Clinical utilisation and survival correlation of blood-based liquid biopsy in sarcoma - A systematic review and meta-analysis

BACKGROUND: Sarcomas are rare malignant tumours of mesenchymal origin that primarily affect connective tissues (soft tissue sarcomas, STS) and bone and cartilage (bone sarcoma, BS). Owing to their marked tumour heterogeneity, deep anatomical location, and the practical challenges of repeated tissue sampling, blood-based liquid biopsy represents a promising minimally invasive approach for tumour detection and longitudinal monitoring, particularly in the context of disease recurrence. This systematic review aims to evaluate the current clinical applications of liquid biopsy in sarcoma and to determine its association with survival outcomes. METHOD: We performed a systematic review to evaluate the clinical utility of liquid biopsy in sarcoma. The review was conducted in accordance with the PRISMA guidelines. Relevant literature was searched from all electronic databases MEDLINE via PubMed, EMBASE, Scopus, and the Web of Science, and data extraction was performed systematically using Covidence. Meta-analysis was performed on studies reporting survival outcomes [(overall survival (OS), disease-free survival (DFS) or relapse-free survival (RFS), and progression-free survival (PFS)]. Either fixed or random effects models were used, taking into consideration heterogeneity. Publication bias was assessed using a funnel plot. RESULTS: Nine studies were included in this meta-analysis. Copy number variation is a common mutation detected in both liposarcoma and leiomyosarcoma. Circulating cell-free DNA (cfDNA) can be detected using quantitative PCR (qPCR), droplet digital PCR (ddPCR), and next-generation sequencing. Three studies reported survival outcomes. Generally, in sarcoma, detectable (or positive) cfDNA (or specifically ctDNA) is associated with poorer overall survival (pooled HR, 1.82; 95% CI, 1.29 - 2.57; p = 0.0006) and disease-free survival (HR 2.23; 95% CI 1.27 - 3.90; p = 0.005). The progression-free survival of patients with detectable cfDNA was not significantly different from those without cfDNA (HR 1.13, 95% CI 0.72 - 1.78; p = 0.60). No evidence of publication bias was detected in this study. CONCLUSIONS: The findings of this study provide the foundation for future studies that will examine the use of cfDNA in assessing the prognosis and monitoring the relapse of sarcoma.