Decoding LNP Characteristics: From CQAs to Better RNA Therapeutics

Why rigorous characterization matters, how to measure it, and when to act on it?

Why this matters. Missed or poorly controlled CQAs (critical quality attributes) are the fastest route to delays in the lead development, batch rework, and clinical risk for RNA/LNP assets.

Non-negotiables to lock early. Size & PDI, encapsulation efficiency, RNA identity/integrity, lipid identity/content, sub-visible particles, sterility/endotoxin.

Morphology. Important but not yet a universal release CQA. Treat as pCQA (potential CQA) and link to potency/stability with orthogonal methods. Morphology should be understood more broadly as a function of rational design choices.

Ask your team/CDMO. Are CQAs linked to Critical Process Parameters (CPPs) & the Quality Target Product Profile (QTPP)? Do orthogonal methods align? How is morphology tracked lot to lot? Are we benchmarking to shared references (BioPhorum, EU, US-NCL cascades, NIST RGTMs)?

This article is for the innovators who turn RNA science into products: biotech executives and BD teams pressure-testing CMC plans, analytical/formulation scientists (including those moving from academia to industry) choosing fit-for-purpose assays, CMC/QA/QC leaders writing specs, process-development and manufacturing teams (in-house or CDMOs) scaling mRNA–LNPs, and program managers who must link CPPs→CQAs→QTPP with defensible data. This article provides the guidance towards a practical playbook: what to measure (release-critical vs. characterization-only), how to measure it with orthogonal methods, how to treat morphology (incl. “blebs”) as a pCQA, and how to align work with regulatory guidelines.

CQAs in LNP design: Measuring what matters, when it matters

CQAs (Critical Quality Attributes) are the measurable properties of your product that must be controlled to ensure safety, efficacy, and quality.

Once you list candidate CQAs, you classify each one into two buckets:

Release-critical CQA

·       Appears on the certificate of analysis (CoA) with a numerical acceptance criterion.

·       The method is validated/qualified for its intended use.

·       If the batch fails this spec, the batch cannot be released.

Characterization-only CQA (often called pCQA = potential CQA)

·       You measure and trend it to build understanding, compare lots, link to the process, and inform specs.

·       It does not gate release (no CoA limit yet), because clinical relevance, control leverage, or method maturity is still being established.

·       It can be promoted to release-critical later when you have a robust CPP→CQA→clinical link and a defined method.

Figure 1 – a gif representing Important questions on LNP CQA

How teams decide which bucket a CQA belongs to

Promote an attribute to release-critical when most of these are true:

• Strong rationale that it impacts patient safety or clinical performance (potency, safety, PK/BD, storage stability).
• Clear process leverage (you can control it via CPPs).
• Method is suitable (accuracy/precision/LoD (limit of detection)/LoQ (limit of quantification) known; matrix effects understood).
• Attribute is stable enough to set a realistic, defendable spec (edge-of-failure understood).
• Regulators expect it for the dosage form/route (e.g., sterility/endotoxin for parenteral).

Keep as characterization-only/pCQA when:

• Clinical link is plausible but not fully defined yet (e.g., morphology/“blebs”, zeta-potential in some cases).
• Method is early (operator-dependent, artifacts, no reference standard).
• You can’t control it reliably yet (no robust CPP linkage).
• Variability is too high to set a meaningful spec.

Developability means translating the Quality Target Product Profile (QTPP) into tractable CQAs (critical quality attributes) and fit-for-purpose assays. For mRNA–LNPs, the BioPhorum cross-company guidance (Pfizer, Roche, Janssen, Regeneron, etc.) now offers an actionable baseline of which attributes to control at DS (drug substance)/DP (drug product), when to test (release vs. stability), and with which classes of methods. Briefly:

• Strength/Identity/Potency. RNA sequence identity; lipid identity & content; potency (validated cell-based or surrogate), at release and on stability as risk dictates.
• Product quality & characteristics. Size, PDI, zeta potential, encapsulation efficiency, morphology - directly influence biodistribution, uptake, safety/efficacy. Use orthogonal tools; do not lean on a single technique.
• Purity/impurities. In vitro transcription (IVT) residuals/fragments; lipid impurities/degradants.
• Safety. Sterility, endotoxin, bioburden, Container Closure Integrity (CCI).

Characterization that stands up in due diligence (and in front of regulators)

The field is not short on techniques — the challenge is picking the right combination, validating them, and ensuring orthogonality (complementary methods that converge on the same answer).

Regulators are clear: any method used for CQAs must be validated for accuracy, reproducibility, and detection limits (ICH, FDA, EMA). This principle, long established in monoclonal antibody (mAb) development, now applies to RNA–LNPs. Unlike antibodies, which are single-component, globular proteins, LNPs are multi-component assemblies composed of RNA, lipids, and potentially other components.

