Combinatorial Targeting of Avapritinib-Driven MAP Kinase Activation in High-Grade Glioma

Technical Summary: Combinatorial Targeting of Avapritinib-Driven MAP Kinase Activation in High-Grade Glioma

Problem Statement
High-grade glioma (HGG), particularly subsets harboring PDGFRA alterations (mutations or amplifications), remains incurable with less than 10% of patients surviving beyond two years. While PDGFRA is a validated therapeutic target, monotherapy with tyrosine kinase inhibitors (TKIs) has historically yielded limited and transient benefits. Previous trials of PDGFRA-targeting agents in HGG have largely failed, often due to resistance mechanisms such as secondary mutations, upregulation of alternative receptor tyrosine kinases (RTKs), or reflexive increases in downstream pathway activity. Specifically, the CNS-penetrant inhibitor avapritinib, which targets exon 18 PDGFRA mutations, showed initial promise but failed to provide durable responses in preclinical models and clinical cohorts, necessitating a deeper understanding of the adaptive signaling networks that emerge upon treatment.

Methodology
The study employed a multi-modal approach combining high-dimensional kinome profiling, in vitro and in vivo preclinical models, and a retrospective clinical case series:

• Kinome Profiling: High-throughput kinase activity mapping (HT-KAM) was performed on avapritinib-treated HGG models (PPK and BT245) to capture global kinase activity changes. This included both acute (8-hour) and chronic (resistant) treatment conditions.
• Preclinical Models: The study utilized syngeneic mouse-derived neurosphere cultures (PPK, PPK*) generated via in utero electroporation (IUE) expressing PDGFRA D842V, H3K27M, and dominant-negative TP53. Patient-derived cell lines (BT245, DIPG-XIII-P*, 154C) and organoids were also used. Isogenic models were generated in neural progenitor cells (NPCs) to compare wild-type amplified PDGFRA against specific mutations (D842V, Y288C).
• In Vivo Studies: Orthotopic and flank xenograft models in NSG mice were treated with avapritinib alone or in combination with MEK inhibitors (trametinib, selumetinib), imipridones (ONC201, ONC206), or ERK inhibitors (ulixertinib). Survival, bioluminescence, and multiplexed immunofluorescence were monitored.
• Single-Cell and Genomic Analysis: Single-cell RNA sequencing (scRNA-seq) was performed on treated tumors to map pathway activity across cellular subpopulations. Whole-exome sequencing (WES) and RNA-seq were used to assess genetic and transcriptional changes in resistant cells.
• Clinical Correlation: A small case series of five patients with PDGFRA-altered brain tumors treated with avapritinib combined with a MEK inhibitor (trametinib or selumetinib) under compassionate use protocols was analyzed for response and tolerability.

Key Contributions and Results

• Identification of Adaptive MAPK Activation: Contrary to the expectation that PDGFRA inhibition would suppress downstream signaling, the study found that avapritinib treatment induces sustained activation of the MAPK pathway. HT-KAM and Western blot analyses revealed robust, dose-dependent ERK activation following avapritinib exposure, persisting even when PDGFRA phosphorylation was fully inhibited. This adaptive response was observed in both acute and chronic treatment settings.
• Cellular Specificity: scRNA-seq analysis demonstrated that this MAPK reactivation is most pronounced in cycling oligodendrocyte precursor cell-like (OPC-like) tumor populations, which exhibit high PDGFRA expression and innate sensitivity to avapritinib.
• Mechanism of Resistance: The study elucidated that ERK-driven activation stabilizes the anti-apoptotic protein MCL-1 and degrades the pro-apoptotic protein BIM. Knockout of MCL1 in PDGFRA-altered cells restored sensitivity to avapritinib-induced apoptosis, confirming that MCL-1 stabilization is a key mechanism limiting monotherapy efficacy.
• Variant-Dependent Sensitivity: The magnitude of MAPK activation varied by PDGFRA variant. D842V-mutant cells exhibited heightened and sustained MAPK activation compared to wild-type amplified models, suggesting that D842V tumors may be particularly dependent on this adaptive pathway and thus more amenable to combination therapy.
• Combinatorial Efficacy: In vitro, the combination of avapritinib with MEK inhibitors (specifically trametinib and selumetinib) resulted in synergistic suppression of p-ERK and significant reductions in cell viability. In vivo, while avapritinib monotherapy extended survival modestly, the combination of avapritinib and selumetinib in orthotopic PPK models significantly prolonged median survival (149 days vs. 65 days for controls).
• Clinical Translation: In a small cohort of five patients, the combination of avapritinib and a MEK inhibitor demonstrated feasibility and early clinical activity. Notably, one patient with an unresected PDGFRA D842V-mutant HGG treated with avapritinib and selumetinib achieved a sustained complete regression, and another patient with a D842V-mutant diffuse midline glioma achieved a 40% tumor reduction and prolonged disease control (17 months progression-free survival), exceeding historical benchmarks.

Significance
The paper claims that adaptive MAPK reactivation represents a targetable liability in PDGFRA-driven high-grade glioma. By identifying that avapritinib monotherapy inadvertently promotes survival signaling through ERK-dependent MCL-1 stabilization, the study supports a shift from single-agent therapy to rational combinatorial strategies. The findings suggest that concurrent inhibition of PDGFRA and the MAPK pathway (specifically via MEK inhibition) can overcome this adaptive resistance, offering a promising therapeutic strategy to extend survival in a patient population with historically poor outcomes. The study emphasizes that optimizing combination regimens, potentially favoring selumetinib over trametinib for improved tolerability, is critical for translating these preclinical insights into clinical benefit.