Advancing AAV Gene Therapies Through Analytical Innovation

Key Takeaways

• Adeno-associated viral (AAV) vectors used in the formulation of gene therapies must be supported by detailed analytical data clearly demonstrating their identity, purity, potency, safety, and stability.

• Owing to the complexity and heterogeneity of AAV vectors, multiple orthogonal methods must be used to fully characterize them and ensure final products produced in different batches have consistent properties.

• The complexity, heterogeneity (full, partial, and empty capsids), and instability of AAV vectors create numerous analytical challenges, both for in-process and product release analytics.

• The best analytical strategies for AAV vectors build in flexibility so methods can be adapted to support required levels of characterization, speed, precision, sensitivity, and robustness for in-process monitoring during process development and clinical and commercial manufacturing and ultimately validation and qualification for use as release assays.

• The emergence of advanced technologies, such as mass photometry, nanopore sequencing, and nanoneedle array technology, is allowing AAV therapy developers to advance products from the lab to clinical and commercial production more quickly and cost-effectively, providing safe and effective treatments to patients sooner.

Adeno-associated viral (AAV) vectors used in the formulation of gene therapies must be supported by detailed analytical data clearly demonstrating their identity, purity, potency, safety, and stability. Specific critical quality attributes (CQAs) linking AAV characteristics to therapeutic safety and efficacy include genomic, capsid/particle, and functional/infectious titers, full/empty/partial capsid ratio, residual host-cell DNA and host-cell protein (HCP) content, replication-competent virus level, and potency.^1,2

Owing to the complexity and heterogeneity of AAV vectors, multiple orthogonal methods must be used to fully characterize them and ensure that final products produced in different batches have consistent properties. Commonly used techniques include droplet digital polymerase chain reaction (ddPCR) and quantitative PCR (qPCR), enzyme-linked immunosorbent assays (ELISA), analytical ultracentrifugation (AUC), mass photometry, cryoelectronic microscopy, and various cell-based assays.

Analytical strategies for AAVs must be designed to overcome variability in raw materials and both product and process complexity. Gathered data must support development of deep process understanding and product characterization so that material inputs, critical process parameters (CPPs), and CQAs can be linked to product performance. Effective methods are therefore required for raw material analysis (most notably cell lines, plasmid DNA, and transfection reagents), in-process testing (CPPs, such as pH, dissolved oxygen, temperature, metabolites, as well as AAV degradation, yield, purity, viral particle content, etc.), and product release.^3,4 In addition to being appropriate for these applications, they must also provide the level of data needed depending on the development stage, from preclinical through early- and late-stage clinical development to approval and commercialization.

The complexity, heterogeneity (full, partial, and empty capsids), and instability of AAV vectors create numerous analytical challenges for gene therapy developers, both for in-process and product release analytics.^2,5–7 AAV vectors exist as multiple serotypes and both upstream and downstream process conditions can impact vector properties, as can buffer and other formulation ingredients.² Consequently, AAV vector attributes can vary from vector to vector and batch to batch. The need to confirm the capsid protein composition, the nature of the encapsidated genetic material (or lack thereof), particle size, levels of product- and process-related impurities, posttranslational modifications (PTMs), and numerous other properties makes comprehensive characterization and deep process knowledge difficult to achieve.

Adaptation of analytical methods initially developed for recombinant proteins and antibodies has contributed to some of the issues, as many of these techniques do not take into account the unique requirements of AAV vector analysis. One main challenge is the often-limited availability of AAV sample material. Many traditional methods require large sample volumes.³ Separately, access to only small amounts of sample material limits the ability to conduct numerous orthogonal analyses and creates the need for effective multiplexed techniques.⁶ The extensive time required to obtain results for cell-based assays is another challenge.⁴ Methods for viral vector analysis may also suffer from inadequately understood measurement variability and lack of an easy mechanism for establishing comparability.⁶

Lack of established reference standard materials (RSMs) for assessing analytical performance is another challenge for AAV analytics.^5,6 Reference standards developed in different labs often do not provide reproducible results, making it difficult to standardize product activity and dosage potency and compare data. The constant evolution of regulatory expectations — the stringency of which is steadily increasing — also creates difficulties. Analytical capabilities are also increasing, but development of rapid, robust, reliable, and sufficiently sensitive fit-for-purpose methods that can be applied across the full development cycle remains a strong need, particularly with respect to potency assays and methods for distinguishing full, partial, and empty capsids and PTMs, detecting helper virus contamination, and confirming genome integrity (e.g., damaged inverted terminal repeats (ITRs) or damaged/mutated single-stranded DNA (ssDNA)).

