Skip to main content

Liquid Biopsy Shows Promise for Monitoring Childhood Brain Tumors

NEW YORK (GenomeWeb) – New research by investigators at the University of California, San Francisco and the Children's National Health System, has provided early evidence that liquid biopsy testing could help doctors monitor how well treatments are working in kids with diffuse midline gliomas.

Brain cancers present a challenge for longitudinal monitoring, because obtaining repeat biopsy samples is dangerous and difficult. But liquid biopsy techniques have now opened the possibility of tracking these and other tumors over time based on analysis of tumor genetic material that is shed into the blood or other body fluids.

"At a time when neuro-oncologists are reimagining treatment … and turning to targeted therapies that match a tumor's mutations, there is a heightened need to monitor tumor growth, evolution, and response to treatment," co-senior author Javad Nazarian, a researcher at Children's National Health System and George Washington University School of Medicine, said in a statement.

Monitoring cancer using blood-borne DNA mutations has already been explored in other cancer types, including melanoma, lung, breast, and colon cancers, and some adult brain tumors. But this is the first use of the technique in a pediatric brain tumor population in a clinical trial, Nazarian wrote.

In the group's study released online in Clinical Cancer Research today, investigators reported on their analysis of blood and cerebral spinal fluid samples from 48 children diagnosed with diffuse midline glioma (DMG) — 30 CSF, 79 plasma, and one cyst fluid sample in total.

The CSF samples were all collected at a single time point either at pretreatment, during therapy, or at postmortem from 27 patients. Plasma specimens were obtained starting at diagnosis, and longitudinally throughout treatment from patients with an aggressive subtype of DMG, called diffuse intrinsic pontine glioma (DIPG), enrolled in PNOC003

Using BioRad's droplet digital PCR technology, the team created an assay to quantify levels of a mutation, H2K27M, which is known to be present in over 70 percent of patients with DMG, and which correlates with a poorer clinical outcome and is used as a criterion for enrollment in clinical trials.

The authors reported that they were able to identify H3K27M in pre-treatment samples from 42 of the 48 patients, a frequency that is comparable to what is seen in tissue samples. Cerebral spinal fluid had more mutated DNA present, overall, than blood, they added.

In a subset of children with DIPG who had liquid biopsy testing before and after radiation treatment in a clinical trial, investigators saw that the circulating tumor DNA decreased significantly, indicating that the therapy was working, and the tumor had receded. Researchers calculated that there was an agreement of 75 percent between ctDNA response (a 50 percent decrease or more) and MRI tumor volume measurements (at least 10 percent decrease).

In nine of the children with DIPG who underwent precision therapy with drugs that matched the mutations identified by the liquid biopsy testing, authors reported that this treatment led to a drop in circulating tumor DNA. Three of five patients also showed an increase in plasma ctDNA at progression.

"Physicians rely on clinical exams and MRIs to assess response to therapy, although both of these methods are limited in sensitivity and specificity," Sabine Mueller, a pediatric neuro-oncologist at UCSF Benioff Children's Hospital and a co-author of the study, said in a statement.

Subtle tumor growth, for example, may not be evident on MRI, authors argued. In addition, clear growth in an imaging scan may not actually mean a tumor is not responding, because radiation and certain other treatments can cause inflammation that masquerades as growth.

Besides just looking at levels of H3K27M, the team also explored the feasibility of multiplexed testing of other driver mutations, including mutant and wild-type alleles for oncohistone and obligate partners in ACVR1, PIK3R1, or BRAF.

The authors wrote that they believe that their proof of principle has shown CSF and plasma ctDNA analysis in children with DMG to be feasible as a tool for detecting mutational load, providing additional means for molecular disease characterization, and "most importantly" as an additional method for assessing tumor response to treatment.

The team also suggested that the method could have similar utility in other other childhood central nervous system cancers.