ASO/siRNA RNA-seq Off-Target Detection

ASO/siRNA RNA-seq Off-Target Detection

1. Background

Antisense oligonucleotides (ASOs) and small interfering RNAs (siRNAs), as next-generation precision nucleic acid therapeutics, achieve targeted gene regulation at the transcriptomic level through sequence-specific recognition of target RNAs. They have emerged as critical modalities for the treatment of rare diseases, hereditary disorders, neurodegenerative conditions, and metabolic diseases. ASOs primarily exert their effects via RNase H-mediated mRNA degradation or steric hindrance mechanisms, whereas siRNAs rely on the RNA-induced silencing complex (RISC) to cleave target mRNAs, conferring superior targeting specificity. Compared with conventional small molecule drugs and antibody-based therapeutics, nucleic acid drugs offer distinct advantages including broad target coverage, short design cycles, and sequence customizability—enabling therapeutic development against targets traditionally considered undruggable. These features have secured a pivotal strategic position for nucleic acid drugs in the global biopharmaceutical pipeline.

With multiple ASO and siRNA drugs achieving regulatory approval worldwide—spanning indications such as spinal muscular atrophy, hereditary transthyretin amyloidosis, acute hepatic porphyria, and hypercholesterolemia—the clinical value and commercial viability of nucleic acid therapeutics have been robustly validated. The global nucleic acid drug development pipeline continues to expand, with a large number of candidate molecules advancing into preclinical and clinical development stages, reflecting broad market prospects and establishing this field as one of the fastest-growing segments within biopharmaceuticals.

However, the success rate of nucleic acid drug development is critically dependent on precise sequence design and comprehensive safety evaluation. The safety risks of ASOs/siRNAs arise primarily from two dimensions: (1) sequence-dependent (hybridization-dependent) off-target effects—due to high sequence similarity, drugs may engage in unintended hybridization with non-target genes, causing hybridization-dependent off-target effects (HDOTs) that result in aberrant expression of non-targeted genes, representing a key source of preclinical toxicity signals; and (2) sequence-independent effects—for example, class effects associated with ASOs such as complement activation and coagulation interference, as well as immunostimulatory responses triggered by siRNAs through innate immune receptors (e.g., TLR7/TLR8). In recent years, ICH, major global regulatory authorities (FDA, EMA, PMDA, CDE), and the Oligonucleotide Safety Working Group (OSWG) have issued dedicated technical guidance documents imposing clear and systematic requirements for the safety evaluation of nucleic acid drugs:

Figure 1. Overview of Key Guidance Documents from ICH/FDA/EMA/PMDA/CDE

In November 2024, the FDA issued the Guidance for Industry: Nonclinical Safety Assessment of Oligonucleotide-Based Therapeutics, explicitly requiring that prior to IND submission, comprehensive hybridization-dependent off-target assessments covering the transcriptome, nuclear genome, and mitochondrial genome must be conducted for candidate sequences and their metabolites, with cross-analysis integrating both RNA-seq transcriptomic experimental validation and in silico computational prediction. Systematic and standardized safety evaluation has become a fundamental prerequisite for the successful advancement of nucleic acid drugs toward IND filing.

Figure 2. FDA Requirements for Sequence-Dependent Off-Target Hybridization Assessment of Oligonucleotide Therapeutics

(The guidance explicitly requires systematic assessment of potential off-target sites using in silico methods combined with in vitro approaches, including RNA-seq.)

2. Detection Principles

RNA-seq off-target detection is based on high-throughput transcriptomic sequencing technology. By comparing genome-wide expression profiles between ASO/siRNA-treated groups and control groups, the assay systematically identifies unintended gene expression changes caused by sequence homology.

2.1 Core Detection Strategy

For siRNA: The primary focus is detection of off-target effects at the mature mRNA level. Because siRNAs mediate mRNA degradation via the RISC complex in the cytoplasm, an mRNA-enrichment library construction strategy (Oligo-dT magnetic bead enrichment of poly(A)⁺ RNA) is employed to efficiently capture off-target-induced mRNA downregulation events.

For ASO: Library construction strategies are flexibly selected based on ASO type. Gapmer-type ASOs (RNase H-dependent) can employ mRNA-enrichment library construction; whereas splice-switching ASOs (which modulate pre-mRNA splicing in the nucleus) are recommended to use whole-transcriptome library construction (rRNA depletion) to comprehensively assess both nuclear and cytoplasmic off-target risks.

2.2 Analytical Workflow

Significant downregulated genes are identified using professional differential expression algorithms (|Log2FC| > 1, p < 0.05). Results are integrated with sequence homology alignment (allowing ≤3 nucleotide mismatches) to cross-validate computational predictions against RNA-seq experimental findings, enabling precise identification of sequence-dependent off-target sites. Functional enrichment analysis is also performed to evaluate the biological significance and potential safety risks of the off-target effects.

