Highly modified nucleic acid therapeutics require analytical strategies that go beyond simple identity or purity testing.
Impurity profiling for oligonucleotide therapeutics must account for deletion/addition sequences and modified backbone, sugar, and nucleobase variants.
mRNA analytics must address sequence confirmation, integrity, 5′ capping, poly(A) tail attributes, fragments, and double-stranded RNA.
Orthogonal techniques, such as chromatography, liquid chromatography–mass spectrometry, high-resolution mass spectrometry, and ion mobility mass spectrometry, can provide complementary insight.
Fit-for-purpose analytical validation is central to developing robust control strategies for nucleic acid therapeutics.
The Rising Analytical Burden of Modified Nucleic Acid Therapeutic
Nucleic acid therapeutics have expanded well beyond the earliest concepts of synthetic strands designed to bind RNA or encode protein. Today, the category includes multiple therapeutic oligonucleotide classes, including antisense oligonucleotides (ASOs), small interfering RNA (siRNA), microRNA, aptamers, and immunostimulatory oligonucleotides. Approved oligonucleotide therapeutics have incorporated chemical modifications, such as 2′-fluoro-RNA, 2′-O-methyl RNA, phosphorothioates, 2′-O-methoxyethyl RNA, and phosphorodiamidate morpholinos.1 Delivery has also evolved, with lipid formulation and GalNAc conjugation supporting efficient delivery and robust, long-lasting gene silencing for modified oligonucleotides.
That progress has changed the analytical burden. Highly modified nucleic acid therapeutics are not defined only by base sequence. They may include backbone modifications, sugar modifications, nucleobase modifications, conjugated delivery ligands, stereochemical heterogeneity, or structural elements such as a 5′ cap and poly(A) tail. For sponsors and contract development and manufacturing organizations (CDMOs), the central analytical question is no longer simply whether a product has the expected identity or meets a purity threshold. It is whether the analytical strategy can show what was made, what related species are present, how those variants are controlled, and whether methods are fit for their intended use.
Synthetic oligonucleotides pose unique scientific and regulatory challenges.2 That statement reflects a reality now visible across the field: the same modifications that improve drug-like properties can also complicate characterization. For nucleic acid therapeutics, analytics is therefore not a late-stage release-testing exercise. It is an enabling component of product understanding, process development, control strategy, and life cycle management.
Nucleic Acid Therapeutics Are Outgrowing Simple Analytical Models
Traditional analytical models are strained by the diversity of modern nucleic acid therapeutics. ASOs, siRNA, microRNA, aptamers, and immunostimulatory oligonucleotides differ in structure and intended function, while approved oligonucleotide therapeutics already span multiple modification chemistries.1,3 The analytical strategy suitable for one product may not fully address the risks of another.
For example, phosphorothioate chemistry introduces one kind of challenge; GalNAc conjugation introduces another; mRNA products introduce still others. A short, chemically modified oligonucleotide may require detailed impurity profiling and sequence confirmation. A GalNAc-conjugated siRNA may require attention to conjugation, duplex architecture, and phosphorothioate diastereomers. An mRNA product may require control of sequence, 5′ cap, untranslated regions, poly(A) tail, integrity, fragments, and double-stranded RNA.4,5
This diversity makes a platform mindset useful but not sufficient. Sponsors and CDMOs can benefit from established analytical toolkits, but the final control strategy must be product specific. The relevant question is not which analytical technology is most advanced in the abstract but which combination of methods can adequately characterize the product’s critical attributes and process-related risks.
The Impurity Challenge: Sequence Variants, Chemical Variants, and Process-Related Species
Impurity profiling is especially challenging for modified oligonucleotide therapeutics because impurities may be closely related to the intended molecule. The U.S. Food and Drug Administration (FDA) has identified product-related synthetic oligonucleotide impurities that can include deletion/addition sequences and modified backbone, nucleobase, and sugar variant.2 These species are not generic contaminants. They may be near-neighbor molecules that differ by a missing residue, an added residue, or a change in a chemical modification.