Figure 2 – a gif showcasing characterization techniques

Application of biophysical techniques

Size & polydispersity (DLS, MADLS, NTA, MALS):

• What they measure: particle size distribution, aggregation, and colloidal stability. 
• Experience in mAbs: used mainly to track aggregation; antibodies are isotropic scatterers so DLS alone is less informative. 
• For LNPs: essential, but orthogonal confirmation is beneficial; Might need to combine the detector with a separation technique to have more precise information about particle populations (e.g. AF4, SEC, AUC). AF4 + MALS emerging as a gold standard for heterogeneous populations. 

Charge (ELS (Zeta-Potential), Capillary Electrophoresis, CiEF):

• What they measure: colloidal stability, toxicity.
• mAbs parallel: CiEF is a mainstay to map charge variants; for LNPs, surface charge is more dynamic and also buffer-dependent, making interpretation harder, Zeta-Potential measure through ELS is the golden standard.

Morphology (Cryo-TEM, SAXS/SANS):

• What they measure: structure, lamellarity, bleb formation, bilayer organization. 
• mAbs parallel: morphology characterization is largely irrelevant for antibodies;  for LNPs, it is central. Cryo-TEM and SAXS are the only ways to directly visualize intraparticle characteristics and structural heterogeneity and thus, allow deep insight into LNP morphology. NanoVation’s approach integrates these tools early in formulation development to inform rational design and ensure consistency of structure.

Chemical stability of mRNA (RP-HPLC- UV, LC-MS, AEX-UV, CGE) and lipids (HPLC-CAD, LC-MS). 

• What they measure: lipid composition, degradation, RNA integrity. 
• mAbs parallel: chromatography is foundational for glycosylation and variant mapping. for LNPs, lipid as well as RNA degradation and ratios require equally careful chromatographic validation.

Encapsulation efficiency (RiboGreen, dye assays, IP-RP,AEX UPLC-UV,- SECFFF-MALS-UV):

• What they measure: ratio of encapsulated vs. free RNA.
• mAbs parallel: no analogy, but regulators treat EE as analogous to protein content/purity.. for LNPs EE is critical for dosing accuracy, stability, and functional delivery. Poor EE can result in degradation of unencapsulated RNA and variable in vivo performance.

Computational tools (MD simulations):

• What they measure: lipid–RNA interactions, lipid packing, endosomal escape hypotheses.
• mAbs parallel: widely used for antibody folding and stability predictions;. for LNPs still early but promising.

 Empty VS mRNA-filled LNPs (Nano-Flow Cytometry, AUC)

• What they measure: ratio of empty LNPs Vs LNPs containing mRNA
• For mAbs not relevant
• Emerging issue in the LNP field, growing interest.

Cutting through complexity

With so many methods available, the challenge for R&D teams is not knowing what exists but knowing what matters. To make this easier, NanoSphere has compiled a side-by-side comparison table of all key techniques - showing what each measures, where it excels, where it fails, and how it has been applied to both LNPs and mAbs.

Table 1 – mRNA–LNP Characterization Matrix: Property-to-Technique Map

Choosing the right assay is hard - there are dozens of overlapping techniques, each with trade-offs. We condensed 30+ LNP characterization methods into a one-page matrix showing what each measures, how it works, key pros/cons, and a QC-usability rating, so you can decide fast and defend choices in QbD/CMC.

Figure 3 – Methods for LNP Characterization

Morphology & structure - signal, noise, or lever?

From curiosity to differentiated technology

To be clear: Structure isn’t noise - it’s a lever. Rational LNP design must incorporate morphology early, not just to understand what’s made, but to make what’s needed. With LNPs moving beyond the liver and toward lower doses, longer circulation times, and precision cell targeting, structure-function relationships matter more than ever. Morphology - including bilayer structures, blebs, empties, and lamellarity - should be treated as a potential CQA (pCQA). That means:

• Monitor and trend morphology using orthogonal methods.
• Correlate to potency and stability outcomes where possible.
• Avoid locking morphology as a release spec until robust correlations are demonstrated.

The key is not whether specific structures are “good” or “bad,” but whether your team is tracking them systematically and exploring links to function. What is curiosity today may become tomorrow’s differentiator.  At NanoVation Therapeutics™, this principle underpins the development of the lcLNP technology, a platform designed to extend systemic circulation and support extrahepatic delivery: Unlike conventional solid-core LNPs that exhibit a lipid monolayer around a compacted RNA core, lcLNPs adopt a liposomal-like morphology — with a defined lipid bilayer structure similar to classical liposomes. These particles resemble "fried egg" or "fish-eye" morphologies under cryo-TEM, an indication of bilayer formation around a hydrated or partially structured core.

