Poliovirus for Glioblastoma Grabs National Attention

Get Permission

Matthias Gromeier, MD

What is remarkable is that we get these types of responses with a single administration of virus without further therapy.

—Matthias Gromeier, MD

Researchers at The Preston Robert Tisch Brain Tumor Center at Duke University are being barraged by patients wishing to enroll in their clinical trial of an engineered poliovirus for recurrent glioblastoma. This comes as a result of a CBS 60 Minutes interview with lead researcher Matthias ­Gromeier, MD, Associate Professor of Surgery, Molecular Genetics and Microbiology, and Medicine at Duke.

The results he shared engendered hope within the glioblastoma community, and understandably so. When glioblastoma recurs, median survival is 6 to 8 months. Using the engineered poliovirus (PVS-RIPO), Dr. Gromeier and his team have extended survival beyond 3 years in some patients.

“What is remarkable is that we get these types of responses with a single administration of virus without further therapy,” said Dr. Gromeier.

Dr. Gromeier recently discussed the approach at the American Society of Gene and Cell Therapy (ASGCT) Annual Meeting in New Orleans.1 He also spoke with The ASCO Post.


PVS-RIPO is a live attenuated, oral serotype 1 poliovirus (Sabin) vaccine that contains a heterologous internal ribosomal entry site from human rhinovirus type 2. The vaccine recognizes nectin-like molecule-5 (CD155), the poliovirus receptor and a confirmed tumor antigen that is widely expressed in solid malignancies.

Dr. Gromeier explained the rationale for this therapy, which has been evaluated so far in 23 patients, with patients continuing to enroll in the phase I trial. He said that researchers are interested in the poliovirus because of its tropism. Simply put, the poliovirus attacks tumor cells, leading to cell lysis (which does not need to be extensive) and then presentation of tumor antigen from these lysed cells.

“Poliovirus induces a form of cell death that does not resemble any orchestrated form of demise (such as apoptosis, necrosis, or necroptosis). It’s a highly unusual, highly danger-signaling form of death,” he indicated. “At the same time, it elicits antiviral immunity or interferon response, so the combining of the danger signals and pathogen signals helps to recruit cytotoxic T cells to recognize these tumor antigens.… What you need is activation of innate antiviral response plus cell killing.”

In its very fast-paced life cycle, the poliovirus behaves more like a toxin than a virus, ultimately producing viral polypeptides that exert a toxic effect. Events conspire, he believes, “to set up a classic immune effector response against the tumor that we believe is protecting some of our patients durably.”

Details of Treatment

PVS-RIPO is delivered intratumorally via a form of convection-enhanced delivery, which takes advantage of the natural flow of the cerebral spinal fluid and delivers drug by a stereotactically placed catheter directly into the tumor. Once in the brain, the virus triggers the immune response.

“The true objective of [convection-enhanced delivery] is to cover vast areas of the brain, but in this case, it’s not what we necessarily need,” he explained. “We just want to inject the virus in a way that the tumor is exposed to some degree, so we deliver 3 mL of the virus over about 6 hours.”

At the Society for Neuro-Oncology Annual Scientific Meeting in Miami this past November, Dr. Gromeier’s team reported phase I dose-finding and safety results on 15 patients.2 Median survival was 15.2 months, and the 12-month survival rate was 70%; 18- and 24-month survival rates were 43.8% and 29.2%, respectively. Serious adverse events included hemiparesis, intracranial hemorrhage at catheter removal, lymphopenia, seizure, and hyperglycemia, as well as lethargy, headache, diarrhea, paresthesia, and hyperbilirubinemia.

Response and Survival

At the ASGCT meeting, Dr. Gromeier elaborated on these outcomes. “We have very good survival and complete radiographic responses,” he said, including complete responses in patients whose tumors could not be completely resected.

“The first patient we treated had a very large tumor excised. She got a single PVS-RIPO infusion, and 2 months later the tumor was significantly enlarged. The [magnetic resonance imaging (MRI) scans] looked awful, but the patient was doing fine,” he recalled. “By 6 months, we saw glial scarring. By 11 months, the tumor had significantly shrunk. And by a year or so, the tumor was gone.”

He added, “Three years out, the patient is tumor-free, she’s working, she’s completely fine. She has to take anticonvulsants for the rest of her life, but otherwise you would never know this patient had a brain tumor.”

Another patient’s MRI at 3 years is essentially normal. “He recently delivered a half-hour talk at a fundraiser. He’s 74 now, and you would not know that this man had a recurrent [glioblastoma],” he reported.

Lessons Learned

“So what can we learn from our trial? We started out really, really positive, and then things got difficult,” he noted.

At the time of the ASGCT meeting, two patients were alive past 3 years and another past 2 years, all of whom were complete responders to the infusion and not steroid-dependent. On the other hand, patients on high-dose steroids before PVS-RIPO treatment had worse outcomes. This led the researchers to look at the effect of steroids on the treatment.

By observing differences related to prior steroid use and by realizing they created steroid dependency when using the high doses of PVS-RIPO, the researchers concluded that both steroid use and dose of PVS-RIPO were very important to outcomes. They amended their enrollment criteria to exclude patients on high-dose steroids and lowered the treatment dose.

“We can’t leave out steroids entirely because we use a neurosurgical procedure to administer the virus, and the surgeons must give some steroid for that to protect the brain from swelling,” he explained. “But we try to taper the patient off within a week or so. We believe that high doses of steroids are inappropriate because they are too immunosuppressive.”

The investigators are now treating with “dose level minus 1,” which is 5 × 10e7 (50 million infectious units) and half the initial starting dose. They have been able to successfully taper steroids in eight patients treated with this dose.

Dr. Gromeier emphasized in the interview that it is rare to settle on the lowest dose in a dose-finding study. “In cancer, it’s always ‘more is better.’ But the immune system does not work on dose. The reason the high dose doesn’t work is not just due to toxicity. I believe we get the wrong kind of response.”

On the lower dose, he reported, “patients are doing well on this therapy, so we just keep on watching what’s happening.” He acknowledged that one patient died, “which reminds us this is a terrifying disease.”

Possible Mechanism

As a virologist, Dr. Gromeier attributes the therapeutic benefit to eliciting an innate human response to polio, setting up a classic effector immune response against the tumor, stemming from a complex interplay of activated macrophages, dendritic cells, cytokines, and “superfast” proinflammatory cell killing.

“I think the most damaging proposal in the oncolytic virus field has been that tumors lack innate immune responses. I don’t believe that’s true. Plenty of evidence says the opposite. I actually believe the innate response in the tumors themselves plays a very important role in what we’re doing,” he maintained.

Since his interview on 60 Minutes aired, he said, “We have received thousands of calls, and we answer all of them.” All patients are considered; however, given that this is a phase I safety study, the eligibility criteria are quite narrow, he indicated.

While the researchers are very encouraged by preliminary data, they acknowledge they need to wait at least another year and a half “to be confident.” ■

Disclosure: Dr. Gromeier reported no potential conflicts of interest.


1. Gromeier M: Cancer Immunotherapy: Cutting Edge – Clinical. Scientific Symposium. 2015 American Society of Gene and Cell Therapy Annual Meeting. Presented May 14, 2015.

2. Desjardins A, Sampson J, Peters K, et al: Final results of a phase 1 trial of an oncolytic polio/rhinovirus recombinant (PVSRIPO) against recurrent glioblastoma (GBM). 2014 Society for Neuro-Oncology Annual Scientific Meeting. Abstract AT-21.