To Evade Treatment, This Brain Cancer Mimics Neurons

Article Summary
  • Sylvester Comprehensive Cancer Center’s Dr. Antonio Iavarone and collaborators have identified how glioblastoma tumors change to mimic healthy cells.
  • The researchers looked at the tumor proteome (full set of proteins) and matched tumor samples from 123 glioblastoma patients at the time of diagnosis and recurrence after initial therapy.
  • Dr. Iavaone’s Sylvester colleague, Dr. Simona Migliozzi, then found the BRAF kinase inhibitor vemurafenib in combination with the chemotherapy drug temozolomide had an impact on the treatment-resistant tumors.

Cancer cells are really good at playing dress-up. Tumors have developed many ways to evade being killed by drugs or detected by our immune systems by disguising themselves as different kinds of healthy cells.

Illustration of cancer cell mimicking a healthy cell.
Cancer cells can evolve to make active connections with healthy neurons in the brain. (Illustration courtesy of Cancer Cell.)

Incurable brain cancer glioblastoma can mimic human neurons, even growing axons and making active connections with healthy neurons in the brain. With an average patient survival time of slightly more than a year from diagnosis, glioblastoma is so deadly in large part because it nearly always recurs after initial treatment and the recurrent tumors are always resistant to therapy.

Now, a new study from Sylvester Comprehensive Cancer Center, part of the University of Miami Miller School of Medicine, and collaborating institutions has found that this neuron mimicry seems to be essential for the cancer’s treatment resistance. The study, published Jan. 11 in the journal Cancer Cell, also identifies a class of therapeutics, BRAF inhibitors, that could prevent the tumors’ transition to drug resistance.

A Unique Approach to Glioblastoma

These findings were made possible by the researchers’ unique approach to studying glioblastoma. The research team, led by Antonio Iavarone, M.D., deputy director of Sylvester and professor of neurological surgery and biochemistry and molecular biology at the Miller School, and Jong Bae Park, Ph.D., of the National Cancer Center in Korea, used a platform they designed to study glioblastoma cells’ full set of proteins, also known as the proteome.

Sylvester Comprehensive Cancer Center researcher Dr. Antonio Iavarone
Dr. Antonio Iavarone and the the National Cancer Center’s Jong Bae Park, Ph.D., developed a platform to study glioblastoma proteins.

The researchers looked for certain modifications on those proteins that indicate enzyme activity in the cell.

“These platforms can provide you a landscape of alterations in individual tumors that you cannot get from genetics alone,” Dr. Iavarone said.

The collaborative research team assembled what is now the largest dataset of its kind—matched tumor samples from 123 glioblastoma patients at the time of diagnosis and when their cancers recurred after initial therapy.

By studying the tumors’ proteomes and protein modifications in these samples, the researchers were able to spot important changes not previously seen in similar studies of the cancer that examined the tumors’ genomes, or their transcriptomes (the set of RNA molecules in the cancer cells). Even though genes need to be converted or transcribed into RNA to result in a protein, there are additional steps after transcription that can result in different levels of proteins or different activity states. A cell’s transcriptome is not necessarily a readout of its final behavior.

Tumor Cells Change to Evade Treatment

This study is the first time scientists have used proteomics to study how glioblastomas transition from treatable to treatment resistant. By looking at cancer proteins and a specific chemical modification known as phosphorylation, the researchers were able to show that before treatment, glioblastoma cells were in a proliferative state. The cells expend their energy toward replicating themselves.

Many chemotherapies work by targeting the cellular functions involved in self-replication, as cancer cells typically grow faster than healthy cells. But once the tumors recurred in these patients months later, the cells looked very different, and much more like healthy neurons.

Sylvester Comprehensive Cancer Center researcher Dr. Simona Migliozzi
Dr. Simona Migliozzi and colleagues used a machine-learning approach to find the most active kinases in glioblastoma tumors.

There seems to be something about this replicating-to-neuronal transition that helps the cancer cells evade the normal initial course of treatment for glioblastoma, a combination of chemotherapy, radiation and surgery.

“The tumor cells actually resemble normal brain cells,” said Simona Migliozzi, Ph.D., an assistant scientist at Sylvester and one of the lead authors on the study. “Why? Because the tumor cells want to survive, they want to live, and it seems that they’re able to do this, to acquire therapy resistance, by mimicking the normal brain.”

Looking for Therapeutic Weak Points

The scientists used their new dataset to identify therapies that could kill these resistant cancers. Looking at kinases—enzymes responsible for phosphorylating other proteins—Dr. Migliozzi and her colleagues used a machine-learning approach they’d previously developed to find the most active kinases in the neuron-like glioblastoma tumors. Kinases are important for many different cellular functions and are a key target for many FDA-approved cancer drugs.

One kinase popped to the top of their list: BRAF. The gene encoding for this kinase is commonly mutated in some cancers, including melanoma. But in glioblastoma, BRAF protein levels increase without corresponding changes in the gene. The team wouldn’t have identified its importance in the brain cancer without looking at the cancer proteome.

The researchers then tested an existing BRAF inhibitor, vemurafenib, in treatment-resistant glioblastoma cells in a petri dish and in a patient-derived xenograft tumor in mice. In both cases the drug, in combination with the chemotherapy drug temozolomide, knocked down the formerly resistant tumors. In the mouse model, the BRAF inhibitor extended the animals’ survival over chemotherapy alone.

Their machine-learning algorithm to predict glioblastoma’s most active kinase could also apply to other cancers, Dr. Iavarone said. The researchers are working to develop a clinical test that would use artificial intelligence to identify therapeutic weaknesses in a variety of cancers by finding each tumor’s most active kinase and pairing it with an existing kinase-inhibitor drug.

Now, Dr. Iavarone and his colleagues are in discussions to plan a clinical trial testing vemurafenib or another BRAF-inhibitor drug for glioblastoma. Patients would need to be treated with the inhibitor from the beginning of treatment to prevent the cancer from transitioning to the resistant state.

“Proteomics gives us a much more direct prediction of the proteins’ activity,” Dr. Iavarone said. “We hope that this type of analysis can be seamlessly translated into the clinic as a next-generation precision therapy approach for this very challenging disease and for other resistant cancers as well.”

Tags: Dr. Antonio Iavarone, Dr. Simona Migliozzi, glioblastomas, Sylvester Comprehensive Cancer Center