When Anne Donguy’s son, Anatole, was diagnosed with diffuse midline glioma in May 2018, she joined the ranks of many parents around the world looking for any effective treatment for their child. Living in Stockholm, Sweden, Anne and her family eventually learned that the most hopeful drug treatment, Onc201, was not available in Sweden nor elsewhere in Europe. To receive it, they would have to travel to the US as part of a clinical trial.
The decision to pursue that treatment led Anne to discover a new virtual trial organization for patients with brain cancer – The Musella Foundation for Brain Tumor Research and Information, Inc which would eventually connect her son with the treatment.
“I started this organization back in 1993 and we’ve now created a patient-centric, precision oncology platform trial for patients with brain cancer where we learn from every patient,” says Al Musella, Director. The Foundation’s Brain Tumor Virtual Trial is a registry of brain tumor patients, their treatments and outcomes. To date, more than 1,500 patients are part of the registry. The purpose is to find which treatments and combinations of treatments are working the best, and which are not. Participants get access to the ongoing results.
Artificial intelligence plus molecular tumor board
The Brain Tumor Virtual Trial is an unusual combination of molecular tumor board and matchmaker through its partnerships with two organizations: Cancer Commons, which provides personalized action plans to patients with metastatic cancer, and xCures. The xCures platform transforms complex unstructured medical data directly from the patient’s medical records into structured data suitable for analysis and machine learning. The company’s AI engine then cross-references this data against the digital library of oncology data to match patients with potential treatments and predict outcomes.
“When a patient is referred to us, we get the patient’s medical records, which include scans, genetic testing results, and all their relevant diagnostic and treatment information,” says Musella. Then, a team of brain cancer experts at Cancer Commons discusses each case to make treatment recommendations. Currently, between 20-30 people rotate in and out of the trial as experts.
Outcomes inform the software
What differs with the Musella Brain Cancer Virtual Trial approach is that patient results are returned to the registry database to improve its usability. “We rely on artificial intelligence software trained on data from experts, the medical literature, and the Virtual Trials registry that analyze the smallest pieces of information about each patient’s tumor,” adds Musella. The program relies on thousands of structured data points from blood tests, MRI scans, adverse clinical events, genetic mutation testing, earlier treatments, clinical evaluations, and more.
That program generates suggestions for and against certain treatments. Many genetic testing companies already supply a similar service for cancer patients. “However, the team of experts at Cancer Commons make specific recommendations to our families. Once we have that final list of plan options, we give that to the referring doctor and patient who decide which plan to follow,” Musella says. “Sometimes the plan options are clinical trials, sometimes it’s standard treatments. But often it is a combination of treatments, sometimes experimental and sometimes off-label.” In this way, the trial has been able to consider multiple treatment options and combinations that have not been studied before.
“Everything felt overwhelming already and we felt we had no clue of what to do and where to go,” says Anne who confirms that the virtual trial approach also determined Onc201 was the best choice for Anatole. She adds, “While we already knew about Onc201 it gave us even more confidence in the approach we chose.”And if patients cannot find or get access to that treatment – as in Anatole’s case – the Brain Tumor Virtual Trials helps connect them.
“When we first got to the United States, we found that Anatole was not accepted as a suitable candidate treatment site in Texas,” recalls Anne. The Musella Foundation stepped in to help find the drug through a compassionate use program at Children’s National Hospital in Washington, DC. He has been receiving the drug for the past several years making trips from Sweden to the US every few months – even shuttling the drug overseas where he receives it with his Stockholm-based treatment team.
In working with families, the trial receives information in return about patient outcomes. This information becomes part of the artificial intelligence program, shortening the time and expense needed to evaluate treatments in such a high-risk cancer.
A neurooncologist’s perspective
As a practicing neurooncologist at Smilow Cancer Hospital and an assistant professor clinical neurology at Yale School of Medicine, Nicholas Blondin, MD, cares for many patients diagnosed with glioblastoma (GBM). Because of the nature of the disease, Blondin believes it has been difficult to make advances with therapies because the rate of enrollment in clinical trials has been low. It has been estimated that 10% of eligible patients take part in a formal clinical trial.
“In my practice I’ve seen patients with unexpected outcomes due to their individual treatment. The Brain Cancer Virtual Trials platform provides a way that this information can be collected and perhaps generate a signal across a population of patients with similar therapies that suggests a new treatment approach could be promising.I believe this virtual trial could be especially helpful for patients who do not have access to a neurooncologist or who may live in an area with limited expertise in treating brain cancers, particularly with new treatments in development. This can make them aware of treatments that their doctors may not be familiar with or if they do not meet certain inclusion criteria for a particular trial. The virtual trial platform can help both patient and doctor become aware of other therapy options. The way it is designed, it will work better with more people participating.”
A cancer learning network
The Brain Cancer Virtual Trial – working with Cancer Commons and xCures – is creating a brain cancer learning network; an attempt to gather all the possible information about brain cancer tumors, treatments, and outcomes so that none of it is lost or dispersed. “We hope to ultimately provide the best treatment options for every patient, help them get access to it, and record the results in a high-grade registry,” says Musella, who adds that the program learns from every single patient. “Ultimately, we are trying to prove that a system like this can have better outcomes than just randomly picking a clinical trial or staying with the standard of care.”
For Anatole, the virtual trial process has been life-saving. When diagnosed, his grapefruit-sized tumor housed an H3 K27M mutation which classified his as grade 4, the most aggressive. The usual patient survival rate is 10% at two years, 1% at five years. Nearly four years later, after treatment with Onc201, Anatole is a vibrant, almost 12-year-old boy, actively participating in school, after-school clubs, and otherwise enjoying his life. His tumor has been stable for about two years to about the size of a grape.
“As long as there are options, there is hope,” says Anne. “Connecting with the Musella Foundation and then Cancer Commons helped immensely to the point that our son is not only here but thriving.”
Alice McCarthy is a science writer based in Massachusetts. As a scientific storyteller with a business background, she enjoys translating research and medical discoveries into stories for a wide range of readers—from patients to physician scientists to business executives. For the past decade she has particularly focused on telling the stories of the genomic revolution in terms of new treatments and research strategies.