Phosphorothioate oligonucleotides illustrate the scale of the challenge. Therapeutic phosphorothioate oligonucleotides present analytical difficulties because of relatively large molecular size, diastereoisomeric nature, and complex impurity profiles.6 Modified small RNA-based therapeutics add further complexity, as second- and third-generation chemical modifications can include backbone, 2′ sugar, and base modifications.7
For these molecules, impurity characterization must do more than separate a main peak from related peaks. It must help determine what those related species are. Are they truncated sequences? Extended sequences? Backbone variants? Sugar variants? Nucleobase variants? Low-abundance impurities that co-elute or overlap with expected product signals? The answer matters because different impurity classes may require different process controls and different analytical methods.
Sequence fidelity is part of this same problem. For modified nucleic acids, confirming a nominal molecular weight is not necessarily equivalent to confirming the correct molecule. Ion-pair reversed-phase liquid chromatography coupled to high-resolution tandem mass spectrometry can support sequence verification and confident identification of low-abundance impurities in modified single-stranded antisense oligonucleotides.7 That type of sequence-level confirmation becomes increasingly important as modification density increases.
The analytical goal is therefore not only to measure purity but to generate molecular understanding. A robust impurity profile should help connect analytical signals to specific product-related or process-related species and give development teams the information needed to refine processes, justify specifications, and support comparability over time.
mRNA Analytics: Integrity, Cap, Tail, UTRs, and dsRNA
mRNA-based products require a different but equally demanding analytical lens. For mRNA, sequence confirmation remains important, but it is not sufficient. The World Health Organization (WHO) states that the correct sequence of mRNA should be confirmed and defines mRNA integrity as the proportion of mRNA that is the correct size and contains the 5′ cap and poly(A) tail.4 That definition makes clear that mRNA quality depends on both sequence and structural completeness.
Regulatory considerations for mRNA products also extend to defined structural features. The European Medicines Agency’s (EMA) draft mRNA vaccine quality guideline says the full mRNA sequence should be provided, including the 5′ cap, 5′ and 3′ untranslated regions, and poly(A) tail; the 5′ cap structure should be described; and nucleoside modifications should be indicated.5 The WHO identifies control needs including mRNA integrity, 5′ capping efficiency, 3′ poly(A) tail presence or length, percentage intact mRNA, percentage mRNA fragments, and percentage double-stranded RNA.4
These attributes make mRNA analytics multidimensional. A product can have the intended encoded sequence but still raise quality concerns if the transcript is fragmented, incompletely capped, missing expected structural elements, or associated with double-stranded RNA. The EMA’s draft mRNA vaccine guideline identifies mRNA integrity as a purity/impurity consideration and states that the proportion of intact full-length mRNA should be determined using an appropriate method, such as capillary gel electrophoresis, agarose gel electrophoresis, or ion-pair reversed-phase high-performance liquid chromatography (IP-RP-HPLC).5
The broader lesson for nucleic acid therapeutics is that each modality carries its own definition of “correct.” For an mRNA product, correctness includes sequence, size, integrity, cap structure, untranslated regions, and poly(A) tail attributes. For a modified oligonucleotide, correctness may focus more on sequence, modification pattern, impurity profile, and stereochemical or conjugation-related features. Analytical strategy must reflect these modality-specific definitions rather than treating all nucleic acid products as variations on the same test article.
Orthogonal Methods for a Multidimensional Analytical Problem
No single analytical method can resolve every relevant attribute of a highly modified nucleic acid therapeutic. Analytical approaches for oligonucleotide characterization include anion exchange chromatography, IP-RP-HPLC, mixed-mode chromatography, hydrophilic interaction liquid chromatography (LC), two-dimensional liquid chromatography, and mass spectrometry (MS).3 Each approach can offer different selectivity, resolution, or structural information.