Figure 3. RNA-seq Off-Target Detection Technical Workflow

3. Technical Advantages and Innovation

3.1 Whole-Transcriptome Coverage for Comprehensive Off-Target Risk Identification

Leveraging high-throughput sequencing technology, our assay achieves unbiased detection of genome-wide expression profiles. Compared with conventional qPCR-based validation methods, this approach simultaneously monitors expression changes across tens of thousands of genes, eliminating the risk of missing potential off-target sites and ensuring comprehensive and accurate off-target assessment.

3.2 High Sensitivity for Precise Quantification of Off-Target Magnitude

With a sequencing depth of 5–10 Gb per sample, the assay reliably detects expression changes in low-abundance genes and accurately quantifies the degree of upregulation or downregulation at off-target loci, providing precise quantitative data for subsequent therapeutic window evaluation.

3.3 Multi-Dimensional Analysis for In-Depth Mechanistic Interpretation of Off-Target Effects

By integrating sequence homology analysis, differential expression analysis, functional enrichment analysis, and protein-protein interaction network analysis across multiple bioinformatics layers, our platform not only identifies off-target loci but also deeply dissects the biological functions, signaling pathway associations, and potential toxicity risks of off-target genes—providing comprehensive data support for safety evaluation.

3.4 Regulatory Compliance to Support IND Submission

Detection methods and analytical workflows strictly adhere to the FDA's 2024 latest guidance and OSWG industry consensus. Delivered analytical reports include complete data quality assessment, off-target site identification, functional annotation, and risk evaluation, and can be directly incorporated into the nonclinical safety module of IND applications, meeting regulatory review requirements.

3.5 Flexible Library Construction Strategies for Optimized Detection Performance

Tailored library construction approaches are provided based on the mechanisms of action of different oligonucleotide drug types: siRNA projects employ mRNA-enrichment library construction; ASO projects offer a choice between mRNA-enrichment or whole-transcriptome library construction depending on the compound type—ensuring detection comprehensiveness while optimizing cost-effectiveness and achieving precise strategy matching.

3.6 Cross-Validation of Computational Prediction and Experimental Verification

Innovatively integrating in silico sequence homology prediction with RNA-seq experimental validation, the platform performs cross-verification analysis between computationally predicted potential off-target sites and genes exhibiting actual expression changes. This dual-confirmation mechanism substantially improves the accuracy of off-target site identification, effectively distinguishing true off-target effects from biological noise and reducing false-positive rates.

3.7 Systematic Off-Target Analysis Framework

A standardized off-target analysis pipeline has been established, encompassing full-sequence matching analysis, seed region matching analysis (for siRNA), dose-dependent assessment, tissue expression relevance analysis, gene functional importance evaluation, and other multi-dimensional components. This framework constructs a comprehensive off-target risk assessment matrix, providing clear decision-making support for candidate compound optimization.

4. Application Scenarios

RNA-seq off-target detection technology is applicable across multiple key stages of oligonucleotide drug development:

(1) Candidate Compound Selection: In the preclinical stage, transcriptomic off-target analysis is used to screen candidate sequences with superior safety profiles.

(2) IND Submission Support: Fulfills regulatory requirements for data completeness in off-target evaluation, providing analytical reports compliant with FDA/NMPA standards.

(3) Clinical Safety Monitoring: Provides molecular mechanism-level data support for adverse event tracing during clinical trials.

5. Sample Report Content Overview

To ensure clients gain a comprehensive understanding of service deliverables, we provide professional and standardized analytical reports. The following uses a siRNA RNA-seq off-target analysis report as an example to showcase core analytical modules (Note: ASO off-target analysis differs strategically, primarily in that off-target site identification is based on full-length sequence homology analysis rather than seed region matching, and library construction may use either whole-transcriptome or mRNA-enrichment approaches; the remaining analytical workflow is essentially identical). The report encompasses a complete analytical chain from data quality control to off-target risk assessment, with all conclusions based on rigorous statistical testing, directly supporting IND submission and R&D decision-making.

（1）Data Quality Assessment and Differential Expression Analysis: The report first presents quality control results for sequencing data, including key metrics such as Q20/Q30 ratios and mapping rates for both raw and quality-filtered data, ensuring that all downstream analyses are built upon high-quality data. Differential expression analysis is then performed using the DESeq2 algorithm, with volcano plots providing an intuitive panoramic view of transcriptome-wide expression changes. The number of upregulated and downregulated genes is tabulated, and the distribution of significantly differentially expressed genes is clearly annotated on volcano plots.