The transition from solid-core to bilayer structures is driven by rational formulation design, specifically by increasing the helper lipid content. This compositional adjustment promotes bilayer formation, resulting in a distinct morphology that results in longer systemic exposure, reduced protein corona formation, and lower clearance rates.

By intentionally designing liposomal-like structures, morphology is not just as a structural curiosity, but a performance lever. Extended circulation improves the probability of reaching tissues beyond the liver.

The “bleb” question: Few topics in LNP morphology spark as much debate as “bleb-like” structures - particles with protruding or irregular compartments. What we know (and don’t).

• Potency link? Some studies report that mRNA-rich blebs enhance potency, potentially by modifying release kinetics or shielding cargo in unique ways.
• Instability risk? Other findings suggest the opposite: blebs correspond with reduced stability, faster degradation, or inconsistent activity, especially under serum conditions.

Why the conflict? The outcome depends heavily on buffer composition, pH-shift kinetics, and ionic strength during formulation. Small variations, e.g., buffer type and molarity, pH-shift kinetics (dialysis vs. dilution vs. TFF), serum exposure, and whether mRNA localizes inside or outside the bleb, can completely change outcomes. A recent perspective urges standardizing quantification (single-particle and cryo-TEM strategies) and separating “research handling” from QC. (Simonsen, J. Control. Release, 2024.) The emerging consensus: It’s also crucial to distinguish between blebs generated intentionally and uncontrolled blebs that emerge as a sign of instability.

Scientist’s playbook - quantifying morphology

Define morphology classes (e.g., compact core, lamellar, bleb, empty) and train analysts on blinded sets.

Acquire ≥3 grids × ≥5 images/grid; classify ≥500 particles per lot for robust prevalence estimates.

Report by-number metrics (not intensity-weighted).

Trend vs e.g., buffers/pH/serum challenge; overlay potency and stability outcomes.

From Parameters to performance - Turning design into decisions

A practical guide to CPP→CQA→QTPP mapping

Designing effective LNPs isn’t just about what goes in the vial. It’s about knowing which process and product attributes actually drive clinical performance. The CPP→CQA→QTPP framework offers a way to align formulation and manufacturing decisions with therapeutic goals like dosing frequency, organ targeting, and delivery route. By linking Critical Process Parameters (CPPs) to Critical Quality Attributes (CQAs) and ultimately to the Quality Target Product Profile (QTPP), teams can build control strategies that are scientifically sound, scalable, and ready for regulatory scrutiny.

CPP→CQA→QTPP mapping

• QTPP endpoints (delivery route, dose/frequency, target organ)
• CPPs that move the needle (flow-rate ratio, total flow, lipid concentration, lipid:RNA, buffer/pH, in-line dilution, TFF strategy)
• CQAs locked to release vs. characterization vs. pCQA
• Orthogonality matrix (1 fractionation + 1 imaging + 1 functional)

Escalation rule. Promote any attribute to release CQA only when it shows a reproducible CPP→CQA→ performance link, with a control window you can hold in GMP. Morphology becomes a defensible lever when backed by structural insights and robust correlations to function as implemented in platforms like NanoVation’s lcLNPs, where morphology supports pharmacokinetic goals and enables extrahepatic tissue delivery.

Decision frameworks you can use tomorrow

Figure 4 – Quality by Design Workflow Cycle

Method-selection logic

1. Screen: DLS/NTA (size/PDI/concentration) + RiboGreen EE% + ELS ZP
2. Resolve heterogeneity: AF4-MALS-DLS (Rg/Rh) ± SEC (aggregates)
3. Analyze morphology: cryo-TEM (classes; by-number) ± SAXS/SANS (bulk structure)
4. Verify composition & integrity: LC-CAD/MS (lipids), IP-RP/CGE/AEX (RNA)
5. Function: potency + stability (forced & real time); correlate morphology pCQA

What to Ask Your CDMO (executive sidebar)

• CQAs defined. Which attributes are release-critical vs. characterization-only?
• Orthogonality. Are at least two complementary methods (fractionation + imaging) used?
• Morphology. How are structural features monitored and tied to potency/stability?
• Standards. Are assay choices aligned with EU-NCL/US-NCL cascades and BioPhorum? Are we benchmarking to shared references (e.g., NIST RGTMs)?
• Scale transition. How do analytics transfer from R&D to GMP (controls, acceptance ranges, reference standards)?

