Abstract Background Circulating tumour DNA (ctDNA) isolated from cerebrospinal fluid of children with medulloblastoma has been shown to be prognostic for recurrence1. However, obtaining CSF is invasive and often not feasible. It is generally accepted that ctDNA cannot be detected in plasma using conventional assays. Sensitive plasma-based ctDNA analysis could enable less invasive disease monitoring. Aims We aimed to develop a highly sensitive plasma-based ctDNA assay for brain cancer patients enrolled in the ZERO Childhood Cancer Program. Methods Our personalised ctDNA assay used paired tumour-normal whole genome sequencing data from each individual to design a panel of ∼400 patient-specific tumour biomarkers. We detected ctDNA in plasma down to a tumour fraction of 0.005%. We evaluated 63 patients treated for brain cancer who had plasma collected through the ZERO precision medicine program. Diagnoses included embryonal brain cancers (n = 22); high-grade glioma (n = 30); low-grade glioma (n = 5); ependymoma (n = 8); choroid plexus carcinoma (n = 2); and CIC-rearranged CNS sarcoma (n = 1). For embryonal brain cancer patients, we assessed the survival association between ctDNA positivity at i) 3 months from treatment commencement (“early response”) and ii) the end of treatment. Results Fifty patients (79%) had positive plasma ctDNA, with variable clearance. In embryonal brain cancer, end of treatment ctDNA positivity correlated with progression-free (p = 0.011) but not overall survival (p = 0.18, n = 18 evaluable). Early response ctDNA did not correlate with either progression-free or overall survival (p 0.05). In high-grade glioma, plasma ctDNA was detected in 10/20 patients (50%) with confirmed progression (9/10 at 0.4% tumour fraction), whereas 3 patients with pseudo-progression did not have detectable ctDNA. Conclusion We have developed an ultra-sensitive plasma ctDNA assay capable of detecting brain tumour DNA in peripheral blood. End of treatment plasma ctDNA positivity in embryonal brain tumours predicted subsequent recurrence, supporting its potential role in guiding post-therapy management. Validation in larger, uniformly treated cohorts is planned. 1. Liu, A. P. Y. et al. Serial assessment of measurable residual disease in medulloblastoma liquid biopsies. Cancer Cell. 2021; 39: 1519-1530
Mateos et al. (Tue,) studied this question.