Assessment of circulating tumor DNA in cerebrospinal fluid by whole exome sequencing to detect genomic alterations of glioblastoma
Glioblastoma (GBM) is the most common and deadliest primary brain tumor in adults, with a median survival of just 14.6 months even with surgery, radiation, and chemotherapy. But new research suggests a less invasive way to unlock its genetic secrets—using cerebrospinal fluid (CSF) and whole exome sequencing (WES)—that could transform how doctors diagnose and treat this aggressive cancer.
Circulating tumor DNA (ctDNA) is fragmented DNA from tumor cells that floats in body fluids. For brain tumors like GBM, CSF—which directly surrounds the brain and spinal cord—is a better source of ctDNA than blood plasma, as it avoids the blood-brain barrier that limits tumor DNA in the bloodstream. Previous studies used targeted sequencing (focused on specific cancer genes) to analyze CSF ctDNA, but WES— which sequences all protein-coding genes in the genome—is more affordable and widely available. The question was: Can WES on CSF ctDNA reliably detect the genomic alterations that define GBM?
To find out, a team from Sun Yat-sen University Cancer Center and Jiangxi Cancer Hospital studied 10 patients with GBM who underwent lumbar puncture (to collect CSF) before surgery. One patient’s ctDNA was too low quality for sequencing, so 9 were included in the final analysis. CSF was collected via lumbar puncture (only clear samples, to avoid blood contamination) and stored at -80°C. Tumor tissue was taken from surgical resections and stored in liquid nitrogen. The team used standard kits to extract ctDNA from CSF and genomic DNA from tumor tissue, then ran WES on both samples. They focused on 27 GBM-associated genes—selected from databases like the Catalogue of Somatic Mutations in Cancer (COSMIC) and The Cancer Genome Atlas (TCGA)—and filtered out non-harmful mutations using tools like Polyphen2, LRT, and MutationTaster. All procedures were approved by the Medical Ethics Committee of Sun Yat-sen University Cancer Center (No. GZR2018-244) and followed the 1964 Helsinki Declaration; informed consent was obtained from all patients.
The results were promising. On average, CSF ctDNA detected more GBM-related mutations (3.56 per patient) than tumor tissue (2.22 per patient). While this difference wasn’t statistically significant—likely due to the small sample size—the frequency of mutations was almost identical between CSF and tumor samples (74.1% vs. 73.8%). Crucially, the team found mutations that were already confirmed in patient diagnoses—like the IDH1 R132H mutation (linked to better surgical outcomes) and H3F3A G34V mutation—in CSF ctDNA. These mutations are key to GBM classification and treatment: IDH1-mutant tumors are easier to resect and have better survival with maximal surgery, while H3F3A mutations are associated with aggressive pediatric and young-adult GBM.
Two patients stood out for having far more CSF mutations: one with recurrent GBM who’d received temozolomide (TMZ) chemotherapy before CSF collection (likely due to TMZ-induced “hypermutation,” a known effect of the drug) and another whose tumor had spread to the subventricular zone (SVZ)—a brain region with neural stem cells, where tumors are often more invasive. In some cases, mutations were found only in CSF: for example, a PIK3CA mutation in one patient and a PTEN mutation in another. This suggests CSF ctDNA might capture more of the tumor’s genetic diversity than a single tissue biopsy—a critical advantage, since GBM is notoriously heterogeneous (meaning different parts of the tumor have different mutations).
Why does this matter? Traditional tumor biopsies are invasive, risky, and often miss key mutations because of GBM’s heterogeneity. CSF ctDNA is a less invasive alternative that can reflect the entire tumor’s genome. For patients, this could mean earlier, more accurate genetic testing to guide treatment: for example, knowing a tumor has an IDH1 mutation might lead to more aggressive surgery, while MGMT promoter methylation status (a marker of TMZ sensitivity) could inform chemotherapy decisions. For doctors, it could reduce reliance on biopsies and speed up personalized care.
The study’s biggest limitation is its small size—just 9 patients. Larger, multi-center studies are needed to confirm the findings and see if CSF ctDNA can predict survival or treatment response. But the results are a critical first step: they show WES on CSF ctDNA is feasible for GBM and can detect clinically relevant mutations. For a disease where every month counts, this could be a lifeline.
The research was led by Hao Duan and Ji-Long Hu (co-first authors) from the Department of Neurosurgery/Neuro-Oncology at Sun Yat-sen University Cancer Center and the Department of Abdominal Surgery Oncology at Jiangxi Cancer Hospital, respectively. Yong-Gao Mou (corresponding author) oversaw the study at Sun Yat-sen University Cancer Center. The findings were published in the Chinese Medical Journal in 2020.
doi.org/10.1097/CM9.0000000000000843
Was this helpful?
0 / 0