Artificial intelligence–enabled liquid biopsy in cancer: a systematic review and meta- analysis of diagnostic performance and biological implications
By
Luisana Sisca
Mariam Grazia Polito
Emy Sisca
Michele Iuliani
Alessio Cortellini
Bruno Vincenzi
Giuseppe Tonini
Francesco Pantano
June 17, 2026
Clinical Scorecard: AI-Enhanced Liquid Biopsy for Cancer: A Comprehensive Review and Meta-Analysis of Diagnostic Efficacy and Biological Significance
At a Glance
Category Detail
Condition Cancer detection and monitoring Key Mechanisms Integration of AI to analyze circulating biomarker data Target Population Patients undergoing cancer diagnosis and monitoring Care Setting Oncology
Key Highlights
Pooled AUROC of AI-enhanced liquid biopsy is 0.924 (95% CI, 0.879–0.953) AI models show an absolute AUROC improvement of 0.025 over conventional methods Substantial between-study heterogeneity (I² = 88.8%) observed AI enhances the ability to extract diagnostic signals from complex biomarker data Liquid biopsy provides a minimally invasive alternative to traditional tissue biopsy Guideline-Based Recommendations
Diagnosis
AI-enhanced liquid biopsy should be considered for oncologic diagnosis. Management
Integration of AI models in clinical decision-making for molecular diagnostics is suggested. Monitoring & Follow-up
Liquid biopsy can be used for real-time monitoring of tumor evolution. Risks
Low signal abundance and biological noise may limit diagnostic performance. Patient & Prescribing Data
Patients with various cancer types including NSCLC, colorectal, breast, and hepatobiliary cancers
AI models can improve sensitivity and specificity in cancer detection.
Clinical Best Practices
Utilize AI models to analyze high-dimensional datasets in liquid biopsy. Ensure rigorous validation of AI models before clinical implementation. Consider the impact of pre-analytical variability on liquid biopsy results. Related Resources & Content
