Recurrent high-grade glioblastoma has a poor prognosis, with a median overall survival of 6 to 9 months. Treatment is limited, partly because immunotherapy has not yet been shown to be effective in the immunosuppressive microenvironment of this tumor. A novel treatment approach involving drug-inducible, regulatable, interleukin-12 (IL-12) gene therapy showed promise in a recent phase I study. The findings suggest that production of human IL-12 within the central nervous system can be controlled using a gene transcriptional “switch.”
The phase I findings were recently reported in Science Translational Medicine.1The ASCO Post talked with one of the study authors, Rimas V. Lukas, MD, Associate Professor of Neurology at Northwestern University School of Medicine, about this exciting novel therapy.
Early Results of Phase I Trial
Although this was a small phase I study, the results were impressive. What did you learn?
Our multicenter open-label dose-escalation phase I trial assessed the safety and biologic effects of a human IL-12 gene that was transcriptionally regulated by an oral activator (veledimex) in 31 patients undergoing resection of recurrent high-grade glioma.
After the tumor was resected, the cavity was injected with a fixed dose of a vector, Ad-RTS-hIL-12 (an adenoviral serotype 5 vector encoding the human IL-12 p70 transgene). Preoperatively, we gave one dose of veledimex (10 mg, 20 mg, 30 mg, and 40 mg) and then gave it postoperatively for 14 days.
We think the signal is strong enough to move forward with additional studies that would validate the efficacy [of veledimex and Ad-RTS-hIL12 in recurrent glioblastoma].— Rimas V. Lukas, MD
Tweet this quote
Median overall survival was 12.7 months. The survival rate in the 20-mg cohort was 60% at 12 months, 26.7% at 18 months, and 13.3% at 24 months. One patient receiving 40 mg was alive at the time of data cutoff, around 30 months. We think these preliminary results are encouraging and promising, and the treatment tolerability was acceptable. All adverse events were reversible after veledimex discontinuation.
The caveat is that this is not a randomized study, so we want to make sure the optimism we have is cautious. We think the signal is strong enough to move forward with additional studies that would validate its efficacy. We’ve identified 20 mg as the dose associated with the best survival outcome and optimal tolerability.
Most patients had subsequent treatment, starting within a few months of your treatment. Could this have confounded overall survival as an outcome?
That’s always a concern with regard to survival results in patients who receive subsequent therapies, but it becomes less of an issue in glioblastoma because we don’t have a standard treatment that definitely works. So, I think it’s less of a concern than it may be in other diseases.
The use of dexamethasone seemed to have a negative effect. What are your thoughts on that?
In patients who cumulatively received more than 20 mg of dexamethasone (days 0–14), median overall survival was 6.4 months, compared with 16.7 months in patients receiving up to 20 mg. In the optimally dosed cohort—patients receiving 20 mg—median survival was 17.8 months in those who did not receive steroids. There’s a strong belief that dexamethasone can abrogate the efficacy of immunotherapy, and that may be happening here. I think it’s going against dogma, or tradition, not to give steroids to patients with brain tumors, but I don’t think we necessarily have to. I think neuro-oncologists and neurosurgeons are able to manage clinical symptomatology and avoid or substantially minimize their use.
Pseudo-progression and Cytokine-Release Syndrome
In your study, you saw some cases of pseudo-progression. Could this be a problem in the clinic?
There is certainly a concern about pseudo-progression, which was seen in some patients. However, this isn’t a surprise, as neuro-oncologists are used to seeing pseudo-progression—up to one-third of patients on standard radiation/temozolomide treatment experience it. In turn, we are comfortable with that phenomenon. We just need to be cautious and avoid stopping or switching therapy in these patients; instead, we should continue with the current treatment plan or possibly consider biopsy or resection to better assess what’s going on. The potential for pseudo-progression is not a reason to be overly concerned.
You saw some cases of systemic cytokine-release syndrome. Is this worrisome?
The frequency of grade 3 cytokine-release syndrome was 12.5% in the 20-mg (optimally dosed) cohort, and 25% and 50% in the higher-dose cohorts. It was reversed upon holding or terminating veledimex.
One could extrapolate that if our approach is successful in glioblastoma, it may also be successful in other infiltrating gliomas.— Rimas V. Lukas, MD
Tweet this quote
Cytokine-release syndrome is definitely a concern with this treatment, but what’s good is that it’s predictable. Patients have the most notable symptomatology shortly after their resection. Day 3 or 4 is where they seem to be the “rockiest,” and that improves over time. We know the time frame, so we can monitor patients in the hospital and can stop cytokine-release syndrome. If we start to see a problem with their lab studies or physiologic parameters, we can hold the veledimex and see their numbers return to normal quickly.
This is a different situation than cytokine-release syndrome associated with CAR (chimeric antigen receptor) T-cell therapy, in that we have a clear “on/off switch” for our therapy, which we don’t have with CAR T-cell therapy.
Mechanism of Activity
Tell us more about how this novel immunotherapy works.