Because effective AAV vector process and product development requires access to the right, high-quality analytical data, the analytical strategy for in-process monitoring and product release will directly influence program success.⁶ Optimized analytics are necessary for monitoring and controlling processes and ensuring vector products meet release specifications.⁵ Factors such as the vector serotype, batch size, and process yield must be taken into consideration when selecting specific techniques. Choosing the most appropriate methods early on will enable development of deeper process understanding and better correlation of CPPs and CQAs with product safety and efficacy.

Where possible, development of at least semi-platform methods is recommended, as doing so allows for collection and analysis of larger quantities of data across multiple batches. Fit-for-purpose methods used in research labs, which tend to be highly manual with slow throughput and high risk of variability, need to be adapted for use in a GMP setting to support clinical and commercial manufacturing as programs advance.⁵ They must evolve along with platform processes and scaling activities in a rational manner that continues to support data-driven decision-making.⁷

The best analytical strategies for AAV vectors therefore build in flexibility so methods can be adapted to support required levels of characterization, speed, precision, sensitivity, robustness, and ultimately validation and qualification for use as release assays.^5,7,8 Ideally, methods used in preclinical development can be carried through to clinical and commercial production so no extensive additional method development work or bridging studies are needed.

Comprehensive process and product characterization for AAV vectors requires use of multiple, complementary chemical and biological-based analytical methods. The main techniques used across all phases of AAV development include:²

• qPCR: Vector genomic titer, infectious titer, DNA contaminants

• dPCR: Vector genomic titer, infectious titer, DNA contaminants

• NGS: Methylation, packaged sequences, DNA contaminants

• ELISA: Capsid titer, HCPs

• MS: Capsid stoichiometry, serotype, HCPs and PTMs

• OD: Vector genomic titer, capsid titer, empty/full ratio

• AUC: Aggregation, empty/partial/full ratio

• TEM: Viral-vector titer, aggregation, empty/partial/full ratio

• AEX: Empty/partial/full ratio, capsid identity

• SEC-MALS: Viral-vector titer, empty/full ratio

• SLS/DLS: UV-visible Aggregation, viral-vector titer, empty/full ratio

Increasing automation of some more traditional methods, such as capillary electrophoresis (CE), is benefiting AAV analysis. There is also an ongoing shift from quantitative to digital PCR due to the advantages provided by the latter technique. Liquid chromatography–mass spectrometry (LC-MS) methods provide a chemical- rather than biological-based approach to evaluation of complex vector structures and identification and quantification of capsid proteins, various impurities, and PTMs. Newer methods attracting attention due to their ability to provide information on multiple important CQAs with high throughput and high resolution using small sample quantities include mass photometry, nanopore sequencing, and nanoneedle array technology.

Automated CE–sodium dodecyl sulfate (SDS) analysis enables rapid, label-free determination of AAV identity and purity on a single platform using different cartridges.⁹ One example is the Maurice™ System with Turbo CE-SDS™ and CE-SDS PLUS™ Cartridges, which provides quantitative results in 5.5–40 minutes for high-throughput R&D work and high-resolution release testing, respectively. Charge heterogeneity characterization is also possible when CE separation is combined with imaged capillary isoelectric focusing (icIEF) analysis.

Simple Western™ Technology (bio-techne®) based on automated CE western analysis coupled with immunoassay-based detection, meanwhile, allows rapid analysis of AAV lysates during downstream processing to support process optimization, providing determination of titer, capsid protein ratio, and empty/full status with a minimum number of rapid runs using just 3 μL of diluted unprocessed sample.⁹

PCR is widely used for analysis of multiple attributes of AAV vectors, including titer levels and DNA-based impurities.^10,11 It is highly standardized and thus supports comparison of samples during process development. Quantitative PCR enables quantification with high sensitivity and specificity, but requires the development of a standard curve and can be impacted by DNA amplification efficiency. Digital PCR allows for absolute quantification without the need for a standard curve and typically provides more precise, accurate, and reproducible results compared with qPCR.10 For these reasons, there is growing preference for dPCR. Droplet digital PCR (ddPCR) is most widely used because it supports detection of quantities below what is possible with qPCR and better intra- and inter-assay precision with less variation of results.¹¹ Two-dimensional ddPCR is also under investigation as a means for assessing multiple attributes simultaneously.

