MRI-Guided AAV Delivery in NHP Brain Models For CNS Gene Therapy

Central nervous system gene therapy remains technically challenging because therapeutic agents must overcome the blood-brain barrier, reach precise anatomical targets, and achieve controlled distribution within complex neural circuits. For AAV-based therapies, local intraparenchymal delivery is often required to improve regional specificity and reduce systemic exposure.

The review article ​"Guided delivery of adeno-associated viral vectors into the primate brain"​ by Salegio et al., published in Advanced Drug Delivery Reviews in 2012;64(7):598–604, DOI: 10.1016/j.addr.2011.10.005, summarizes how MRI-guided convection-enhanced delivery, stereotactic targeting, and reflux-resistant cannula design can improve AAV vector administration in the primate brain. The article is especially relevant for biopharmaceutical companies developing CNS gene therapies that require reliable non-human primate translational data. 

The paper describes an NHP CNS delivery platform using AAV vectors, particularly AAV2, administered through convection-enhanced delivery. CED applies a pressure gradient from the cannula tip to distribute macromolecules through brain interstitial spaces, enabling wider tissue coverage than diffusion alone.

A key methodological advance was the co-infusion of AAV2 with Gadoteridol, an MRI-visible tracer. The authors reported that real-time MRI signal distribution correlated with subsequent transgene expression in the NHP brain, supporting Gadoteridol as a surrogate marker for vector distribution during infusion.

The reviewed platform also integrated a skull-mounted aiming device, MRI-based planning software, and a reflux-resistant CED cannula. In NHP validation studies, the system supported delivery into the caudate, putamen, thalamus, substantia nigra, subthalamic nucleus, and hippocampus, with cannula tip placement reported within ​<1 mm of the visually identified target site.

The article highlights several endpoints that are highly relevant for NHP preclinical CNS studies.

First, anatomical targeting accuracy is essential because small placement errors can lead to leakage, off-target distribution, or incomplete tissue coverage. In the putamen, cannula placement within 3 mm or less of certain neighboring structures was associated with poorer containment, while optimized placement zones improved target-specific distribution.

Second, real-time MRI monitoring allows investigators to observe infusate distribution during the procedure rather than relying only on post-mortem analysis. This can help identify reflux, ventricular leakage, and inadequate coverage early.

Third, infusion rate is a critical procedural variable. Early CED-based gene therapy studies used infusion rates of ​<1 μL/min and relatively small infusion volumes of 50 μL because visual feedback was unavailable. Later work showed that reflux-resistant cannulae tolerated infusion rates up to 5 μL/min, while significant reflux was observed at 8 μL/min.

One important finding is that MRI-guided CED can provide intra-operative feedback during AAV infusion, improving procedural control and reducing uncertainty in CNS delivery studies.

The article also describes thalamic delivery as a strategy for broader cortical transgene expression. The thalamus contains 50–60 nuclei, and in one NHP example, 300 μL of AAV2-hGDNF/tracer was infused into the left and right thalamus without reported leakage, reflux, or adverse behavioral effects. Extensive thalamic distribution and robust cortical transgene expression were subsequently observed.

For basal ganglia delivery, the authors emphasize that AAV2-GDNF transport appears predominantly anterograde rather than retrograde. This supports striatal rather than nigral delivery for certain Parkinson's disease gene therapy strategies and highlights the importance of selecting delivery targets based on neural circuitry, not only surgical accessibility.

Although this review provides valuable guidance, several limitations should be considered in modern drug development. First, it is a review article rather than a single controlled efficacy study, so its conclusions are best used to inform study design rather than serve as direct efficacy benchmarks. Second, the article focuses mainly on AAV2, while current CNS pipelines often use engineered capsids, alternative serotypes, and modified expression systems. Each vector may differ in tropism, expression kinetics, biodistribution, and immune profile.

In addition, MRI-visible tracer distribution remains a surrogate readout. For robust translational interpretation, sponsors should combine imaging with vector genome biodistribution, transgene expression, histopathology, neurobehavioral monitoring, and pharmacodynamic endpoints.

At Prisys Biotech, our NHP CNS pharmacology platform supports customized non-human primate CNS delivery study design and integrated MRI-based anatomical assessment with translational efficacy endpoints. These capabilities help sponsors connect delivery accuracy with biological outcomes, including vector biodistribution, target engagement, behavioral changes, and safety observations.

This article reinforces a key principle in CNS gene therapy: delivery strategy is part of the therapeutic product strategy. Vector design, anatomical target selection, cannula placement, infusion volume, infusion rate, and real-time monitoring all influence whether a therapy reaches the intended tissue and produces interpretable efficacy data.

For biopharmaceutical companies, NHP models are particularly valuable when therapeutic performance depends on primate-relevant brain anatomy and neural connectivity. Rodent studies may support early proof-of-concept, but they cannot fully reproduce the scale and structural complexity of primate CNS targets such as the putamen, thalamus, substantia nigra, and cortical projection systems.

Prisys Biotech helps global sponsors translate these principles into executable NHP preclinical programs, including feasibility assessment, stereotactic delivery planning, longitudinal monitoring, sample collection strategy, and integrated efficacy interpretation.

To discuss how our NHP CNS Gene Therapy Preclinical Evaluation Service can support your AAV or biologic pipeline, contact Prisys Biotech for a customized study design consultation.

Contact Prisys Biotech

References

• Salegio EA, Samaranch L, Kells AP, Forsayeth J, Bankiewicz KS. Guided delivery of adeno-associated viral vectors into the primate brain.​ Advanced Drug Delivery Reviews. 2012;64(7):598–604. DOI: 10.1016/j.addr.2011.10.005. 
• Bobo R, Laske D, Akbasak A, Morrison P, Dedrick R, Oldfield E. Convection-enhanced delivery of macromolecules in the brain.​ PNAS. 1994;91:2076–2080. DOI: 10.1073/pnas.91.6.2076.
• Su X, Kells AP, Aguilar Salegio EA, Richardson RM, Hadaczek P, Beyer J, Bringas J, Pivirotto P, Forsayeth J, Bankiewicz KS. Real-time MR imaging with Gadoteridol predicts distribution of transgenes after convection-enhanced delivery of AAV2 vectors.​ Molecular Therapy. 2010;18:1490–1495. DOI: 10.1038/mt.2010.114.
• Richardson RM, Kells AP, Martin AJ, Larson PS, Starr PA, Piferi PG, Bates G, Tansey L, Rosenbluth KH, Bringas JR, Berger MS, Bankiewicz KS. Novel platform for MRI-guided convection-enhanced delivery of therapeutics: preclinical validation in nonhuman primate brain.​ Stereotactic and Functional Neurosurgery. 2011;89:141–151. DOI: 10.1159/000323544.