This article is more than 5 years old. Information may no longer be current.
Autologous cell vaccine prolongs survival in newly diagnosed glioblastoma
Click Here to Manage Email Alerts
ATLANTA — An autologous cell vaccine composed of glioblastoma tumor cells and an antisense molecule against insulin-like growth factor type 1 receptor DNA/mRNA prolonged PFS and OS among adults with newly diagnosed glioblastoma compared with historical data of patients treated with standard of care, according to phase 1b study results presented at American Association for Cancer Research Annual Meeting.
“Glioblastoma is a tragic disease with a very poor prognosis,” David W. Andrews, MD, Anthony Alfred Chiurco professor of neurological surgery and director of the division of neuro-oncologic neurosurgery and stereotactic radiosurgery at Jefferson Health, said during his presentation. “The only study with incremental benefit with the current standard of care was published in 2005. Subsequent prospective randomized trials have shown no increase in OS benefit.
“This is a very disruptive disease for patients and their families,” Andrews added. “They will often lose autonomy and will have a very poor quality of life up until their demise.”
Andrews and colleagues developed a novel, personalized immunotherapeutic that is delivered 24 hours after surgery, as craniotomy disrupts the blood-brain barrier for up to a week.
“This is time enough for the adaptive immune system to invade the tumor microenvironment,” Andrews said.
Researchers combined glioma cells collected during surgery with an antisense molecule against insulin-like growth factor type 1 receptor DNA/mRNA (IGF-1R AS ODN) and encapsulate them in diffusion chambers. They were then irradiated with 5 Gy radiation to promote the release of tumor antigens with an immunogenic form.
“It is our belief that the fully formulated chamber causes the release of tumor antigens, which then lead to an immunologic effect,” Andrews said.
These chambers containing the autologous tumor cells and IGF-1R AS ODN are implanted under the rectus sheath within 24 hours of surgery for up to 48 hours.
Andrews and colleagues evaluated this approach in 33 patients assigned to one of four dose cohorts: 10 chambers implanted for 24 hours (n = 6), 10 chambers implanted for 48 hours (n = 5), 20 chambers implanted for 24 hours (n = 5), or 20 chambers implanted for 48 hours (n = 17). The first 23 patients were randomly assigned to one of these dose cohorts. However, after an interim analysis of pro-inflammatory serum cytokines showed that the most favorable response was seen at the highest dose exposure, random assignment stopped and all remaining patients were treated only at the highest dose.
Patients underwent standard of care with radiotherapy and temozolomide after 4 to 6 weeks.
Safety served as the primary endpoint. Tumor response served as a secondary endpoint, and PFS and OS were exploratory endpoints.
Median follow-up was 13 months (range, 4-29).
Researchers observed eight adverse events possibly related to treatment, including two grade 3 seizures, one grade 2 and one grade 3 deep vein thrombosis, one grade 3 hydrocephalus, one grade 3 elevation in alanine aminotransferase and one grade 3 elevation in aspartate aminotransferase, and one grade 3 encephalopathy.
Among patients who achieved gross total resection, researchers observed complete resolution of enhancing nidus in tumor cells postoperatively with no incidence of recurrence at 6-month and later follow-up.
Patients who achieved subtotal resections — which represent a “more interesting and compelling group,” based on how extensive their tumors were — demonstrated persistent enhancement of nidus, which in some cases was enlarged, at 6 months. However, researchers observed complete resolution and nonmeasurable disease at the most recent follow-up.
“As a neurosurgeon practicing for 30 years, I’ve never seen neurologic or radiographic responses like this,” Andrews said.
Researchers then evaluated T-cell function as an exploratory endpoint. Patients who had good T-cell function (n = 15) — defined as having the median or more T cells expressing interferon-gamma — showed a “compelling, doubling of survival” compared with patients with poor T-cell function. Median OS was 21.9 months for those with good T-cell function vs. 10.1 months for those with poor T-cell function (2-year OS, 38% vs. 27%).
“This gave us preliminary evidence that T-cell function may be a predictor of response to immunotherapy and may inform future trials,” Andrews said.
Compared with a historical cohort of 35 patients from Thomas Jefferson University matched for gender and age who received standard of care, patients in the intention-to-treat population of the current analysis demonstrated significantly longer median OS (17.3 months vs. 12.1 months; P = .016).
When looking at patients treated with the vaccine according to MGMT promoter methylation status, those with methylated MGMT had median OS of 30.9 months compared with median OS of 11.3 months in those with unmethylated MGMT (P = .005).
Median PFS also appeared improved in the current cohort (10.4 months) compared with data evaluating standard of care from Stupp and colleagues (6.9 months; P = .002) and Kong and colleagues (5.4 months; P = .003).
Researchers are planning a phase 2 single-arm trial of IGF-1R AS ODN dose escalation this year.
Researchers noted that, among the 41% of patients who died of other causes in the current analysis, 44% of deaths were attributed to complications from standard of care received after vaccination.
“Standard of care is associated with significant morbidity and immune compromise,” Andrews said. “We are planning a follow-on phase 3 trial that is designed to address this by including an arm involving less radiation after vaccination.” – by Alexandra Todak
Reference:
Andrews DW, et al. Abstract CT038. Presented at: AACR Annual Meeting; March 29-April 3, 2019; Atlanta.
Disclosures: Andrews reports he is the co-founder and chief medical officer of, and holds stock in, Imvax Inc. Other study authors also report financial relationships with Imvax.
Perspective
Back to Top
Kunle Odunsi, MD, PhD, FRCOG, FACOG
Glioblastoma is one of the worst cancers because it is so very aggressive and so difficult to treat. The use of autologous tumor cells is not new, but what is new is its application to a very difficult-to-treat disease in the brain, to harvest brain tissue used in a novel way as a vaccine. This could have tremendous implications for patients if confirmed in later trials. This is a complicated protocol, but to implement it and be successful by showing these good preliminary clinical results and benefit in terms of prolongation of PFS is quite remarkable.
The results were very clear, showing significant shrinkage in the imaging studies. Bringing this treatment upfront so that patients have the ability to generate immune responses as quickly as possible during the course of treatment is particularly impressive.
In general, several cancer vaccines have failed in the past. But the scientific understanding we now have, combined with this platform of delivery, is beginning to make a difference. The vaccine presented in this study is more powerful than conventional vaccines we tested years ago. This really has tremendous promise.
One of the major next steps and current questions is to better characterize and understand how this vaccine works. One can hypothesize that the vaccine is probably generating immune reaction against tumor mutations, essentially neoantigens, because they are present in the autologous tumors. So, the question is, what are the kinetics, and what is the nature of the immune cells that you are generating with this vaccine approach? Are you generating CD4 cells and/or CD8 cells? Do these have effector cells vs. memory attributes? Do they become exhausted? I think we need that kind of detailed analysis so that going forward we don’t make some of the mistakes of the past, whereby we go to a large randomized trial and then it fails. If, for example, the T cells that you are generating ultimately become exhausted, do you need to combine with a checkpoint inhibitor? That plays into how you design the larger trials so that patients can derive maximum benefit.
I think this is a very promising and exciting approach that probably should move to a larger phase 3 trial, but I would caution the investigators to try and further understand its mechanism, which they likely are already working on. What kind of T cells are generated by the vaccine, their quantity, quality and kinetics?
There are several ongoing promising studies in glioblastoma. There are also indications of trying to use adoptive T-cell therapy, which I think is a very promising area, and there are studies trying to combine checkpoint inhibitors with other approaches. I think, overall, glioblastoma is going to have a new face in the next few years based on some of the promising results at this meeting.
Read more about
Click Here to Manage Email Alerts