Glioblastomas Not 'Off-Limits' to Immunotherapy

Alexander M. Castellino, PhD

February 13, 2019

Glioblastomas may no longer be the "cold tumors" unresponsive to immunotherapy. Two small studies report on the benefits of program death 1 (PD-1) inhibitors in patients with recurrent, resectable glioblastoma and, along with a third study, hint at why only some patients respond.

The three studies were published online February 11 in Nature Medicine.

One study, conducted in 35 patients with recurrent resectable glioblastoma, showed that median overall survival was nearly doubled when pembrolizumab (Keytruda, Merck) was given before surgery compared with after surgery (13.7 vs 7.5 months; P = .04). Median progression-free survival was also significantly longer (3.3 vs 2.4 months; P = .03).

"This is one of the first studies to evaluate immunotherapy in the neoadjuvant setting for patients with recurrent glioblastoma," coauthor Aaron Mochizuki, DO, UCLA pediatric hematology/oncology fellow at UCLA Jonsson Comprehensive Cancer Center, told Medscape Medical News.

In the article, the authors note that glioblastoma is the most common brain tumor and carries an abysmal survival rate of just 10% at 3 years. The median overall survival is 14.6 months with standard of care surgery, radiation therapy, and temozolomide. For recurrent glioblastoma, median overall survival is an estimated 6 to 11 months.

Another study, also published in Nature Medicine, also explored immunotherapy in glioblastoma, this time with nivolumab (Opdivo, Bristol-Myers Squibb) given both before and after surgery. The study enrolled patients with newly diagnosed (n = 3) or relapsed (n = 27) glioblastoma, which required primary or salvage surgical resection. Patients received a single dose of nivolumab 3 weeks before surgery as well as postsurgically until disease progression.

"In a way, the regimen is both neoadjuvant and adjuvant," senior author Ignacio Melero, MD, PhD, of the Clinica Universidad de Navarra, Pamplona, Spain, told Medscape Medical News.

"Our data indicate that neoadjuvant nivolumab in patients with relapsed glioblastoma followed by surgery is safe and feasible," the authors conclude.

Notably, of the three cases enrolled with treatment-naive disease who underwent primary surgical resection, two patients remained alive for over 2 years (33 and 28.5 months). This finding warrants further investigation, they add.

The outcomes for patients with recurrent glioblastoma is comparable to previous reports, but the limited sample size prevents definitive conclusions about the clinical impact of treatment, they note.

"Glioblastoma is not going to be off limits for immunotherapy," Melero commented.

Why Immunotherapy Works

These studies were unique because researchers had tumor tissue sample and periodic peripheral blood samples (before and every 2 weeks after surgery) available to map the immune landscape in patients.

"Our thinking is that T cells, which normally are responsible for eliminating cancer cells, are rendered ineffective by the tumor, which hijacks signals that normally help prevent autoimmunity," explained UCLA's Mochizuki.

In their study with pembrolizumab, "the neoadjuvant administration of anti-PD1 blockade nearly doubles overall survival in patients with recurrent glioblastoma by priming tumor-specific T cells for a robust immune response," Mochizuki said.

This robust immune response occurs by an upregulation of mRNA associated with T-cell and interferon-gamma related pathways, an increase in the infiltration of T cells, and a focal induction of PD-1 in the tumor microenvironment.

Mochizuki pointed out that their data along with other preclinical data suggests that the tumor acts as a huge repository of potential antigens, which T cells, once released by PD-1 blockade, can respond to and expand, recruiting other antigen-presenting cells and more T cells via interferon gamma to the resection cavity after surgery.

These observations were corroborated by Melero and colleagues who showed, using flow cytometry and multiplex TCR-sequencing of pre- and post-treatment samples, that tumors show more T lymphocytes in the tumor microenvironment after treatment, with a potential broader repertoire of antigen receptors.

"Surgery and the removal of this significant antigenic burden prior to checkpoint release limit the potential immune response, and the resection cavity remains relatively 'cold'," Mochizuki said. When surgery is performed prior to the drug being given, there are relatively few remaining cells, and thus fewer potential targets for the T cells. Without these extra targets, the immune system isn't as good at detecting leftover tumor cells as it could have been, he elaborated.

"Part of the problem with glioblastoma being in the brain is that it makes monitoring intratumoral/tissue changes impractical," Mochizuki said. He explained that changes in activation of peripheral T cells as well as changes in clonotypes were also seen, indicating that tumor-specific T cells do get out into the periphery and may be patrolling for stray glioma cells.

"We're hoping that with more data through an expansion of the neoadjuvant arm will help us identify a potential biomarker of response that's less invasive than serial biopsies," Mochizuki noted.

Only Some Patients Respond to Immunotherapy

A third study, a retrospective analysis of 66 adults with glioblastoma, indicated that patients who responded to pembrolizumab or nivolumab had significantly longer median overall survival than those who did not (15.5 vs 5.7 months; P = 2.2 x 10-5).

"Personalized medicine approach to cancer immunotherapy might allow some patients a clinical benefit from a drug that does not seem to be effective for glioblastoma when considering all patients," co-corresponding author Adam M. Sonabend, MD, from the department of neurosurgery at the Northwestern Feinberg School of Medicine, Chicago, Illinois, told Medscape Medical News.

"Each tumor is unique, thus the diagnosis of glioblastoma is too general to decide on whether to use immunotherapy for a given patient," he added.

In their study, glioblastomas were sampled before and after immunotherapy, and Sonabend and colleagues performed longitudinal characterization using genomic and transcriptomic analyses. Nonresponders to immune checkpoint inhibitors showed a significant enrichment in PTEN mutations and responders had alterations in the MAPK pathway (PTPN11, BRAF).

Using quantitative multiplex immunofluorescence, Sonabend and colleagues noted that glioblastomas with PTEN mutations have immunosuppressive features that influence the microenvironment and are associated with lack of response to PD1 inhibitor immunotherapy. In contrast, tumors bearing alterations in the MAPK pathway (PTPN11, BRAF) are associated with specific features in the microenvironment, which are associated with response to immune checkpoint inhibitors.

"Contrary to other cancers, the number of mutations did not influence response to this therapy," Sonabend said.

Noting that only a subset of patients with glioblastoma responds to immunotherapy, Sonabend pointed out that the identity of these patients was elusive. "Our results provide an insight into the molecular profile that distinguishes tumors that are responsive to this treatment," he said.

Their observations also provide a glimpse into future approaches that may be more fruitful. "The observation in our cohort that BRAF/PTPN11 mutations are enriched in tumors responsive to anti-PD-1 therapy supports the rationale for combining checkpoint inhibitors with MAPK-targeted therapy," Sonabend and colleagues write.

Funding sources for all studies as well as financial disclosures for all authors are included in each article. The authors cited in this news article have reported no relevant financial relationships.

Nature Med. Published February 11, 2019. Study 1, Study 2, Study 3

For more from Medscape Oncology, follow us on Twitter


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.
Post as: