Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2025 Sep 1;97(36):19570–19577. doi: 10.1021/acs.analchem.5c02727

A Novel Hybrid LC-MS/MS Methodology for the Quantitative Bioanalysis of Antibody-siRNA Conjugates

Zifeng Song 1, Long Yuan 1,*
PMCID: PMC12444745  PMID: 40888416

Abstract

Small interfering RNA (siRNA) is a new class of oligonucleotide therapeutics that is rapidly growing in drug research and development for the treatment of various diseases. One major challenge for siRNA therapeutics is their inefficient delivery to target tissues by systemic administration. Antibody-siRNA conjugates (ARC) are being developed as a promising approach to enhancing the selectivity and delivery of siRNA payloads to target tissues. A sensitive, selective, and reliable method to quantify ARCs in biological samples is critical to understand their pharmacokinetics, pharmacodynamics, toxicity, and biodistribution properties. There are very limited methods reported for the quantification of ARCs, including reverse transcription-quantitative polymerase chain reaction (RT-qPCR) or hybridization enzyme linked immunosorbent assay (hELISA); however, these methods were still proof-of-concept or suffered from poor specificity or assay performance. LC–MS/MS has been used for the quantification of various modalities, including siRNA or oligonucleotides. The technique offers unique advantages of high specificity, fast method development, and robust assay performance. However, no LC–MS/MS-based method has been reported for the quantitative bioanalysis of ARCs. In this work, we developed a novel hybrid LC–MS/MS methodology for the quantification of ARCs using Ab-siRNA01, an antibody-siRNA conjugate, as the test compound. An antihuman IgG antibody was used as the capture antibody to selectively extract the target ARC analyte from the serum samples. The optimized hybrid LC–MS/MS method was successfully qualified for the quantitation of Ab-siRNA01 in mouse serum over the range of 6.70–3350 ng/mL. The developed method was used to support pharmacokinetic studies of Ab-siRNA01 in mice. This work is the first instance where a hybrid LC–MS/MS method was developed and applied to the quantification of antibody-siRNA conjugates. This novel methodology can also be applied to other ARCs and significantly facilitates their research and development.

graphic file with name ac5c02727_0010.jpg

graphic file with name ac5c02727_0008.jpg

Introduction

In 1998, RNA interference (RNAi) was discovered as a conserved process in which double-stranded RNA induces the cleavage of mRNA with complementary sequence to realize efficient and targeted gene silencing. Three years later, small interfering RNA (siRNA), typically 20–24 nucleotide-long, was identified as the mediator of RNAi in mammalian cells. , Since then, siRNA has been widely used as an indispensable tool for gene-function studies in biology research. This starkly contrasts its bumpy journey of being a novel therapeutic reagent due to the poor drug profile of naked unmodified siRNAs, e.g., susceptible to degradation by endogenous enzymes, large molecule weight (>13 kDa), and poor membrane permeability. Still two decades of continuous efforts and enormous investments have made it possible. In 2018, the first siRNA-based therapeutic, patisiran, was approved by the United States Food and Drug Administration (FDA). Within five years, the FDA has approved six siRNA-based therapeutics, all of which target the liver either through lipid nanoparticle (LNP) formulation or N-acetylgalactosamine (N-GalNAc) conjugation. Feasibility of both delivery methods largely relies on the unique physiological and molecular characters liver tissue possesses, such as large blood supply, high abundance of Kupffer cells, and predominant expression of asialoglycoprotein receptors on hepatocytes. Therefore, efficient and targeted extrahepatic delivery remains the central hurdle to advancing broader clinical applications. However, the concept of targeting ASGR on hepatic cells can be transferred to other types of tissues that might also have tissue-specific receptors/antigens expressed. Monoclonal antibodies (mAbs) as highly specific and long-lasting vehicles have been well-established and successfully used to deliver small-molecule drugs to targeted sites (antibody-drug conjugates (ADCs)). , As such, conjugation of mAbs with siRNAs provides a potential solution to address the delivery issue and expand the therapeutic potential of siRNAs to extrahepatic tissues. As this field has been actively evolving, researchers from both academia and industry reported various conjugation methods focusing on site-specific siRNA coupling to antibodies. , At the same time, they also demonstrated that both the specific binding affinity from mAb and gene silencing capacity from the siRNA payload have been retained on the resultant conjugates. Currently, there are multiple antibody-siRNA conjugate (ARC) therapeutic candidates under clinical development.