Unlike CE, which can detect but not quantify nucleic acids, and long-read sequencing, which has low throughput, dPCR enables accurate absolute quantification of both full and partial genomes more rapidly, with 4-plex assays providing data on genome completeness assessment and absolute copy numbers. Comprehensive genome evaluation requires selecting appropriate target regions including promoters, poly-A tails, and the 5′ and 3′ ends or targets set at regular intervals across the entire genome.¹²

With dPCR, sample preparation can impact results, so optimized, standard protocols are essential.¹¹ Use of state-of-the-art multiplex assay designs and analysis software is also essential to avoid negative impacts of cross-talk owing to sample variability and the ability to detect different fragments for multiple genes of interest.¹² One example is QIAGEN’s Cell and Gene Therapy Assays for dPCR.

Separation by liquid chromatography followed by high-resolution MS (HRMS) allows sequence mapping and characterization of capsid proteins, PTMs (e.g., deamidation, oxidation, phosphorylation, acetylation, glycosylation), and protein-based impurities.^4,13 HRMS can also be employed for genome integrity analysis. Another option is size-exclusion chromatography (SEC) combined with optical detection using UV-visible (UV-vis) or multi-angle light scattering (MALS) techniques. These methods can be used to evaluate AAV titer, particle size distribution, aggregation, and capsid full/empty ratios.

A novel Q-based anion exchange resin has also been demonstrated to efficiently separate and allow quantification of empty and full AAV capsids of numerous serotypes using choline-based mobile phase gradient salts.¹³ Interestingly, the method can also be applied at production scale for full capsid purification. Charge-detection mass spectrometry (CDMS), meanwhile, enables analysis of individual AAV particles and provides information on empty, full, partial, and overfull capsids.¹⁴

Mass photometry (MP) represents one of the newer methods attracting significant attention as an improved, rapid and reliable solution for in-process and release testing of AAV vectors.^1,8,14 It is a label-free technique that requires the use of minimal sample volumes (e.g., 10 µL) and preparation, provides results in minutes, and supports quantitation of AAV capsid populations. Like CDMS, it enables analysis of single AAV particles; the molecular mass of each particle is determined in solution via light scattering, allowing differentiation of empty and partial full, full, and overly full capsids and providing detailed insights not possible with bulk methods that evaluate overall capsid populations, such as AUC, SEC-MALS, and qPCR-ELISA.

This information can be used to assess the impact of different process conditions during process development, enabling more rapid and thorough process optimization. Acknowledgement by the United States Pharmacopeia, British Pharmacopoeia, and National Institute of Food and Drug Control in China, meanwhile, suggests MP could at least complement analytical ultracentrifugation (AUC), the current gold standard, in GMP production settings. Its rapid nature can support process monitoring and control as well as faster product release.

Next-generation sequencing (NGS) can be used to determine unknown genetic sequences, such as replication-competent AAV (rcAAV) impurities, which is not possible with PCR techniques that require primers and probes designed based on known target sequences.¹⁰ Multiplexed NGS platform methods have also been developed for AAV genomes and plasmid raw materials using a common library preparation. One issue is a lack of regulatory guidance; however, the Microarray/Sequencing Quality Control consortium at the U.S. Food and Drug Administration (FDA) is working to establish quality metrics and criteria, as well as data analysis protocols, which are complex and often customized, and other considerations needed to ensure robust, reliable application of NGS in gene therapy development and manufacturing.

Advanced particle analysis techniques enable characterization of AAV aggregation and stability, even when only limited sample volumes are available. Dynamic light scattering (DLS) combined with backgrounded membrane imaging (BMI) supports high-throughput particle detection and counting.¹³ Side Illumination Membrane Imaging (SIMI) and Fluorescence Membrane Microscopy (FMM) technology, as offered by Aura Systems, combined with BMI also use membrane microscopy, enabling particle analysis in less than one minute using just 5 µL of sample, which is not possible with traditional flow-based particle analyzers.¹⁵

Two nanoscale techniques are receiving growing attention from AAV developers: nanopore sequencing (Oxford Nanopore Technologies) and nanoneedle array technology (NanoMosaic).¹⁸ The former analyzes single DNA strands as they pass through nanopores, detecting current changes related to specific DNA bases in sequences from thousands to millions of bases long.¹⁶ Recent chemistry and software advances have improved the ability of long-read nanopore sequencing to enable in-depth analysis of mispackaged DNA in AAV particles throughout gene therapy development. Short-read sequencing in comparison offers reduced ITR coverage and cannot resolve the size of fused DNA sequences.¹⁷ The method requires minimal sample volumes, achieves 3–5 million total reads per MinION flow cell, and provides results in minutes.

