Clinical diagnostic utility of cerebrospinal fluid ctDNA in central nervous system lymphoma: A systematic review and meta-analysis.

Abstract

e15067 Background: Diagnosis of central nervous system lymphoma (CNSL), including primary (PCNSL) and secondary (SCNSL) forms, is challenging due to the limited sensitivity of conventional cerebrospinal fluid (CSF) studies. Cerebrospinal fluid circulating tumor DNA (ctDNA) analysis has emerged as a minimally invasive liquid biopsy that may complement standard diagnostic approaches. We performed a systematic review and meta-analysis to evaluate the diagnostic performance of CSF ctDNA in CNS lymphoma. Methods: PubMed, EMBASE, Scopus, and Web of Science were systematically searched for studies evaluating ctDNA detection in CSF from patients with suspected or confirmed CNS lymphoma. Random-effects meta-analyses of proportions were conducted, with prespecified subgroup analyses for PCNSL and SCNSL. Comparative analyses of CSF versus plasma ctDNA detection were performed when available. Risk of bias was assessed using QUADAS-2. Results: Fourteen studies including 370 CSF samples were analyzed. The pooled ctDNA detection rate in CNS lymphoma was 0.70 (95% CI 0.63–0.75; I² = 1.8%). Detection rates were similar in PCNSL (0.68, 95% CI 0.55–0.79; I² = 31.2%) and SCNSL (0.71, 95% CI 0.55–0.83; I² = 0%). CSF ctDNA showed significantly higher detection compared with plasma-based testing (OR 12.54, 95% CI 1.02–154.41; I² = 72.3%). Most studies showed moderate risk of bias, mainly related to patient selection, while index test domains demonstrated low risk. Conclusions: CSF ctDNA analysis shows consistently high diagnostic performance in CNS lymphoma but should not be interpreted as a standalone diagnostic test. When integrated with conventional modalities, it represents a valuable adjunctive tool, particularly in clinically challenging cases. Prospective studies with standardized diagnostic frameworks are needed to define its optimal clinical role.