The way I typically present this to patients is that IL-12, which is a cytokine, is an accelerator on the immune system. IL-12 can stimulate production of interferon-γ and result in more inflammatory T cells and infiltrating CD8-positive cells, without causing disproportionate increases in T regulatory cells, which would downregulate immune activity. We call this “making a cold tumor ‘hot.’” Simply, with IL-12 therapy, we’re getting the immune system cranked up to be aggressive about fighting the cancer.
The gene therapy and the viral vector are a mechanism for allowing us to do this in what we hope is a safe and controlled fashion. Clinical trials were undertaken with systemic IL-12, but they have been stopped because of poor tolerability. Systemically delivered cytokines have a lot of off-target toxicities.
For more on a novel cancer vaccine targeting the tumor-specific antigen survivin in newly diagnosed glioblastoma, see an interview with Manmeet S. Ahluwalia, MD, on The ASCO Post Newsreels at www.ascopost.com/videos.
This approach aims to minimize systemic toxicity by using a ligand-inducible expression switch system to locally control the production of IL-12 in the microenvironment. In this system, transcription of the IL-12 gene only occurs in the presence of the activator ligand, veledimex. The concept is to bring the active agent, the IL-12, to the area where it is theoretically needed the most, which is within the tumor and its microenvironment, and to be able both to “turn on” the delivery of the IL-12 and to quickly “turn it off” as well. Ideally, this leads to a substantial amount of tumor cell death and the ability of the immune system to continue killing the tumor cells.
The oral activator is important in this construct. What are you asking it to do here?
The oral activator veledimex allows for the transcription of the IL-12 gene that was brought into the tumor by the adenoviral vector. Veledimex binds to the construct within the tumor and allows that gene to be transcribed locally. This elicits and sustains an intratumoral immune response through regulatable expression of IL-12, resulting in interferon-γ (IFN-γ) generation and an increase in tumor-infiltrating lymphocytes. When you stop giving the veledimex, you’re no longer transcribing the gene, and now you’ve “turned it off.”
Enhanced Immunogenicity in Tumor Samples
You say there is evidence from tumor samples that you’re accomplishing what you want to and enhancing immunogenicity. Did you find that in your study?
We can measure the downstream product of veledimex in the blood. We can see there is IL-12 elevation above the baseline and the increase in IFN-γ in the blood. This is a proxy for what’s going on in the tumor.
The phase I data also established that veledimex crosses the blood-brain barrier. In addition, there is a positive correlation between the dose of veledimex and serum IL-12 as well as IFN-γ concentrations proportionate to the dose of veledimex. These concentrations return to baseline after veledimex discontinuation.
However, post-treatment tumor samples clearly demonstrate persistent infiltration of CD8-positive cells, showing that the tumor microenvironment remains very “hot,” even after stopping the actual therapy.
Since the aim of much research in the field of immunotherapy is to “make a cold tumor hot,” can this approach can be translated to other tumor types besides glioblastoma?
I definitely think it can. Glioblastoma has been a notoriously “cold” tumor. We’ve consistently had difficulty using immunotherapies in this space, with a number of high-profile studies turning out, unfortunately, to be negative. I think this is the perfect place to explore this. One could extrapolate that if our approach is successful in glioblastoma, it may also be successful in other infiltrating gliomas. And one could also hypothesize that it might make sense in other types of cancer outside of the nervous system as well.
Do you believe that regulatable IL-12 gene therapy may be even more effective in combination with another treatment?
We imagine this approach could be compatible with other immunotherapeutic approaches as well as other nonimmunotherapeutic approaches that we have within the space of glioblastoma. Our phase I substudy cohort is now also being treated with nivolumab. And there is a phase II nonrandomized study that is adding another anti–PD-1 [programmed cell death protein 1] antibody, cemiplimab, to our IL-12 gene-therapy approach. This is based on preclinical studies showing the combination with PD-1 antibodies is superior to the gene therapy alone. We have not yet reported outcomes, but we have reported that we can likely combine these approaches safely.2
We think this is an impressive treatment approach for recurrent glioblastoma, and we’ve shown that it can actually work: you can cut out tumor, inject the virus, bring a gene into the tumor, turn the gene on with a pill, have the gene create the product that it’s supposed to make—IL-12—have the IL-12 increase the downstream product—IFN-γ—presumably along with other downstream products. And then that can lead to a substantial infiltration of immune cells into the tumor, making that cold tumor hot. From my perspective, the fact that all these different steps can come together consistently and in a relatively safe manner is impressive. ■
DISCLOSURE: The study discussed was funded by Ziopharm Oncology and the National Institutes of Health. Dr. Lukas has consulted for AbbVie, Eisai, and Monteris Medical and has received honoraria for serving as medical editor for MedLink Neurology and EBSCO Publishing.
1. Chiocca EA, Yu JS, Lukas RV, et al: Regulatable interleukin-12 gene therapy in patients with recurrent high-grade glioma: Results of a phase 1 trial. Sci Transl Med 11:eaaw5680, 2019.
2. Chiocca EA, Lukas RV, Rao G, et al: Evaluation of controlled IL-12 in combination with a PD-1 inhibitor in subjects with recurrent glioblastoma. 2019 ASCO Annual Meeting. Abstract 2020. Presented June 2, 2019.