Abstract:
Pediatric brain tumors are the leading cause of childhood cancer-related death1. The overall survival for pediatric
lymphoblastic leukemia is over 90% at 5 years 2. In stark contrast, the overall survival of children with pediatric
high-grade gliomas (pHGG) is less than 20% at 5-years. Remarkable progress has been made over the last
decade in elucidating the origin and genomic landscape of childhood brain tumors 3. Despite these advances,
pHGGs are mostly incurable, as current therapies rarely provide a greater survival benefit over the current
standard of care, focal radiation. The few survivors with pHGG are often left with devastating side effects,
including endocrine morbidity, psychiatric and neurocognitive impairments, developmental disorders,
neurological disease, and a high incidence of secondary tumors4-6. These side effects further highlight the
necessity of developing novel treatment modalities, ideally with minimal toxicity while maintaining significant
prognostic outcomes for children with pHGG. Immune checkpoint inhibition (ICI) resulted in an unprecedented
response rate in many cancer types, including cancers in advanced metastatic stages such as melanoma and
non-small cell lung cancer 7-9. Unfortunately, ICI has largely failed to produce benefits in pHGG, with the
exception of patients harboring constitutional mismatch repair (MMR) deficiency syndrome (CMMRD) 10-13.
Based on limited data from literature and our preliminary observations, we hypothesize that biallelic germline
MMR mutations in pHGG result in enhanced ICI response while somatic MMR mutations do not. We further
hypothesize that stromal MMR mutations drive enhanced ICI response by reversing the immunosuppressive
phenotype of innate immune cell called tumor-associated macrophages (TAMs) in the tumor microenvironment
(TME). TAMs are the most-abundant non-neoplastic cell infiltrates in the TME and express the highest levels of
PD-L114,15, a ligand for the programmed cell death-1 receptor (PD-1) on effector T-cells. To test our hypothesis,
we developed genetic models of germline and somatic MMR mutant pHGG and propose to use these models to
compare their expression profiles to human pHGG samples from CMMRD and non-CMMRD patients. We will
also use these models to determine whether there is a casual link between germline biallelic MMR mutation and
response to anti-PD-L1 therapy. The clinical benefits of this high-risk high-reward application are two-fold. First,
it will establish whether there is a causal link between biallelic MMR mutation in pHGG and immunotherapy
response. If the link exists it will contribute to our understanding of the primary resistance to checkpoint inhibitors
in children with pHGG and subsequently, how to target such mechanisms. Second, CMMRD tumors are resistant
to conventional therapies, since several common chemotherapeutic agents, including temozolomide, require
adequate mismatch repair to exert their cytotoxic effects. Patients with CMMRD tumors thus require novel
therapeutic strategies, and our studies will mechanistically establish whether or not ICI elicits an enhanced anti-
tumor response in these tumors.