Ion-pair LC–MS combining ultraviolet and mass spectrometry quantification has been developed and validated for more than 35 oligonucleotide drug substances and products, including commercialized drugs.6 That example highlights the value of pairing quantitative and molecularly informative readouts. Ultraviolet (UV) detection can support quantitation, while MS can provide identity and impurity information that chromatographic retention time alone may not supply.
High-resolution MS can add another layer of resolving power. The FDA has shown that high-resolution MS can resolve impurity information that unit- or low-resolution MS cannot, including coexisting n-U and n-C isobaric impurities and their ratio.2 For highly modified nucleic acid products, where closely related impurity species may be present at low abundance, that added mass accuracy and resolving power can be central to product understanding.
The practical implication is that orthogonality should be built deliberately. Chromatography can separate species based on charge, hydrophobicity, polarity, or other properties. Mass spectrometry can help determine molecular identity. High-resolution mass spectrometry can distinguish species that lower-resolution approaches may not resolve. Gel-based methods may be appropriate for certain mRNA integrity questions. The strongest analytical strategy is not necessarily the one with the largest number of methods, but the one in which each method answers a specific product-relevant question.
Stereochemical and Conjugation Complexity: Phosphorothioates and GalNAc-siRNA
Phosphorothioate chemistry is a major example of how therapeutic performance and analytical complexity can be linked. Phosphorothioate modification is widely used in oligonucleotide therapeutics but produces diastereomers that may have distinct physicochemical and pharmacological properties.8 In siRNA, phosphorothioate modification improves serum stability by increasing resistance to nuclease digestion, but it also creates a chiral center at phosphorus, generating diastereomers.9
That stereochemical heterogeneity is analytically challenging. Techniques used to characterize phosphorothioate oligonucleotide diastereomers include IP-RP chromatography, anion exchange chromatography, mixed-mode chromatography, hybrid LC approaches, capillary electrophoresis, spectroscopy, and other methods.8 The range of methods reflects the difficulty of resolving and understanding these species.
GalNAc-conjugated siRNA adds another layer. GalNAc-siRNA conjugates enable targeted liver delivery because GalNAc sugars are recognized by the asialoglycoprotein receptor carbohydrate recognition domain, and ASGPR is predominantly expressed on hepatocytes.1 Products that employ GalNAc conjugation chemistry include GIVLAARI, LEQVIO, OXLUMO, and AMVUTTRA. This delivery strategy has become a significant part of the nucleic acid therapeutic landscape, but it also creates products that can combine conjugation, duplex structure, chemical modification, and stereochemical diversity.
Separation and characterization of phosphorothioate diastereomers in GalNAc-conjugated siRNA is analytically challenging because of the high number of isomers present. A 2025 comparative study evaluated IP-RP, anion exchange chromatography, hydrophilic interaction LC, and ion mobility MS for diastereomer separation in GalNAc-conjugated siRNA.9 Among the chromatographic techniques, anion exchange chromatography provided the highest diastereomer selectivity, but none of the LC methods completely separated all diastereomers for the selected antisense and sense strand reference compounds. Fragment-based ion mobility MS allowed separation of all antisense-strand diastereomers and partial resolution of the sense strand in the gas phase, and the authors concluded that ion mobility MS can serve as a promising complementary approach to established LC-UV methods for qualitative and quantitative diastereomer analysis.
This study is useful for oligonucleotide drug developers and CDMOs because it shows why advanced nucleic acid analytics cannot be reduced to a single platform decision. GalNAc-conjugated siRNA can require attention to delivery conjugation, strand-specific attributes, phosphorothioate chemistry, and diastereomeric composition. Analytical strategy must therefore be designed around the full architecture of the molecule.
Emerging Manufacturing Routes: Enzymatic RNA Synthesis and New Impurity Questions
Manufacturing innovation can change analytical requirements as much as product design does. RNA oligonucleotides have emerged as a therapeutic modality, and current manufacturing methods may not meet anticipated future demand. In a 2024 paper published in Nature Biotechnology, the authors developed an aqueous-based, template-independent enzymatic RNA oligonucleotide synthesis platform as an alternative to traditional chemical methods.10
The platform used reversible terminator nucleotides with a common 3′-O-allyl ether blocking group and CID1 poly(U) polymerase mutant variants. The authors demonstrated production of natural and therapeutically relevant modified sequences over 10 liquid-phase synthesis cycles. These findings support the idea that enzymatic approaches may become part of the future manufacturing toolkit for RNA oligonucleotides.
However, new manufacturing routes also introduce new impurity questions. In solid-phase enzymatic RNA oligonucleotide synthesis, they observed N+4 and N+6 products as primary impurities, potentially arising from incomplete enzyme access to surface-bound oligonucleotide or nucleoside triphosphate carryover from inadequate washing. They also found that phosphorothioate backbone modifications were not tolerated by the allyl ether deblocking chemistry and that 3′-O-azido-methyl ether blocking may be a viable alternative.
The implication is not that enzymatic synthesis is analytically problematic; rather, it is that each manufacturing route creates its own impurity landscape. Chemical synthesis, enzymatic synthesis, in vitro transcription, conjugation, and formulation can each introduce different quality questions. As new platforms mature, analytical development will need to identify which impurity classes are expected, which are unexpected, and which methods are best suited to detect and characterize them.
Building a Fit-for-Purpose Analytical Strategy for Sponsors and CDMOs
The regulatory and development challenge is to build analytical strategies that are scientifically justified rather than maximally complex by default. ICH Q2(R2) provides guidance on validation of analytical procedures and selection and evaluation of validation tests, and it states that the objective of analytical procedure validation is to demonstrate that the procedure is fit for its intended purpose.11 That fit-for-purpose principle is especially important for nucleic acid therapeutics, where methods may be used for identity, assay, purity, impurity control, sequence confirmation, structural characterization, or stability assessment.
For drug developers and CDMOs, fit-for-purpose analytics should begin with the molecule. A phosphorothioate antisense oligonucleotide, GalNAc-conjugated siRNA, enzymatically synthesized RNA oligonucleotide, and mRNA product do not present the same analytical problem. Each requires a clear understanding of product structure, expected modifications, potential impurity classes, and development-stage needs.
The FDA’s clinical pharmacology guidance for oligonucleotide therapeutics addresses evaluations including QTc prolongation, immunogenicity risk assessment, hepatic and renal impairment, and drug-drug interaction assessment.12 While that guidance is not a chemistry, manufacturing, and controls (CMC) analytics document, it underscores that oligonucleotide therapeutics are being addressed as a distinct development category. On the CMC side, the FDA’s high-resolution MS discussion of synthetic oligonucleotide impurities and ICH Q2(R2)’s fit-for-purpose validation framework provide useful anchors for analytical strategy.1,2
In practice, a strong analytical strategy for highly modified nucleic acid therapeutics should connect several layers of product understanding. First, it should confirm identity and sequence fidelity. Second, it should characterize relevant impurity classes, including deletion/addition sequences and modified backbone, nucleobase, or sugar variants where applicable. Third, it should address modality-specific structural features, such as mRNA cap, poly(A) tail, integrity, fragments, and double-stranded RNA. Fourth, it should use orthogonal methods when a single technique cannot adequately resolve the relevant variants. Finally, it should be validated according to how the method will be used.
For CDMOs, this creates both a technical challenge and a strategic opportunity. Biopharma companies developing nucleic acid therapeutics increasingly need partners that can support method development, impurity identification, advanced mass spectrometry, chromatographic optimization, method transfer, and phase-appropriate validation. Analytical capability becomes part of manufacturing readiness, not a separate quality function added at the end.
As nucleic acid therapeutics continue to evolve, the analytical bar will rise with the molecules themselves. More sophisticated chemistry, delivery, and manufacturing platforms will require equally sophisticated ways to prove what was made, what variants are present, and whether those attributes remain controlled. For highly modified nucleic acid therapeutics, analytics is becoming inseparable from product quality, process understanding, and the ability to move complex medicines through development with confidence.
