Nucleosomes containing mutant K27M histones in diffuse intrinsic pediatric gliomas (DIPG) exclude PRC2 binding and recruit BET bromodomain proteins; however, residual PRC2-dependent repression of specific loci, is required for DIPG oncogenesis. These results provide a rationale for targeting these epigenetic regulators in patients. Diffuse intrinsic pontine glioma (DIPG) is a highly aggressive pediatric brainstem tumor characterized by rapid and uniform patient demise1. A heterozygous point mutation of histone H3 occurs in more than 80% of these tumors and results in a lysine-to-methionine substitution (H3K27M)2,3. Expression of this histone mutant is accompanied by a reduction in the levels of polycomb repressive complex 2 (PRC2)-mediated H3K27 trimethylation (H3K27me3), and this is hypothesized to be a driving event of DIPG oncogenesis4,5. Despite a major loss of H3K27me3, PRC2 activity is still detected in DIPG cells positive for H3K27M6,7. To investigate the functional roles of H3K27M and PRC2 in DIPG pathogenesis, we profiled the epigenome of H3K27M-mutant DIPG cells and found that H3K27M associates with increased H3K27 acetylation (H3K27ac). In accordance with previous biochemical data5, the majority of the heterotypic H3K27M-K27ac nucleosomes colocalize with bromodomain proteins at the loci of actively transcribed genes, whereas PRC2 is excluded from these regions; this suggests that H3K27M does not sequester PRC2 on chromatin. Residual PRC2 activity is required to maintain DIPG proliferative potential, by repressing neuronal differentiation and function. Finally, to examine the therapeutic potential of blocking the recruitment of bromodomain proteins by heterotypic H3K27M-K27ac nucleosomes in DIPG cells, we performed treatments in vivo with BET bromodomain inhibitors and demonstrate that they efficiently inhibit tumor progression, thus identifying this class of compounds as potential therapeutics in DIPG.

RNA interference–based spherical nucleic acid nanoconjugates are a potentially safe and brain-penetrant approach for glioblastoma treatment.

Diffuse midline gliomas (including diffuse intrinsic pontine glioma, DIPG) are highly morbid glial neoplasms of the thalamus or brainstem that typically arise in young children and are not surgically resectable. These tumors are characterized by a high rate of histone H3 mutation, resulting in replacement of lysine 27 with methionine (K27M) in genes encoding H3 variants H3.3 (H3F3A) and H3.1 (HIST1H3B). Detection of these gain-of-function mutations has clinical utility, as they are associated with distinct tumor biology and clinical outcomes. Given the paucity of tumor tissue available for molecular analysis and relative morbidity of midline tumor biopsy, CSF-derived tumor DNA from patients with diffuse midline glioma may serve as a viable alternative for clinical detection of histone H3 mutation. We demonstrate the feasibility of two strategies to detect H3 mutations in CSF-derived tumor DNA from children with brain tumors (n = 11) via either targeted Sanger sequencing of H3F3A and HIST1H3B, or H3F3A c.83 A > T detection via nested PCR with mutation-specific primers. Of the six CSF specimens from children with diffuse midline glioma in our cohort, tumor DNA sufficient in quantity and quality for analysis was isolated from five (83%), with H3.3K27M detected in four (66.7%). In addition, H3.3G34V was identified in tumor DNA from a patient with supratentorial glioblastoma. Test sensitivity (87.5%) and specificity (100%) was validated via immunohistochemical staining and Sanger sequencing in available matched tumor tissue specimens (n = 8). Our results indicate that histone H3 gene mutation is detectable in CSF-derived tumor DNA from children with brain tumors, including diffuse midline glioma, and suggest the feasibility of “liquid biopsy” in lieu of, or to complement, tissue diagnosis, which may prove valuable for stratification to targeted therapies and monitoring treatment response.

Radiation therapy is the most commonly used postsurgical treatment for primary malignant brain tumors. Consequently, investigating the efficacy of chemotherapeutics combined with radiation for treating malignant brain tumors is of high clinical relevance. In this study, we examined the cyclin-dependent kinase 4/6 inhibitor palbociclib, when used in combination with radiation for treating human atypical teratoid rhabdoid tumor (ATRT) as well as glioblastoma (GBM). Evaluation of treatment antitumor activity in vitro was based upon results from cell proliferation assays, clonogenicity assays, flow cytometry, and immunocytochemistry for DNA double-strand break repair. Interpretation of treatment antitumor activity in vivo was based upon bioluminescence imaging, animal subject survival analysis, and staining of tumor sections for markers of proliferation and apoptosis. For each of the retinoblastoma protein (RB)-proficient tumor models examined (2 ATRTs and 2 GBMs), one or more of the combination therapy regimens significantly (P < .05) outperformed both monotherapies with respect to animal subject survival benefit. Among the combination therapy regimens, concurrent palbociclib and radiation treatment and palbociclib treatment following radiation consistently outperformed the sequence in which radiation followed palbociclib treatment. In vitro investigation revealed that the concurrent use of palbociclib with radiation, as well as palbociclib following radiation, inhibited DNA double-strand break repair and promoted increased tumor cell apoptosis. Our results support further investigation and possible clinical translation of palbociclib as an adjuvant to radiation therapy for patients with malignant brain tumors that retain RB expression.

STING agonists can reprogram the tumor microenvironment to induce immunological clearance within the central nervous system. Using multiplexed sequential immunofluorescence (SeqIF) and the Ivy Glioblastoma Atlas, STING expression was found in myeloid populations and in the perivascular space. The STING agonist 8803 increased median survival in multiple preclinical models of glioblastoma, including QPP8, an immune checkpoint blockade–resistant model, where 100% of mice were cured. Ex vivo flow cytometry profiling during the therapeutic window demonstrated increases in myeloid tumor trafficking and activation, alongside enhancement of CD8 ⁺ T cell and NK effector responses. Treatment with 8803 reprogrammed microglia to express costimulatory CD80/CD86 and iNOS, while decreasing immunosuppressive CD206 and arginase. In humanized mice, where tumor cell STING is epigenetically silenced, 8803 therapeutic activity was maintained, further attesting to myeloid dependency and reprogramming. Although the combination with a STAT3 inhibitor did not further enhance STING agonist activity, the addition of anti–PD-1 antibodies to 8803 treatment enhanced survival in an immune checkpoint blockade–responsive glioma model. In summary, 8803 as a monotherapy demonstrates marked in vivo therapeutic activity, meriting consideration for clinical translation.