A brain-persistent DDR2-degrading antibody reverses Alzheimer's pathologies by restoring brain fluid dynamics and metabolic clearance

Alzheimers disease (AD) is defined by A{beta} deposition, yet cerebrovascular and glymphatic dysfunction are early drivers of progression. We identify discoidin domain receptor 2 (DDR2), a collagen-sensing receptor tyrosine kinase, as a central mediator of neurovascular impairment in AD. Integrative analysis of human single-nucleus RNA (snRNA)-seq data and immunohistochemical validation across human, non-human primate, and mouse AD models revealed that DDR2 is markedly upregulated in reactive astrocytes, perivascular fibroblasts (PVFs), and choroid plexus epithelial cells (CPECs). Astrocytic DDR2 overexpression in APP/PS1 mice exacerbated A{beta} deposition, reduces cerebral blood flow (CBF), and impaired glymphatic function. In CPECs, DDR2 upregulation accompanied increased type X collagen (Collagen X). Central nervous system (CNS) delivery of a monoclonal antibody (HL2) that promotes lysosomal degradation of DDR2 ablated pathogenic signaling, reduced Collagen X, and rescued cognitive, vascular, and glymphatic deficits in APP/PS1 mice. A DDR2-targeted PET probe enabled in vivo visualization of target engagement in these models. These findings establish DDR2 as a key driver of neurovascular dysfunction and support CNS-delivered DDR2-targeted therapy for AD. INTRODUCTIONAlzheimers disease (AD) is defined by amyloid-{beta} (A{beta}) plaques and tau tangles, yet the modest clinical benefits of recently approved immunotherapies underscore its multifactorial nature. Cerebrovascular dysfunction and impaired glymphatic clearance are now recognized as early and integral contributors to AD pathogenesis, but the molecular drivers disrupting these vascular and clearance networks remain poorly defined. RATIONALEDiscoidin domain receptor 2 (DDR2), a collagen-binding receptor tyrosine kinase, drives fibrosis in peripheral organs and has been linked to AD pathology, yet the specific cell types and mechanisms through which DDR2 contributes to disease progression remain unknown. Furthermore, strategies for sustained in vivo suppression of DDR2, together with tools for non-invasive monitoring, are lacking. We therefore mapped DDR2 expression across human AD brains, modeled its overexpression in mice, and developed a targeted therapeutic and imaging strategy. RESULTSIntegrative analysis of human single-nucleus RNA-seq data and immunohistochemical validation across human, non-human primate, and mouse AD brains revealed pronounced DDR2 upregulation in three functionally distinct cell populations: reactive astrocytes, perivascular fibroblasts (PVFs), and choroid plexus epithelial cells (CPECs). In astrocytes, DDR2 expression correlated positively with Braak stage, A{beta} burden, and cognitive decline. Astrocyte specific Ddr2 overexpression in APP/PS1 mice exacerbated A{beta} deposition via {beta}-amyloid cleaving enzyme 1 (BACE1) upregulation, reduced cerebral blood flow (CBF), disrupted blood-brain barrier (BBB) integrity, and impaired glymphatic function. In CPECs, DDR2 upregulation was accompanied by increased expression of type X collagen (Collagen X), a non-fibrillar collagen that is both a DDR2 activator and a marker of calcification, suggesting a potential link to choroid plexus calcification. We further developed a monoclonal antibody (HL2) that drives efficient internalization and lysosomal degradation of DDR2 regardless of collagen occupancy, by binding a unique "waist" epitope distinct from the collagen-binding site. Adeno-associated virus (AAV)-mediated central nervous system (CNS) delivery of HL2 in APP/PS1 mice ablated DDR2, reduced Collagen X expression, reversed cognitive impairment, diminished A{beta} plaque burden. This was accompanied by restored cerebral perfusion, enhanced glucose supply, and normalized CSF dynamics. A DDR2-targeted PET tracer, validated in human idiopathic pulmonary fibrosis (IPF) studies, enabled in vivo visualization of target engagement in mouse AD models. CONCLUSIONThis study identifies DDR2 as a master regulator of multi-compartmental failure in AD, orchestrating dysfunction across cerebral vasculature, CSF dynamics, and the choroid plexus. We developed a monoclonal antibody that drives durable internalization and lysosomal degradation of DDR2; when delivered via a neuron-tropic AAV, it achieves sustained CNS expression and confers broad therapeutic efficacy against AD. A companion DDR2-targeted PET tracer enables noninvasive visualization of target engagement in vivo. Together, this integrated platform establishes a new paradigm for disease-modifying therapy in AD. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=185 SRC="FIGDIR/small/26348575v1_ufig1.gif" ALT="Figure 1"> View larger version (79K): org.highwire.dtl.DTLVardef@1de2ea4org.highwire.dtl.DTLVardef@1c82eaaorg.highwire.dtl.DTLVardef@16fb054org.highwire.dtl.DTLVardef@1a165ca_HPS_FORMAT_FIGEXP M_FIG DDR2 drives multi-compartmental dysfunction in Alzheimers disease. The upper region depicts an AD brain where DDR2 is upregulated in astrocytes, PVFs, and CPECs. DDR2-activating ligands Collagen I and Collagen X are highly expressed in PVFs and CPECs, respectively. In astrocytes and PVFs, DDR2 impairs CBF and energy supply, disrupts CSF circulation, and promotes A deposition. In CPECs, DDR2 induces Collagen X expression, linking to choroid plexus dysfunction. The middle region shows therapeutic intervention with AAV-delivered HL2, a monoclonal antibody that binds a unique epitope on DDR2 non-competitive with collagen binding and promotes its lysosomal degradation, thereby rescuing cognitive and vascular deficits. The site of action of DDR2 blockade by HL2 is also depicted. C_FIG