Analysis of Subvisible and Visible Aggregation in Lentiviral Vectors with Aura GT System

Analysis of Subvisible and Visible Aggregation in Lentiviral Vectors with Aura GT System — Executive Summary

Introduction — Significance of the topic

Lentiviral vectors (LVVs) are an increasingly important modality for cell and gene therapies because of their large payload capacity, stable genomic integration, and ability to transduce non‑dividing cells. Their structural complexity (lipid envelope, surface proteins, and genomic nucleic acid) and scarcity of material pose analytical and formulation challenges. Subvisible and visible particulate formation is a critical quality attribute linked to product stability, potency, and patient safety. Analytical methods that use minimal sample volumes and can both quantify particles and determine their biological composition are therefore highly valuable during LVV development and release testing.

Objectives and overview of the application note

This application brief demonstrates the capabilities of the Aura GT System to:

• Quantify and size visible and subvisible particles in LVV drug product candidates using backgrounded membrane imaging (BMI) with very low sample volumes (system can operate with as little as 5 µL).
• Differentiate the biological/chemical nature of large aggregates via fluorescence membrane microscopy (FMM) after staining with nucleic‑acid and protein dyes.
• Provide a practical case comparison of three commercially sourced LVV preparations (Samples 1–3) to illustrate stability ranking and to identify probable mechanisms of aggregation.

Used methodology and workflow overview

The study analyzed three commercial LVV drug candidates. Key steps included:

• Baseline imaging of empty black membrane wells to obtain background images.
• Application of sample to membrane wells (study used 20 µL per well; the Aura GT System supports analyses with as little as 5 µL for imaging workflows).
• BMI acquisition to enumerate and size particles and to collect morphological parameters.
• Post‑stain FMM imaging after application of SYBR Gold (nucleic acid dye) and thioflavin‑T (general protein stain) to distinguish nucleic‑acid–containing aggregates from predominantly proteinaceous particles.
• Automated image processing and particle characterization using Particle Vue Software with Waters’ proprietary analysis algorithms.

Used instrumentation

• Aura GT System (Waters Corporation) implementing BMI and FMM imaging modes.
• Black membrane plates for particle capture and background correction.
• Particle Vue Software for image analysis and particle sizing/morphology quantitation.
• Fluorescent stains: SYBR Gold (nucleic acid selective) and thioflavin‑T (general protein stain).

Main results and discussion

• Comparative stability: BMI images and automated analysis showed clear differences between the three LVV products: Sample 1 exhibited the fewest subvisible/visible particles and was judged most stable; Sample 3 had an intermediate profile with mostly particles <20 µm but some large particles; Sample 2 displayed extensive aggregation and was the least stable.
• Particle size and counts: Sample 2 contained the highest particle burden, including thousands of particles per mL in size bins below 25 µm and reported >310,000 particles/mL larger than 10 µm — a level well above typical USP release expectations and a major red flag for product safety and stability.
• Aggregate identity by FMM: SYBR Gold staining of Sample 2 produced strong fluorescence in many particles, indicating the presence of nucleic acid within aggregates. This supports a root cause hypothesis of LVV capsid instability or leakage of genomic material, which can nucleate aggregate formation. FMM thereby distinguished LVV‑derived aggregates from extrinsic particulate contamination (e.g., glass or plastic) or purely protein aggregates.
• Operational performance: The Aura GT workflow enabled rapid imaging (approximately one minute per sample) and required minimal sample quantities — advantageous when LVV material is limited. Particle Vue provided automated morphological metrics to support quantitative comparisons.