While the combination of two different modalities (e.g., small molecule or oligonucleotide with antibody) synergizes the advantages of both, it also dramatically increases molecule complexity, which will usually pose various bioanalytical challenges. These challenges can be further amplified in ARCs due to the high hydrophilicity, multiple negative charges, and large size brought by the siRNA payloads. As such, bioanalytical methodologies applied to ADCs may not be directly translated to ARCs. The heterogeneous nature of ARC leads to increased complexity of the potential analytes; the antibody-siRNA conjugate, free (unconjugated) or total antibody, and free or total siRNA may need to be monitored to understand fully the pharmacokinetics (PK), toxicokinetics (TK), pharmacodynamic (PD), efficacy, and safety of ARC drug candidates. Based on the experience of bioanalysis for ADCs, − multiple assays are typically required to characterize different components of the ARCs in a single PK/PD or safety study in animals. Among the various components of ARC, the antibody-siRNA conjugate, free siRNA payload, and total antibody are the three critical and most monitored analytes. Measurement of total antibody and antibody-siRNA conjugate in circulation are recommended to provide information on PK and in vivo stability of ARCs. It is also important to assess siRNA exposure in target tissues (e.g., brain for CNS targeting ARCs) and major accumulation sites (liver, kidney, etc.) in animal studies to better understand the PK/TK and safety of the ARCs. Free siRNA clears quickly from circulation; thus, the measurement of free siRNA in circulation is often challenging and may not be needed. Bioanalytical methods for total antibody and free siRNA have been well established. Ligand binding assays (LBA) are commonly used for the quantitative bioanalysis of total antibody. Various bioanalytical methods, including hybridization enzyme-linked immunosorbent assay (hELISA), liquid chromatography mass spectrometry (LC–MS), and stem-loop reverse transcription-quantitative polymerase chain reaction (RT-qPCR), have been applied for the measurement of siRNA. − However, for the quantification of ARC, there were only very limited methods reported. − One is hELISA-based assay, in which a locked nucleic acid (LNA)-containing probe was used to hybridize with the siRNA payload to form triplexes for capture and an anti-Fc antibody for detection, or vice versa, an anti-Fc antibody for capture and LNA-containing probe for detection. This method is still a proof-of-concept and needs to be further assessed and optimized to achieve robust utility. The other is an RT-qPCR-based method, which utilized a capture antibody or antigen of the target ARC to capture and enrich the ARC from the biological samples and stem loop RT-qPCR for quantification. , These hELISA or RT-qPCR-based methods can achieve highly sensitive detection of ARCs, typically in the range of ng/mL to mid-pg/mL level. Unfortunately, these methods usually suffer from a lack of specificity and may be affected by the presence of metabolites or degradants in the samples. They also require the development or generation of suitable reagents (e.g., primers, capture, or detection antibodies), which may take significant time and resources and may not be easily available. This may impact the study timelines, especially for discovery stage projects, which often require fast turnaround. In addition, the accuracy and precision of qPCR-based assays usually are not on par with the levels of LBA or LC–MS assays, and their applications in regulated bioanalysis is limited. A liquid chromatography tandem mass spectrometry (LC–MS/MS) assay has the advantages of high specificity, multiplexing, and fast method development. It has been widely used for the quantification of oligonucleotides, including antisense oligonucleotides (ASO) and siRNA. ,,− However, no LC–MS/MS-based method has been reported for the quantitative bioanalysis of ARCs.

