased age (OR = 1.071). The model showed good discrimination: C-statistic 0.817 (training), 0.804 (bootstrap-corrected), and 0.817 (validation). Calibration and decision curve analysis supported the model's accuracy and suggested potential clinical utility.

A nomogram integrating tumor staging, inflammatory, and coagulation markers was developed and internally validated as a preliminary tool for predicting ICAS severity in CRCI patients. External validation is required before clinical use.