Dendritic Vaccine for Glioblastoma: Hope Hyped, Say Some

Alexander M. Castellino, PhD

June 13, 2018

When a public figure is diagnosed with glioblastoma (as happened with US Senator John McCain) or dies of it (as happened with the recent death of UK cabinet minister Tessa Jowell), positive press coverage on glioblastoma gets a boost.

A recent example was the news about an experimental vaccine (DCVax-L) for glioblastoma, reported to extend survival by almost 2 years, which was very enthusiastically highlighted by the UK press.

The findings were published May 29 in the Journal of Translational Medicine. Coauthor Keyoumars Ashkan, MD, a professor of neurosurgery at King's College Hospital in London, United Kingdom, said in a statement that the interim results of the trial gave new hope to patients.

The news boosted the share price of the company developing the vaccine, Northwest Biotherapeutics Inc.

However, experts contacted by Medscape Medical News expressed misgivings about the research. They were critical about how the study was conducted and reported and concerned that the publication was not peer-reviewed; it was published 2 weeks after it was submitted. One expert who expressly asked not to be quoted said that the report was more for Wall Street to applaud than for scientists to react to.  

Two glioblastoma experts not associated with the study expressed concern. Roger Stupp, MD, chief of neuro-oncology at Northwestern University's Feinberg School of Medicine, Chicago, Illinois, pointed out that the publication has withheld significant information and it is not possible to determine the significance of these findings. 

One concern was that the survival was shown for a pooled patient population. "In the Kaplan-Meier survival graph, it is difficult for me to determine its significance when I am not provided information such as numbers at risk [typically provided at each survival time point]," he said.

Howard A. Fine, MD, director of the Brain Tumor Center at Weill Cornell Medicine and New York-Presbyterian, New York City, agreed.

"It was premature to publish the data at this time," Fine said. "I will wait to see data for its primary endpoint, progression-free survival," he added.

"The study should have been published after the data were unblinded," Stupp said in agreement. Both expressed disappointment that the study did not contain more data, including those related to the primary endpoint as stratified by the randomization group.

"Immunotherapy is of great interest in modern oncology, but at least the way we deliver it currently, it has failed to demonstrate repeated successes or sustained responses in primary brain tumors," Stupp told Medscape Medical News.

Medscape Medical News reached out to the company developing the vaccine, Northwest Pharmaceuticals Inc, and was offered a conference call with chief technical officer, Marnix Bosch, MBA, PhD, and chief executive officer, Linda Powers.

Why publish blinded data? Why not wait until the primary endpoint could be reported? Medscape Medical News asked.

"With the interim data, doctors and patients can get a glimpse into what is developing," Powers said. Although the data are interim, Powers pointed out that it was important for doctors and patients to know what was "coming along," and she also commented that "Survival is the gold standard."

To Stupps' comment that the numbers at risk are not available, Powers said that they were specific in stating that they were Kaplan-Meier estimates.

The trial was "more complex" and not close to completion yet, she added. "More than 108 patients are alive at the time the data were collected," Powers told Medscape Medical News.

As of yet, a Biologics License Application has not been filed with the US Food and Drug Administration. "We will finish the trial before that happens," Powers noted.

"Conducting and reporting research in such a way is nonscientific and should there truly have been an effect of DCVax-L, the conduct of the trial makes it certain that we will never know," Stupp said.

Study Details

The trial was conducted at 80 sites in four countries and, although initiated in 2007, it was paused from 2009 to 2011 for economic reasons. The last patient was enrolled in 2015, and this represents the first blinded interim data analysis.

All patients with glioblastoma underwent surgical resection and 6 weeks of standard-of-care chemotherapy (SOC) with temozolomide before enrollment and randomization, which was done centrally and stratified by clinical site and MGMT (O6-methylguanine-DNA methyltransferase) gene promoter methylation status.

Patients were randomly assigned to receive DCVax-L and SOC (n = 232) or SOC and peripheral blood mononuclear cells (PBMCs) (n = 99). PBMCs served as placebo to visually make it indistinguishable from dendritic cells.

Patients in both groups received monthly adjuvant temozolomide at standard dosage along with DCVax-L or PBMCs given on days 0, 10, and 20; then at months 2, 4, and 8; and every 6 months starting at month 12.