Figure 4. Summary Table of Differentially Expressed Genes

Figure 5. Volcano Plot of Differential Expression Analysis

（2）Target Gene Knockdown Efficiency Validation: For the core target gene designated by the client (exemplified as nudt21, Gene ID: ENSG00000167005), the report provides a dedicated assessment of expression changes. Example results show that the target gene exhibits a log2FoldChange of −0.653 (FoldChange = 0.636, padj = 0.0012), demonstrating a significant downregulation trend and validating the on-target knockdown efficacy of the siRNA.

Figure 6. Target Gene Differential Expression Analysis Table

（3）Off-Target Site Prediction and Cross-Validation Analysis: Based on the seed region sequence matching principle for siRNA (nucleotides 2–8), the report systematically identifies four potential off-target categories: mer8 (complete 8-nucleotide match), mer7m8 (7-nucleotide match including position 8), mer7A1 (7-nucleotide match including position 1 adenosine), and mer6 (6-nucleotide match), with the predicted number of off-target genes tabulated for each category. A key innovation is the cross-comparison of computational predictions with actual differentially expressed genes, filtering for high-confidence off-target genes that demonstrate both sequence complementarity and confirmed expression changes. In the example report, the four match types yield 383, 2,129, 1,023, and 2,245 predicted off-target genes, respectively; intersection with differentially expressed genes reveals 1 shared gene (of the mer7m8 type), which is identified as the high-priority authentic off-target site requiring focused attention.

Figure 7. Intersection Table of Differentially Expressed Genes and Predicted Off-Target Genes

（4）Statistical Evaluation and Visualization of Off-Target Effects: The Kolmogorov-Smirnov (K-S) test is applied to quantify the overall impact of the siRNA on each category of predicted off-target genes. In the example report, all four match types yield p-values < 0.05, indicating that the effect of the siRNA on these predicted off-target genes is statistically significantly different from background gene expression, thereby confirming the existence of off-target effects. The report also provides empirical cumulative distribution function (ECDF) curves, visually demonstrating expression distribution differences between each off-target gene category and background genes; the degree of curve shift reflects off-target risk magnitude. This statistical evidence represents a key data element scrutinized by regulatory authorities.

Figure 8. K-S Test Statistical Results Table (including D-statistic and p-value for all four mer-k categories)

Figure 9. ECDF Cumulative Distribution Curve Plot

6. Service Options and Workflow

Zhuhai GeneRulor Biotechnology offers flexible service models; clients may select the most appropriate option based on their specific circumstances:

6.1 Service Options

Option 1: Client-Provided Samples

(1) Applicable Scenario: Client has completed cell transfection experiments.

(2) Sample Submission Requirements: Post-transfection cell samples or extracted RNA samples (treatment group + control group; ≥3 biological replicates).

(3) Service Scope: RNA quality inspection, library construction, high-throughput sequencing, and bioinformatics analysis.

(4) Project Turnaround: 35 business days.

Option 2: GeneRulor Full-Process Service

(1) Applicable Scenario: Client provides ASO/siRNA compounds; GeneRulor Biotechnology handles the complete experimental workflow.

(2) Sample Submission Requirements: ASO/siRNA compound and target information; specification of cell line selection and experimental design requirements.

(3) Service Scope: Cell culture, compound transfection, RNA extraction and quality inspection, library construction, high-throughput sequencing, and bioinformatics analysis.

(4) Project Turnaround: 45 business days.

6.2 Deliverables

(1) Professional RNA-seq off-target analysis report (PDF format), encompassing data quality assessment, differential expression analysis, off-target site identification, functional enrichment analysis, and complete content.

(2) Raw sequencing data (FASTQ format).

(3) Intermediate analysis files and result files.

(4) Supplementary technical documentation meeting IND submission requirements, available upon client request.

7. Sample Requirements

8. References

[1] Yoshida T, et al. Evaluation of off-target effects of gapmer antisense oligonucleotides using human cells. Genes Cells. 2019; 24(11):827-835.

[2] Andersson P, et al. Assessing Hybridization-Dependent Off-Target Risk for Therapeutic Oligonucleotides: Updated Industry Recommendations. Nucleic Acid Ther. 2024.

[3] Goyenvalle A, et al. Considerations in the preclinical assessment of the safety of antisense oligonucleotides. Nucleic Acid Ther. 2023; 33(1):1-16.

[4] Lindow M, et al. Assessing unintended hybridization-induced biological effects of oligonucleotides. Nat Biotechnol. 2012; 30(10):920-923.

[5] Kamola PJ, et al. In silico and in vitro evaluation of exonic and intronic off-target effects form a critical element of therapeutic ASO gapmer optimization. Nucleic Acids Res. 2015; 43(18):8638-8650.

[6] U.S. Food and Drug Administration. Nonclinical Safety Assessment of Oligonucleotide-Based Therapeutics Guidance for Industry. Draft Guidance. November 2024.