Fig. 5: Tumor growth and survival following monotherapy and combination therapy in a DIPG4423 syngeneic orthotopic DMG murine model.
a, Schematic of stereotactic injection coordinates for establishment of the DIPG4423 syngeneic orthotopic DMG murine model (top). Tumor cells were injected 1 mm inferior and lateral to the lambda suture (purple circle; gray dot indicates injection site). Representative sagittal hematoxylin and eosin (H&E) staining confirms accurate tumor localization within the pons (bottom). b,c, Representative H&E images of normal brain parenchyma (b) and tumor periphery (c), showing diffuse infiltration of tumor cells into adjacent normal tissue. H&E staining was performed for all mice included in survival analyses, when feasible. d, UMAP projection of VIPER-inferred protein activity from 8,013 tumor single cells across two untreated DIPG4423 pontine tumors. Cells are colored by assigned human DMG cell state based on the strongest enrichment of human state-specific MRs per cell (one-sided GSEA, FDR-adjusted). e, OncoTarget and OncoTreat scores for drugs tested in this model, shown as −log10(Bonferroni-adjusted P values) derived from one-sided aREA, demonstrating conservation of predicted cell-state-specific activity between human tumors and the model. f, Representative MRI of tumors 37 days post treatment with monotherapies that significantly extended survival, compared with vehicle-treated controls imaged at 25 days. g, High-power H&E images of end-stage tumors from vehicle-treated, trametinib-treated and avapritinib-treated mice, showing diffuse infiltration with minimal necrosis in avapritinib-treated tumors. h, Tumor volume measured 2 weeks post treatment across independent biological replicates treated with OPC-targeting drugs (left) or AC-targeting drugs (right). OPC-targeting drugs included avapritinib (n = 4), dinaciclib (n = 5), mocetinostat (n = 5), trametinib (n = 5) and etoposide (n = 3). AC-targeting drugs included ruxolitinib (n = 4), larotrectinib (n = 3), venetoclax (n = 5) and napabucasin (n = 3). Drugs are ranked by statistical significance relative to vehicle control (n = 9; two-sided Welch’s t-test). i, Kaplan–Meier survival analysis of monotherapy arms demonstrating significant survival benefit compared with vehicle control (log-rank test): avapritinib (P = 1.9 × 10−3), dinaciclib (P = 3.3 × 10−3) and trametinib (P = 0.03). j, Representative MRI images 37 days post treatment show reduced tumor burden following treatment with OPC-targeting and AC-targeting drug combinations compared with vehicle or corresponding monotherapies from f. k, Tumor volume 2 weeks post treatment for avapritinib (n = 5), trametinib (n = 5) and dinaciclib (n = 5), administered alone or in combination with AC-targeting drugs, including avapritinib + venetoclax (n = 5), avapritinib + ruxolitinib (n = 3), avapritinib + larotrectinib (n = 3), trametinib + ruxolitinib (n = 4), trametinib + venetoclax (n = 5) and dinaciclib + ruxolitinib (n = 3). Statistical significance between combination therapies and their respective monotherapies was assessed using two-sided Welch’s t-test. l, Kaplan–Meier survival analysis comparing vehicle, monotherapy and combination therapy arms for avapritinib, trametinib and dinaciclib. Four out of six combinations significantly improved survival relative to the corresponding monotherapy (log-rank test): avapritinib + ruxolitinib (P = 0.02), trametinib + ruxolitinib (P = 0.01), dinaciclib + ruxolitinib (P = 0.01) and avapritinib + larotrectinib (P = 0.02). Asterisks denote statistical significance: *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; n.s., not significant. Panel a created in BioRender; Calvo Fernandez, E. https://biorender.com/nhid41b (2026).
