In preclinical experiments, treatment with the experimental antibody Hu5F9-G4 (Forty Seven Inc.), which is designed to block inhibitory interactions between tumor cells and macrophages, shrank several different types of pediatric brain tumors. The study findings, reported in the March 15, 2017, issue of Science Translational Medicine, mark the first time that the innate immune system has been targeted in pediatric brain tumors.

The team tested the antibody in models of "five very distinct types of highly malignant brain tumors, which essentially have no standard of care yet," co-corresponding author Siddhartha Mitra told BioWorld Today.

In patient-derived xenograft models, treatment with Hu5F9-G4 inhibited the growth of group 3 medulloblastoma, atypical teratoid rhabdoid tumor, primitive neuroectodermal tumor, pediatric glioblastoma and diffuse intrinsic pontine glioma. It was also effective against metastatic medulloblastoma that had spread within the brain.

Pediatric brain tumors, collectively, are challenging to treat. Radiation can be a treatment, but mostly, "there's nothing defined," said Mitra, who is a senior scientist at the Stanford Institute for Stem Cell Biology and Regenerative Medicine.

Like adult brain tumors, their location next to or intertwined with brain structures that are critical for survival often makes them inoperable.

An additional challenge, co-corresponding author Samuel Cheshier told BioWorld Today, is that pediatric tumors "seem to have much less mutational burden than adult tumors," whose cells have had a much longer time to accumulate the mutational burden that is an inevitable part of aging. "That limits your ability to do targeted therapies against tumor antigens."

Cheshier is assistant professor of neurosurgery at the Stanford University School of Medicine.

T-cell directed immunotherapy in the form of checkpoint blockade, too, is challenging in brain tumors and more challenging in pediatric brain tumors.

"It's quite a challenge to do immune therapy in the brain, and some of the current immune therapies are running into problems with swelling," Cheshier said. When T cells attack and kill tumor cells, those cells lyse, or burst open, releasing their contents. "That further spurs the immune system," Cheshier said, "but it also causes a very profound inflammatory reaction." One of the byproducts of that inflammatory reaction is swelling, and because brain tumors are confined by the skull, such swelling becomes a problem very quickly, as pressure compromises brain function.

When pediatric brain tumors are treated with checkpoint blockers, though, there is less swelling because there is less of an effect in the first place. T cells are part of the adaptive immune system, which matures relatively late in development.

The antibody tested by the Stanford team targets CD47, a surface protein that is expressed on many different tumor types and interacts with SIRP-alpha. SIRP-alpha is an inhibitors signal on macrophages, innate immune system cells that are a bit like the boa constrictors of the immune system: they swallow their prey whole.

As a result, when the CD47-SIRP-alpha axis is disrupted, "the tumor cells aren't dumping their contents out into the environment, which could cause a very severe [inflammatory] reaction," Cheshier said.

And "because the innate immune system develops much earlier in life, it's actually fairly mature" and better able to respond to disinhibition than T cells, Mitra added.

A January report by the cancer biology reproducibility project that attempted to reproduce an earlier experiment in which disrupting the interaction between CD47 and SIRP-alpha shrank breast tumors in mice was unsuccessful. Mitra, who was a co-author on the original paper, and Cheshier, though, were unconcerned.

Mitra noted that due to resource constraints, the authors attempted to reproduce only one experiment that was part of a much larger five-year study, and that the reviewers of the replication attempt "very clearly" stated in their comments that "the original study was not . . . ambivalent, but the reproducibility study was."

And while there are always multiple reasons for why replication attempts might go awry, Cheshier noted that one distinct possibility is that the replication attempt was done with a mouse version of Hu5F9-G4. "It's essentially a different drug," he said.

Hu5F9-G4 is in multiple phase I trials for both solid and hematological tumors. "We are awaiting the results of the phase I trial," Cheshier said. "Once the phase I trial is done, we plan to lobby the company which has rights to the antibody, Forty Seven Inc., to pursue a trial in pediatric brain tumors."

Stanford University and its Lucile Packard Children's Hospital are part of the Pediatric Brain Tumor Consortium, a cooperative research organization "devoted to the study of correlative tumor biology and new therapies for primary [Central Nervous System] tumors of childhood," according to its web site. The Pediatric Brain Tumor Consortium has "very efficient mechanism to make [trials] happen," Cheshier said. But "that will only happen if the company wants to move in that direction."