DCVax-L was prepared by using tumor tissue processed into a lysate and then collecting, purifying, differentiating, activating, and loading autologous dendritic cells. Two grams of tumor tissue produced 10 doses for the 36-month treatment and follow-up. The vaccine was cryopreserved in aliquots, centrally stored, and shipped individually to the trial sites.

Each DCVax-L treatment was given at a dose of 2.5 million autologous tumor-lysate pulsed dendritic cells, which were administered intradermally in the upper arm, with arms alternating between visits.

After tumor progression or recurrence, patients were allowed to receive DCVax-L and approved treatments per local practice.

Questions Arising About Study Design

Stupp and Fine expressed concern with the study design. With a crossover allowed at progression, approximately 90% of patients received the vaccine, and so they suggested that it would be difficult to tease out any benefit that the patients may get from the vaccine.

"We all very much hope that the treatment ultimately demonstrates a true benefit to our patients, but I am concerned it may be difficult to ultimately assess any survival advantage from the treatment given the 90% crossover to the vaccine," Fine said.

Powers indicated that the trial was started 11 years ago and no one was thinking of immunotherapies. "In order to enroll patients and retain them, we felt it was important to build into the trial design the crossover element," she said.

Of 1599 patients screened, only 331 were randomly assigned. Fine had additional concerns with the patients included in the study. He pointed out that patients for this trial were highly selected through exclusion of patients who had tumor progression, declined leukapheresis, withdrew consent, and had insufficient tumor lysate (a likely marker for patients with less than total gross surgical resection) — all potential and likely surrogates for patients with worse prognoses.

"Thus, this study selected for the 20% of patients most likely to have the more favorable prognostic factors," he said, compared with most upfront GBM trials that randomly assign, treat, and analyze a much higher proportion of screened patients.

Powers disagreed, pointing out that 306 patients screened did not have glioblastoma and 201 did not have sufficient tumor lysate, indicating these tumors were not large. She further added that the demographic profile of the patients reflects the general glioblastoma population.

Study Findings

Although enrollment was complete in 2015, the trial is still ongoing and will remain blinded until the required number of events has occurred.

The median overall survival (OS) from surgery was 23.1 months, which compares well with the median OS of 15 to 17 months from surgery that is typically seen in clinical practice. Median OS was longer in patients with tumors having methylated MGMT (n = 131), at 34.7 months, than in those with unmethylated MGMT (n = 162), at 19.8 months. The 2-year and 3-year OS was also higher for these patients.

For patients with total surgical resection (n = 209), median OS from surgery was 25.4 months, compared with 21.1 months for those with partial surgical resection.

Stupp indicated that if time from surgery to randomization were deducted, OS would not differ from what is currently seen in this patient population. Powers pointed out that they were very clear about their assessment. "It is not very different from reports, which provide data from time of diagnosis," she said. "Survival from time of diagnosis is similar to time from surgery," she said.

Median OS was highest in patients with both MGMT methylated tumors and gross total resection: 36.5 months.

The authors noted that approximately a third of patients (n = 100) had an extended median OS — 40.5 months — which is not explained by positive prognostic factors, such as MGMT methylated tumors, complete resection, and age younger than 50 years. For the authors to analyze the tail of the Kaplan-Meier OS curve, data must further mature, the authors contend.

DCVax-L treatment was well tolerated. Before DCVax-L, the most common side effect after chemoradiation was lymphopenia (51%). Only seven patients (2.1%) experienced grade 3 or 4 adverse events that were possibly treatment related; these included cerebral edema (n = 3), seizures (n = 2), nausea (n = 1), and lymph gland infection (n = 1).

With such a safety profile, the study authors point out that "this DC vaccine may be suitably given in a wide range of clinical settings, and may be potentially combined with other treatment agents…without resulting in undue toxicities for patients."

Before that can happen, further studies are warranted and mature data from this study are desired. As coauthor Ashkan said in his statement, "[D]efinitive judgment needs to be reserved until the final data is available."

Northwest Biotherapeutics Inc is the trial sponsor and had a role in the design and conduct of the study, along with academic advisers; collection, management, analysis, and interpretation of the data; preparation, review, and approval of the manuscript; and the decision to submit the manuscript for publication. Authors Bosch and Maida are employees of Northwestern Biotherapeutics Inc, hold stock or stock options in and receive travel expenses. 

J Transl Med. Published online May 29, 2018. Full text

Follow Medscape Oncology on Twitter: @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.