As next-gen LNPs move toward extrahepatic and cell-targeted applications, rigorous characterization including understanding and engineering morphology will be central to unlocking functional performance. Take a look at the two real-life examples below to make this article easier to understand:

Figure 5 – IM RNA-LNP vaccine, liquid at 2-8 °C

Figure 6 – IV RNA-LNP therapeutic, lyo drug product (DP)

Ready to dive deeper? If you haven’t yet begun with our foundational overview of nanomedicine basics, start here with our first article - then deciphering the science (and the art) behind LNP technologies: from the ionizable and helper lipids that give LNPs their biocompatibility and endosomal-escape power, to the PEG-lipid and cholesterol components that stabilize particle structure; we’ll also walk you through nucleic acid therapeutics – check it out here The Rise of Nucleic Acid Therapies — NanoSphere.

Coming in October: How LNP pharmacokinetics, biodistribution, and the protein corona shape delivery success. Stay tuned!

Written by

Dr. Dominik Witzigmann 

Dominik is an entrepreneurial scientist with deep expertise in nanomedicines and nucleic acid delivery. In 2024, he was named Highly Cited Researcher, recognizing him among the world’s most influential researchers in the field. Dominik obtained his Ph.D. in Pharmaceutical Technology from the University of Basel in Switzerland, and held research positions at leading institutions including University College London (safety/tox), German Cancer Research Center (RNAi & cancer), University of Basel (targeted nanomedicines & DNA delivery) and the University of Zurich (mRNA-based genome editing). To focus on extrahepatic RNA delivery, he later joined the team of Prof. Pieter Cullis at the University of British Columbia. Dominik has held leadership roles in Canada’s NanoMedicines Innovation Network (NMIN) and served on the Board of the CRS Gene Delivery and Genome Editing Focus Group. To translate next-generation LNP technologies into the clinic, Dominik co-founded and leads the LNP-nucleic acid company NanoVation Therapeutics. 

Dr. Jayesh Kulkarni

Dr. Kulkarni obtained his PhD from the University of British Columbia and has over 15 years experience in the nanoparticle drug delivery field. He has published over 40 peer-reviewed articles and is a co-inventor on numerous patents. Dr. Kulkarni’s research has focused on the role of the various lipid components in LNP and the biophysics that governs particle formation. His work has contributed to clinical translation, including scale-up and manufacturing of LNP systems in accordance with GLP and GMP regulations. Dr. Kulkarni is a leader in the design and development of lipid nanoparticle (LNP) formulations of small molecule and nucleic acid therapeutics. He currently serves as the Chief Scientific Officer of NanoVation Therapeutics, an LNP-RNA formulation developer.

Daniel Kurek, MSc

Daniel has over 10 years experience in organic synthesis, formulation development, and scale-up of LNPs. He started his career in lipid nanoparticles at Evonik Canada in formulation and process development. In his current role as Associate Principal Scientist, Formulation – Team Lead at NanoVation Therapeutics, he oversees the development of novel long-circulating LNPs for targeting extrahepatic tissue, improving the potency and characteristics of LNPs encapsulating various nucleic acid payloads, and is a co-inventor on several of NanoVation’s key patents. Daniel holds an MSc in organic chemistry from the University of British Columbia.

Dr. Allegra Peletta

Allegra is a vaccine formulation scientist and Project Manager at the Vaccine Formulation Institute. She previously completed a postdoctoral fellowship at Ghent University in collaboration with CEPI, following her PhD in Pharmaceutical Sciences at the University of Geneva. Her research focused on accessible intramuscular vaccine formulations, with a strong emphasis on the characterization and stability of mRNA–LNPs and freeze-dried platforms. She has worked across multiple delivery systems, vaccine types, and adjuvants, with a particular commitment to formulations designed for Low- and Middle-Income Countries. During the COVID-19 pandemic, she was the only researcher still active in her Geneva laboratory, contributing directly to the development of COVID vaccines. Allegra combines rigorous scientific expertise with strong organizational and project management skills, having led and collaborated with interdisciplinary and international teams.

Dr. Marija Petrovic 

Marija is a pharmacist with a PhD in Biopharmacy from the University of Geneva, and a cancer research (ISREC)–trained professional through EPFL, with over seven years of experience in nanomedicine. During her PhD, she worked on miRNA and STING ligand nanocomplexes for cancer immunotherapy, gaining deep expertise in nanoparticle characterization and translational workflows. Certified by the EU-NCL in nanobiotechnology and awarded by Innosuisse (Swiss Innovation Agency) with two prizes (jury and public) for the best life science project, she also earned support from FONGIT, Geneva’s leading deep-tech incubator. As the founder of NanoSphere and an active contributor to the Controlled Release Society (Communication Chair for the Gene Delivery and Editing Group (GDGE), and Industry representative at Nanomedicine and Nanoscale Delivery (NND)), Marija focuses on making next-gen medicine scientific advances more visible, understandable, and useful to the communities that can turn them into impact.