Liquid biopsies may transform brain cancer care through exosome analysis

It was almost serendipity, says Leonora Balaj, PhD, Assistant Professor of Neurosurgery at Massachusetts General Hospital and Harvard Medical School. Earlier this year, she and her colleagues reported a new liquid biopsy approach to monitor the effects of CAR-T cell therapy in patients with glioblastoma.

Glioblastomas, a deadly type of brain cancer, account for about 50% of all malignant brain tumors, according to the US National Brain Tumor Society, and patients have an average 7-year survival rate of less than 7%.

The current gold standard to diagnose solid tumors is a tissue biopsy. In addition to being invasive, however, it only provides a sample at a single point in time.

Many studies have shown that tumors, including gliomas, shed biological material that can be detected in biofluids. In brain cancer in particular, much research has focused on detecting DNA and RNA biomarkers of brain cancer in cerebrospinal fluid (CSF).

But to obtain the fluid, clinicians must perform a spinal tap – a procedure that requires considerable skill and is uncomfortable for patients. By contrast, diagnostic tests that sample biofluids such as blood present a minimally-invasive method of repeat sampling. They can be used not only to diagnose cancer, but also to monitor the effectiveness of therapy.

Exosomes have recently been discovered to carry biomarkers from such brain tumors. Balaj and her team measured in blood, levels of Epidermal Growth Factor variant III (EGFRvIII) mRNA in extra cellular vesicles (EV) also referred to as exosomes.

How are exosomes related to cancer?

Exosomes are nanometer sized packets of cellular components (incl. proteins and nucleic acids), enclosed in a lipid membrane, released by many types of cells, including tumor cells. They were previously thought to ferry waste products. Recent research, however, suggests that they play a role in cellular communication. In cancer, they may participate in tumor metastasis.

In several recent studies, Balaj and her colleagues have shown that RNA from EVs can reliably predict brain cancer patients’ response to treatment (1,2). Now the team has shown that it is possible to measure levels of EGFRvIII RNA from exosomes in blood to monitor how well glioblastoma patients are responding to CAR-T cell therapy. This has never been seen before.

“A blood test provides an easy way to detect and track a tumor, paving the way for testing new therapeutics in this population,” says Balaj. “It eliminates the need for repeat surgeries to monitor the status of a patient’s cancer over time.”

The finding that EGRFvIII RNA in blood served as a proxy for the success of CAR-T cell therapy was unexpected, says Balaj. The primary goal of the trial was to test whether the CAR-T cell infusions would be effective against this difficult to treat cancer – the biomarker aspect was an aside. “But it turned out to be extremely exciting because the data was so clear and robust.” When the trial investigators saw Balaj’s results, “everyone was blown away,” she says.

The trial is ongoing and Balaj and her colleagues continue to analyze blood and CSF from patients. And although early results from the CAR-T infusions were positive – patients’ tumors shrank dramatically and one patient’s tumor practically disappeared – two of the patients’ cancer returned after several weeks.

The glioblastoma

Glioblastomas are tumors of glia – central and peripheral nervous system cells that support and protect neurons. Because of poor treatment options, researchers at the Massachusetts General Hospital Cancer Center chose to study CAR-T cell therapy – a type of immunotherapy – in glioblastoma patients with EGFRvIII mutation, the deadliest form.

To be detectable in blood, the tumor-specific EGFRvIII variant mRNA must first cross the blood-brain barrier (BBB), which is a significant challenge. Presumably, the mRNA is shuttled across the BBB within EVs. Although several groups have previously claimed that brain-derived EVs can reach the blood, the evidence in this case is particularly compelling. Not only is EGFRvIII a tumor-specific variant, but the timing of its detection in the blood strongly suggests that it was released within the brain, where the treatment was performed, and subsequently made its way into the bloodstream.

In a recent study published in the New England Journal of Medicine (4), Balaj and her colleagues showed that the levels of EGFRvIII RNA in the CSF and blood of patients with glioblastoma undergoing CAR-T cell treatment not only correlated with each other, but also with the effects of treatment. Most surprising was the fluctuation in levels of EGFRvIII, an effect that appeared to give information about the cancer, she says. 

“At first we saw an increase in the EGFRvIII copy number,” she explains. “I didn't really know what to make of that.” But she later understood that it was likely a reflection of the tumor releasing its cargo, she says, “which correlates with tumor cell death and has been shown for other cancers.”

After the initial increase, levels of EGFRvIII RNA dropped over the ensuing weeks of treatment to a point where it could barely be detected. In addition, the signature of RNA in blood correlated with the signature found in CSF over time, but in blood it was slightly delayed. This might have something to do with the rate of clearance, she hypothesizes; the body might clear tumor cells in CSF before the effect shows up in blood.

The study showed for the first time that it is possible to monitor response to CAR-T cell treatment in glioblastoma patients using exosomes extracted from blood. Moreover, it provided validation that blood, which is much easier to collect, can be used instead of CSF, Balaj says. 

Finding the potential for therapeutics

Balaj and her colleagues’ protocol takes advantage of QIAGEN’s exoRNeasy kit for extracting RNA from exosomes and other EVs in blood plasma and highly sensitive digital PCR for RNA analysis. The researchers reported an overall sensitivity of 72.8% and a specificity of 97.7% for detecting EGFRvIII in plasma compared with tumor tissue.

Many variables can affect the quality of the patient samples in a clinical setting that are difficult to control, Balaj says. QIAGEN’s exoRNeasy kit is the only one on the market “that provides a robust and reproducible way of doing exosome analysis and delivers results within 2 hours,” she says.  

Another important aspect, she adds, is that QIAGEN’s kit adheres to the US Food and Drug Administration’s Good Manufacturing Practice Guidelines. It has thus already passed a major hurdle towards being validated by the Clinical Laboratory Improvement Amendments of 1988.  Such validation is a crucial step companies must take before an assay can be commercially marketed.  

“Our goal is to develop a method that could be used routinely in clinics,” says Balaj. “QIAGEN’s exoRNeasy kit can help us reach that goal.” 

ExoRNeasy utilizes a column-based procedure to isolate RNA from inside extracellular vesicles, such as exosomes from bodily fluids like CSF and plasma. This method significantly increases standardization and reproducibility over traditional techniques like ultracentrifugation. It involves affinity membrane binding technology, which ensures high specificity and excludes non-vesicular protein complexes, leading to cleaner preparations.

QIAGEN has also performed in house research to optimize preanalytical workflows. For successful RNA analysis, it’s important to avoid blood collection tubes that contain heparin, says Martin Schlumpberger, PhD, Director of Product Development at QIAGEN, which is very difficult to exclude and acts as a potent inhibitor of enzymatic assays . A fresh sample is also key, he adds. The longer a sample sits around, the more EVs will be released by blood cells, effectively diluting tumor-derived analytes.