New imaging technique could spot patients with aggressive brain tumours who would benefit from immunotherapy

• Survival for patients with glioblastoma brain tumour is 12–18 months from diagnosis.
   
• Some tumours may respond well to immunotherapy, but there is currently no way to assess response and adapt treatment without a risky biopsy.
   
• New research shows immuno-PET imaging successfully maps levels of PD-L1 in the brain – a protein inhibited by immunotherapy – meaning drug treatments can be administered and altered accurately. 
   

A new imaging technique could mean patients with aggressive brain tumours benefit from cutting edge immunotherapy treatments, according to researchers.

A team from The Institute of Cancer Research, London, has developed a new immuno-PET imaging technique that could spot which glioblastoma patients would respond well to immunotherapy, and track the response over time.

Glioblastoma is a type of brain tumour with a very poor prognosis, with an average survival time of 12–18 months – only 5 percent of patients survive beyond five years.

Research indicates that certain patients – especially those with aggressive tumours – may respond well to immunotherapy drugs, but there is currently no way to assess this without a tumour biopsy.

Due to the risks of infection and bleeding, biopsies are rarely carried out for glioblastomas prior to surgery to remove the tumour, meaning these patients are missing out on improved treatments.

Higher levels of the PD-L1 protein have been found in the rapidly progressing glioblastoma tumours. This protein acts as the brakes on the body’s immune system, so targeting PD-L1 and blocking its function could kickstart the immune system into fighting the cancer.

Until now, a biopsy has been the only way of assessing the levels of PD-L1 in the brain tumour. However, a biopsy only gives a static snapshot of the protein level in the tumour and its environment at the time it was taken. There is also a time-lag between taking the biopsy and making treatment decisions, by which time protein levels may have changed.

Due to these challenges in assessing PD-L1 levels – and therefore whether immunotherapy would work – without a biopsy, patients with newly diagnosed primary brain tumours are regularly excluded from early phase clinical trials.

The research team, funded by The Institute of Cancer Research (ICR), which is a charity as well as a research institute, developed a radiotracer – a radioactive molecule attached to an antibody – to bind to the PD-L1 protein and then measure its levels in the body of patients with glioblastoma.

In findings published in the journal Neuro-Oncology, the researchers tested their radiotracer to see how well it would bind to the PD-L1 protein on the tumour and immune cells and show up in a PET scan.

A clinical phase was conducted in collaboration with the Medical University of Silesia in Katowice, Poland, and the Maria Sklodowska-Curie National Research Institute of Oncology in Gliwice, Poland, and funded by the Medical Research Agency in Warsaw, Poland.

Eight newly diagnosed glioblastoma patients were given the tracer intravenously, followed by PET scans after 48 and 72 hours. Their PET scans showed that the tracer successfully binds to PD-L1 positive cells within the tumour and throughout the body. 

These results were compared with biopsies taken during surgery to remove the tumour, after the PET scans.

Five of the patients were randomly allocated to also receive pembrolizumab before surgery. Pembrolizumab is a monoclonal antibody that blocks the function of PD-L1 by targeting a protein it interacts with, called PD-1.

The researchers saw that these patients had lower levels of the tracer in their tumours – suggesting that the drug is acting on the PD-L1 protein and removing the brakes for the immune system so it can fight the cancer. They also saw that these patients had higher levels of the tracer in the lymph tissues, indicating that the drug was activating immune cells around the body.

Three of these five patients have seen their cancer stabilise and not grow any further. The researchers will be investigating the connection between their response to the drug, and the levels of PD-L1 in their tumour before treatment.

The clinical trial aims to recruit 36 patients diagnosed with glioblastoma to evaluate whether pembrolizumab given before surgery is effective, and whether PET imaging using the radiotracer can be used to monitor progress and adjust treatment as needed.

The researchers have also developed another radiotracer that they believe will be even more effective than the antibody used in this study. The molecule is smaller and so passes through the blood-brain barrier – the toxin-preventing membrane that protects the brain – more easily, meaning PET scans could be taken just one hour after the injection. The team hopes to test this molecule in similar studies in the future.

Dr Gabriela Kramer-Marek, Associate Professor and Group Leader in Preclinical Molecular Imaging at The Institute of Cancer Research, London, said:

“This study could revolutionise glioblastoma treatment, as we’ve shown that it is possible to image an immunotherapy target with our radiotracer. Being able to take a scan of the patient’s body and see the levels of this target means that we can predict the patients’ response, see their immune system responding to the treatment, and alter treatment where necessary – providing a personalised treatment plan based on the unique characteristics of their tumour, all without the need for a pre-surgery biopsy.

“I look forward to seeing the results of our larger clinical trial to assess how effective this immunotherapy is in glioblastoma patients – and I hope that our radiotracer will tell us more about the biology behind why some tumours are more responsive than others.”

Professor Kristian Helin, CEO of The Institute of Cancer Research, London, said:

“Glioblastoma is a devastating disease, and treatments haven’t significantly changed for decades. Although immunotherapies seem like they might be effective, progress has been stalled by not having a biomarker test to show who might benefit from them, or a way to monitor each patient’s response to treatment. These challenges mean that patients with newly diagnosed brain cancers are often excluded from early phase trials, and it may be the reason that so many brain cancer clinical trials are unsuccessful.

“It’s fantastic to see the advances in technology that mean that innovative imaging techniques could soon guide personalised treatment. This research could be part of a paradigm shift in terms of how clinical decisions for glioblastoma patients are made – from selecting patients for clinical trials, to monitoring the response to treatment in real time.”