The NanoMosaic Tessie™ platform, meanwhile, uses nanoneedle array technology to directly measure full-length transgene integrity and capsid titers, including determination of full and partial AAV capsid quantities, as well as semi-quantitative assessment of the degree of truncation in partial packaged DNA.¹⁸ The nanoneedle arrays interact with AAV vector particles across nearly 100,00 sensors per plate, allowing proteome-wide association studies (PWAS) and affording a femtogram/ml sensitivity — a level not possible with conventional proteomic methods. The technique is relatively high-throughput (provides full results in 3.5 –4 hours), requires small sample volumes, and can be used across early discovery through preclinical and clinical development, process development and optimization, production process monitoring, and QC and release testing using crude or purified samples. Furthermore, the NanoMosaic Tessie™ platform received the FDA’s Advanced Manufacturing Technology (AMT) designation in January 2026 for multiplex testing of vector genome and capsid titers in AAV manufacturing.

Potency of AAV-based gene therapies must be assessed in vitro and in vivo and requires deep understanding of the biological mechanism of action at the molecular level and the development of relevant cell lines and animal models.^1,5 Results must show transfer of the genetic material (viral transducibility and infectivity, tissue selectivity), and confirm the occurrence of the desired biological consequence (typically specific protein expression). The tests must be quantitative and determined against established reference standards. Analytical techniques used include ddPCR, qPCR, western blots, ELISA, immunohistochemistry, and cell-based assays.

The complexity of genetic diseases and the modes of action involved in gene therapy treatments make developing potency assays challenging. Limited availability of material for testing and the often-poor stability of AAV vectors further complicate potency assay development and implementation. The lack of suitable reference standards present further issues. The length of time required for cell-based assays is a particular problem, as they are typically used for product release and often contribute to extended wait times.

Automation of potency testing can help address many of these issues. High-throughput options that require small sample volumes and minimal operator intervention can reduce testing time and the risk of human error.⁹ One example is bio-techne’s Leo™ System leveraging its Simple Western™ Technology, which provides absolute quantification of protein content with high sensitivity (only 3 μL of sample) while analyzing up to 96 samples in three hours. The system requires a single validated antibody for detection and works with all AAV serotypes and target proteins and can be used for both early-phase research and batch-release testing of gene therapy products. Importantly, the results obtained with the Leo System have been shown to align well with those obtained using traditional flow cytometry methods.

Key to improving AAV vector process performance and product quality is having access to real-time data on CPPs and CQAs. Process analytical technologies (PATs) that enable real-time monitoring can provide important insights during process development and support effective process control during clinical and commercial production. At-line, online, and in-line analytics that provide rapid non-destructive measurements of relevant CQAs is necessary to implement effective PAT solutions for AAV vectors.³

In-line Raman spectroscopy —when used in conjunction with chemometric models —can support real-time monitoring and control of some CPPs, but does require advanced software or expertise in calibration curve generation and data analysis. MP and nanopore sequencing are promising options for real-time monitoring of important AAV capsid properties, as is membrane microscopy for particle size determination. Waters has introduced ultraDAWN™ real-time MALS technology for in-process monitoring of particle characteristics including weight-average molar mass, z-average size (radius), polydispersity, particle concentration (physical titer), and payload attributes.¹⁴

Other methods of interest, particularly for cell characterization, include electrical impedance, various imaging techniques, UV-Vis and near-infrared (NIR) spectroscopy, and Laser Force Cytology™ from LumaCyte (based on the behavior of cells when exposed to precisely controlled laser light and microfluidic flow).³ These methods are attractive because they are label-free, non-destructive, and rapid. Combination of these approaches with microfluidic devices, flow cells, and dynamic sample platforms could improve their applicability as PAT solutions for AAV vector manufacturing.

Researchers at the CGT Catapult are seeking to identify means for correlating data obtained using in-line techniques such as Raman spectroscopy with results obtained using high-throughput, high-content screening methods, such as flow cytometry, gene expression profiling, LC-MS, and NGS.³ The data could be used to construct digital twins for use in predicting and controlling AAV vector processes. Other work is focused on identifying surrogate biomarkers that can be used to determine CQAs that currently cannot be measured in real time.