Interpretation, limitations, and implications

• Regulatory and formulation impact: High counts of large subvisible/visible particles (as in Sample 2) would likely fail USP particulate guidelines and require substantial formulation and/or process changes to stabilize the product prior to clinical use or release.
• Biological mechanism insights: Detection of nucleic acid in aggregates suggests biologically driven aggregation (capsid disruption, nucleic acid exposure) rather than inert foreign particulate contamination; this informs targeted mitigation strategies (e.g., buffer composition, excipient selection, process controls to preserve envelope integrity).
• Analytical scope and limitations: The described membrane‑imaging approach characterizes subvisible and visible particles (micrometer scale). It does not detect intact single LVV particles at their nominal ~100 nm diameter; therefore, it complements but does not replace orthogonal nanoscale assays (e.g., nanoparticle tracking analysis, electron microscopy, or qPCR for genomic integrity). The application brief also lacks exhaustive experimental detail (limited replicates and protocol parameters), so quantitative conclusions should be validated in a fuller experimental context.

Benefits and practical applications of the method

• Low sample consumption enables testing of scarce LVV materials during early process development and formulation screening.
• Rapid throughput and automated image analysis facilitate comparative stability ranking across multiple formulations or process conditions.
• Combined BMI + FMM approach provides both quantitative particle sizing/counts and compositional discrimination (nucleic acid vs. protein), supporting root cause analysis and targeted mitigation.
• Useful across development stages: suitable for formulation screening, in‑process monitoring, and supporting release or investigational stability assessments when used alongside orthogonal assays.

Future trends and potential applications

• Integration with orthogonal analytics: Combining membrane imaging with nanoscale and biochemical assays (e.g., NTA, SEC, TEM, PCR) will deliver a more complete stability and potency profile for LVVs.
• Stain panel expansion: Use of additional or more specific fluorescent probes (lipid dyes, capsid protein antibodies) could refine identification of aggregate composition and assembly state.
• Automation and data analytics: Enhanced software models and machine learning applied to morphological and multiparameter fluorescence data may improve classification of particulate sources and predict instability pathways.
• Application to other viral modalities: The workflow is extensible to other enveloped viral vectors (AAV, adenovirus) and to complex biologics where distinguishing nucleic‑acid–containing aggregates is valuable.

Conclusion

The Aura GT System, combined with Particle Vue analysis and targeted fluorescent stains, offers a practical low‑volume workflow to quantify subvisible and visible particulate burdens in LVV preparations and to identify aggregates that contain nucleic acid. In the presented case series, the platform rapidly discriminated relative product stability among three LVV samples and identified nucleic‑acid–positive aggregates consistent with capsid instability in the worst‑performing sample. While membrane imaging does not resolve individual intact LVV particles at the nanoscale, its capability to detect and characterize micrometer‑range aggregates with compositional information provides actionable data for formulation and process development. Complementary assays and more detailed experimental validation are recommended when using these results for regulatory submissions or definitive root‑cause assignments.

References

1. Gándara C; Affleck V; Stoll EA. Manufacture of Third Generation Lentivirus for Preclinical Use, with Process Development Considerations for Translation to Good Manufacturing Practice. Human Gene Therapy Methods. 2018;29(1):1–15.
2. Milone MC; O’Doherty U. Clinical use of lentiviral vectors. Leukemia. 2018;32(7):1529–1541.
3. Stripecke R, et al. The use of lentiviral vectors in gene therapy of leukemia: combinatorial gene delivery of immunomodulators into leukemia cells by state‑of‑the‑art vectors. Blood Cells, Molecules, and Diseases. 2003;31:28–37.
4. Jarraya B, et al. Dopamine gene therapy for Parkinson’s disease in a nonhuman primate without associated dyskinesia. Science Translational Medicine. 2009;1(2):2ra4.
5. Crespo‑Barreda A, et al. Viral and Nonviral Vectors for In Vivo and Ex Vivo Gene Therapies. In: Translating Regenerative Medicine to the Clinic. 2016:155–177.
6. Rapidly Distinguish Protein from Non‑Protein Particles in Biologic Formulations. Waters Application Note. 2025.
7. AAV Aggregate Quantitation and Identification with the Aura System. Waters Application Note. 2025.