Early Potential Seen With Immunotherapy in Glioblastoma, But Questions Remain

David Reardon, MD

Initial findings from clinical trials separately exploring the safety and efficacy of immunotherapy agents nivolumab (Opdivo) and pembrolizumab (Keytruda) as potential treatments for patients with glioblastoma are promising, according to David Reardon, MD. However, checkpoint inhibition in this area requires deeper research.

"We have treatments that can help our patients, but the problem is the durability of the benefit; sooner or later the tumor acquires the ability to no longer respond to what we are treating it with and becomes resistant,” said Reardon, clinical director, Center for Neuro-Oncology, physician, Dana-Farber Cancer Institute and associate professor of Medicine, Harvard Medical School. "Durable benefit is now being seen with checkpoint blockade in metastatic melanoma, lung cancer, and even potentially in renal cell carcinoma, and I think that is a very exciting aspect of these therapies that we hope will be achieved for our brain cancer patients.”

OncLive: What data have we seen regarding immunotherapy in brain cancer that you are excited about?

In an interview with OncLive, Reardon shares the early promise for immunotherapy in brain cancer and what preliminary findings have already shown regarding checkpoint inhibition. Additionally, he sheds light on the significance of vaccines in brain cancer, and how a combination strategy that both accelerates the antitumor immune response and diminishes immunosuppression may be the most effective option.Reardon: A number of the immunotherapy trials had preliminary data presented at ASCO. The efficacy data of these studies had not matured sufficiently to be presented at ASCO, but there was some initial safety data that was presented for the CheckMate-143 study with nivolumab (Opdivo), a MEDI4736 study, and a KEYNOTE study with pembrolizumab (Keytruda). Some preliminary efficacy data were also recorded, and the main body of that is forthcoming. We might get some data by the end of this year or the first or second quarter next year.

Is there any understanding at this point how immunogenic brain cancer is?

We are very excited about these checkpoint inhibitors; PD-1 and PD-L1 targeting inhibitors. There is significant growing preclinical data using immunocompetent intracranial glioblastoma models that show that these 3 agents can have a significant benefit for animals with glioblastoma growing in the brain. It is always exciting to see preclinical validation that this intervention has some promise. It is a big step to then take it into the clinic and to validate it in patients, but at least in the preclinical models, there is very exciting data to support all of the trials that are ongoing in the clinic.For glioblastoma, the most common primary brain cancer that adults get, in general the immune infiltrate associated with it is certainly in the lower end of the spectrum of cancers. A higher degree of immune infiltrate is typically referred to as a hot tumor microenvironment, while a lesser degree of immune infiltrate is typically referred to as a cold tumor microenvironment. Glioblastoma falls in the middle there, but probably more toward the spectrum toward colder or less immune infiltrate typically seen.

Another important factor looks at the mutational burden associated with different cancers as a factor that may predict how immunogenic they might be. Glioblastoma again falls toward the middle of the spectrum of cancers in terms of average mutational burden. It is not at the low end, but not at the very high end like melanoma or lung cancer—it is kind of in between. Most of the data are pointing toward glioblastoma being in the middle range toward immunogenicity, maybe favoring the lower side. But we know much of the standard treatments that we use, particularly radiation therapy, can significantly enhance inflammation within the tumor microenvironment.

Are there other areas outside of checkpoint inhibition that you see a lot of potential for in brain cancer?

When we talk about immune activity within tumors, we have to keep in mind that this is a very dynamic process, not something like genotyping where the mutation of the tumor is analyzed and that is what it demonstrates. For immune activity, it varies over time and can be a very dynamic process. A tumor that is taken out in the operating room after the patient has radiation therapy may be very different in terms of the cytokines that are released that are driving, stimulating, or potentially inhibiting immune responses. The immune infiltrate can also change very significantly over the course of the disease and is associated with different interventions. We need to keep in mind that this is a very dynamic measurement, which can change depending on a variety of factors and influences.There is a lot of exciting work ongoing with vaccine strategies. This has been an area for 15 years that we’ve been working at, where we are trying to generate vaccines for patients with brain cancer. A variety of different centers have been heavily involved. I think a lot of optimization has occurred, as we’ve learned from earlier studies that have been done. One of the exciting vaccine approaches is the identification of tumor-specific neoantigens. These are the mutations that present within a given tumor that give rise to expressed mutated proteins that are only going to be present within the tumor cells and not present in the normal cells. We know from a lot of basic research that these coding mutations and associated mutated peptides can be very immunogenic, and we are trying to utilize these coding mutations, taking full advantage of the technology that is now currently available for high-throughput genomic sequencing analysis. The identification of coding mutations and algorithms can help predict which of the neoantigens is most likely to be immunogenic for patients. Taking advantage of all of that state-of-the-art technology is now a reality, and translating it into a patient-specific neoantigen-based vaccine is a very exciting application for vaccine approach for patients with brain cancer.

There are also a number of vaccine trials that are now in advanced phase III development, and I think the results of those trials should be reporting out in the next year or so. We are hopeful we will see some promise from those ongoing phase III studies that are nearing completion.

What challenge would you like to see tackled next in brain cancer?

The most promising thing, though, is going to require combination treatments where we can use a strategy to enhance the immunogenicity along with, or in combination with, a therapeutic strategy to decrease immunosuppression associated with the tumor. This will allow us to accelerate the anti-tumor immune responses, and at the same time diminish the immunosuppressive factors that in tumors, including in glioblastoma, tend to deploy quite promptly to protect themselves from anti-tumor responses. But we will need a combination where we have strategies trying to do both things. That is the approach that is the most likely to be successful here, and it will be required to optimize the benefit of these agents for our patients.The critical next step is to move beyond the initial single-agent studies, and based on ongoing research, identify effective combinations that have the best likelihood of being able to synergize and compliment one another. I think that is a critical area to focus on and invest in.

The area of understanding the unique aspects of tumor metabolism is also important. Much of that also impacts the immune response associated within the tumor microenvironment. There is a great rationale for thinking about combining immunotherapy treatments with treatments that impact and exploit some of the unique aspects of tumor metabolism. I think that is on the horizon and is a very exciting area.