AAV-based gene therapies cannot reach patients without being proven safe and effective. That evidence can only be obtained through the use of robust, reliable analytical methods throughout the entire life cycle of these advanced products. Traditional methods that enable determination of important AAV CQAs tend to require tradeoffs between speed, reliability, accuracy, and depth of insight. Fortunately, newer rapid, high-resolution, multiplexed analytical technologies are enabling more comprehensive characterization of AAV vectors at earlier development stages, contributing to more efficient development of more manufacturable candidates and optimal production processes and higher-quality gene therapy products with reduced risk.¹⁸

The emergence of advanced technologies, such as MP, nanopore sequencing, and nanoneedle array technology, us allowing AAV therapy developers to advance products from the lab to clinical and commercial production more quickly and cost-effectively, which ultimately means patients have access to safe and effective treatments sooner.⁷

Further advances in AAV vector analytical technologies are expected going forward. In particular, more rapid techniques that require small sample sizes, greater automation, and more multiplexing — all with maintenance of reproducibility, specificity, and sensitivity — will contribute to more efficient product and process development and product release.^3,15 Use of these methods at earlier development stages will, meanwhile, minimize risk and increase the likelihood of success. Incorporation of artificial intelligence (AI) tools has the potential to facilitate more rapid method development and analysis of data generated by complex techniques.¹⁰

1. Russell, Erika. “Analytical Release Testing for Viral Vectors: AAV and Lentiviral Quality Control.” Technology Networks. 24 Jun. 2026.

2. Kontogiannis, Theodoros, et al. “Characterization of AAV vectors: A review of analytical techniques and critical quality attributes.” Molecular Therapy – Methods & Clinical Development. 32(3): 101309 (2024).

3. Pachoumi, Olympia. “Unlocking analytics in cell and gene therapy: from complexity to control.” Team Consulting Insights. 25 Jul. 2025.

4. “Current State of Gene Therapy Development and Key Process & Analytical Challenges.” Emery Pharma. 9 Dec. 2025.

5. Gacheru, Stephen. “Challenges In Analytical Development And Testing For Gene Therapies.” Cell & Gene. 13 Oct. 2021.

6. Chadwick, Jennifer. “Overcoming Analytical Challenges for Gene Therapies.” ProtaGene. Accessed 12 Oct. 2026.

7. Foster, Bree. “How advanced analytics are reshaping gene therapy development.” Drug Discovery News. 16 Mar. 2026.

8. Andelyn Biosciences. “Identifying The Right Analytical Strategy For AAV Production: A Q&A With A CDMO Expert.” Cell & Gene. Accessed 12 Oct. 2026.

9. “Analytical Technologies for Gene Therapy.” Bio-Techne. Accessed 12 Oct. 2026.

10. Blay, Emmanuel, Elaine Hardyman, and Wesley Morovic. “PCR-based analytics of gene therapies using adeno-associated virus vectors: Considerations for cGMP method development.” Molecular Therapy – Methods & Clinical Development. 31: 101132 (2023).

11. Russell, Erika. “Analytical Release Testing for Viral Vectors: AAV and Lentiviral Quality Control.” Technology Networks. 24 Jun. 2026.

12. QIAGEN. “A Q&A on Viral Vector Integrity Analysis: Advancements in Gene Therapy Development.” ISCT Telegraft Hub. 8 Oct. 2025.

13. “Viral Vector Solutions.” Waters. Accessed 12 Oct. 2026.

14. Victoria, Maria Jacintha. “Transformative AAV analytics.” LinkedIn Pulse. 23 May 2025.

15. Mason, Donald. “The Importance of AAV Characterization in Ensuring Safe Gene Therapy Development.” Waters. 4 Jan. 2024.

16. Dunker-Seidler, Florian, et al. “Recombinant AAV batch profiling by nanopore sequencing elucidates product-related DNA impurities and vector genome length distribution.” Molecular Therapy – Methods & Clinical Development. 33(1): 101417 (2025).

17. Breunig, K, et al. “The Good, the Bad, & the Chimeric: Using Nanopore Sequencing to De-Risk rAAV Genome Packaging.” Ascend Advanced Therapies, ASGCT 2025, Poster No. 829. 15 May 2025.

18. Andelyn Biosciences. “A Platform For Advanced AAV Analytics.” Cell & Gene. 27 May 2026.