HEMO-GNN: A Physics-Informed Graph Neural Network Platform for Patient-Specific Injectable Microvascular Scaffolds for Local Factor VIII Delivery in Hemophilic Arthropathy

Read full text Summarise with VIRAT
Authors: Basmalaa Thabet
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Hemophilic arthropathy (HA), the dominant source of long-term morbidity in severeHemophilia A, arises from recurrent hemarthrosis that perpetuates a self-reinforcingpathological loop: FVIII deficiency impairs endothelial stability and ECM regulationvia the FAK/Src pathway (Olgasi et al., 2024), driving synovial hypertrophy,hemosiderin deposition, and aberrant VEGF-driven neovascularization that amplifiesjoint fragility and bleeding susceptibility. No existing therapy simultaneouslydismantles both axes of this vicious cycle. We introduce HEMO-GNN — the firsthemophilia-specific, physics-informed Graph Neural Network (PI-GNN) platformthat couples AI-driven patient-specific microvascular topology optimization with aninjectable, photocrosslinkable GelMA/Fibrin hybrid scaffold delivering dual-payloadFactor VIII therapy directly into the hemophilic joint space. The PI-GNN integratesmultimodal patient data — joint MRI hemosiderin maps, Doppler ultrasound vascularprofiling, and plasma VEGF/FVIII biomarkers — into a patient digital twin, generatingoptimal three-dimensional vascular topologies constrained by hemophilia-adaptedMurray's Law, low-shear-stress avoidance targets (>0.5 Pa), hemosiderin exclusionzones, and anti-pathological angiogenesis objectives. The resulting scaffold designis embedded in a visible-light photocrosslinkable GelMA/Fibrin hydrogel (10%/0.5%w/v) for minimally invasive intra-articular injection, eliminating the surgical paradoxinherent to implantation in coagulopathic patients. A dual-payload strategy providesimmediate hemostatic coverage via PLGA microsphere–encapsulated recombinantFVIII (burst release: 0–7 days) and sustained biosynthetic FVIII production viapatient-derived iPSC-endothelial cells (HA-FLF8-iECs), previously validated at~600% of physiological FVIII secretion (Neumeyer et al., 2019). The validationframework spans multiscale CFD simulation across five scaffold geometry groups,acellular microfluidic benchmarking against conventional lattice, Murray's Law, andStanford 2025 organ-scale algorithm controls (Sexton et al., 2025), PLGA releasekinetics, and endothelial biocompatibility assessment. To our knowledge, this is thefirst integrated computational-biomaterial platform to target hemophilic arthropathyas a spatially addressable microvascular disorder, repositioning joint-directed FVIIIdelivery from a systemic replacement paradigm to a localized, AI-personalizedregenerative strategy with direct translational implications.

Fields of Study & Subjects

Hemophilia Treatment and Research Hydrogels: synthesis properties applications Tissue Engineering and Regenerative Medicine

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
30 March 2026
Language
EN