ctDNA-Based Liquid Biopsy of Cerebrospinal Fluid in Brain Cancer

The optimal treatment and management of patients with brain cancer depend on the molecular characteristics of their tumour. Since the tumour changes with time, it is, therefore, essential to characterise the tumour of each patient at the exact time of selecting the most suitable therapeutic strategy. However, obtaining a tumour biopsy for its characterisation is a risky and invasive procedure and, sometimes, not even feasible, leading to a lack of information about the tumour. These challenges can be overcome by using a liquid biopsy of cerebrospinal fluid. Brain cancer cells release DNA into the cerebrospinal fluid, and the analysis of the cell-free circulating tumour DNA can reveal the genetic profile of brain cancer in a relatively noninvasive manner. In this review, we revise the recent results in this field that show how circulating tumour DNA in cerebrospinal fluid can provide diagnostic and prognostic information, identify potential therapeutic targets, monitor the tumour response or resistance to treatment, and help to identify tumour relapse. The correct characterisation of central nervous system (CNS) malignancies is crucial for accurate diagnosis and prognosis and also the identification of actionable genomic alterations that can guide the therapeutic strategy. Surgical biopsies are performed to characterise the tumour; however, these procedures are invasive and are not always feasible for all patients. Moreover, they only provide a static snapshot and can miss tumour heterogeneity. Currently, monitoring of CNS cancer is performed by conventional imaging techniques and, in some cases, cytology analysis of the cerebrospinal fluid (CSF); however, these techniques have limited sensitivity. To overcome these limitations, a liquid biopsy of the CSF can be used to obtain information about the tumour in a less invasive manner. The CSF is a source of cell-free circulating tumour DNA (ctDNA), and the analysis of this biomarker can characterise and monitor brain cancer. Recent studies have shown that ctDNA is more abundant in the CSF than plasma for CNS malignancies and that it can be sequenced to reveal tumour heterogeneity and provide diagnostic and prognostic information. Furthermore, analysis of longitudinal samples can aid patient monitoring by detecting residual disease or even tracking tumour evolution at relapse and, therefore, tailoring the therapeutic strategy. In this review, we provide an overview of the potential clinical applications of the analysis of CSF ctDNA and the challenges that need to be overcome in order to translate research findings into a tool for clinical practic