In this work, a novel hybrid LC–MS/MS methodology is developed for the quantification of antibody-siRNA conjugates. Ab-siRNA01, an antibody-siRNA conjugate, was used as the model compound for developing the methodology. siRNA01, an HPRT tool siRNA, was used as the siRNA payload and conjugated with a humanized antibody to generate Ab-siRNA01. For LC–MS/MS quantification of siRNAs, antisense strand of the siRNA analyte has been commonly used as the surrogate analyte for the quantification of siRNA. For ARCs, we applied a similar approach using the antisense strand of the siRNA payload on the conjugate as the surrogate analyte of the ARC analyte. The small size of the antisense strand can significantly improve the sensitivity of the LC–MS/MS assay. If the intact ARC analyte is directly analyzed by LC–MS/MS, due to the large size of ARCs, it is impractical to achieve a sensitive and reliable bioanalytical assay. An anti-ARC antibody (a commercially available antihuman IgG antibody in this case) was used to selectively capture and extract the target ARC analyte from the samples. The antisense strand of the siRNA was then eluted from the captured ARC by heat denaturation and analyzed by LC–MS/MS. The immunocapture sample preparation removes potential endogenous inferences and generates a clean sample extract, in combination with the high specificity brought by LC–MS/MS, enabling sensitive, specific, and reliable quantification of ARCs. In addition, one potential issue using the antisense strand as the surrogate analyte is that any free siRNA or other siRNA-containing components in the samples may cause interference with the assay. By using immunocapture with an anti-ARC antibody, mass spectrometry readouts would only be obtained from captured antibodies having the siRNA payload attached. Therefore, only the Ab-siRNA conjugate is measured by the current assay; neither the free antibody nor the free siRNA will be detected. Moreover, the method utilized an elevated temperature (heat denaturation) to unwind the siRNA duplex and release the antisense strand for analysis. This method can be applicable to both cleavable and noncleavable linkers. Key factors that affect the method performance (amount of capture antibody, reagent lot-to-lot variation, and elution temperature) were thoroughly evaluated and optimized. With the optimized conditions, a highly sensitive and reliable hybrid LC–MS/MS method was developed and qualified for the quantification of Ab-siRNA01 in mouse serum. The qualified method has been successfully applied to animal studies. To the best of our knowledge, this is the first report that a hybrid LC–MS/MS methodology was successfully developed for the quantitative bioanalysis of antibody-siRNA conjugates. The developed hybrid LC–MS/MS methodology can also be applied to other ARCs and greatly supports the research and development of ARCs.

Experimental Section

Chemicals and Reagents

Standard materials, including Ab-siRNA01 (Ab-siRNA conjugate analyte), siRNA01 (siRNA payload on the conjugate), AS01 (antisense strand of siRNA01, a 23-mer oligonucleotide), and ASO-IS (an analogue internal standard (IS) that is a 20-mer oligonucleotide), were either purchased from WuXi AppTech (Tianjin, China) or obtained from Biogen (Cambridge, MA). Biotinylated mouse antihuman IgG (H + L) secondary antibody (capture antibody), Dynabeads MyOne streptavidin C1, 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), and phosphate buffered saline (PBS), pH 7.4, were ordered from Thermo Fisher Scientific (Waltham, MA). N,N-Dimethylcyclohexylamine (DMCHA) was obtained from TCI America (Portland, OR). Tween 20 was ordered from Sigma-Aldrich (St. Louis, MO).

Solution and Standard Preparation

Stock and working solutions of Ab-siRNA01 were prepared in 1% bovine serum albumin (BSA) in PBS spiked with 0.05% Tween 20 (PBST). Calibration standards were freshly prepared by appropriate dilution of Ab-siRNA01 standard working solutions with blank mouse serum to generate final concentrations ranging from 6.70 to 3350 ng/mL with ten levels of standards on the calibration curve (equivalent siRNA concentration ranging from 1.00 to 500 ng/mL). Quality control (QC) samples, including a lower limit of quantitation (LLOQ) QC at 6.70 ng/mL, low QC at 20.1 ng/mL, mid QC at 536 ng/mL, and high QC at 2512.5 ng/mL (equivalent siRNA01 concentrations of 1.00, 3.00, 80.0, and 375 ng/mL, respectively), were also prepared in blank mouse serum and stored in a −70 °C freezer.

Sample Extraction

50 μL of standards, QCs, and study samples were added to a 96-well plate and diluted with 400 μL of PBST. Streptavidin Dynabeads (10 mg/mL) were coated with biotinylated mouse antihuman IgG (20 μg of antibody per 100 μL of beads) at room temperature for 2 h. After 3 cycles of washing and resuspending with PBST, tested volumes of capture antibody-coated beads were added into each sample for immunocapture of the ARC analyte. The mixture was incubated at room temperature for 2 h. After washing at room temperature twice with 500 μL of PBST and one time with 500 μL of washing buffer (10 mM Tris, 50 mM NaCl, 1 mM EDTA and 0.05% Tween 20 in water), the captured analyte was eluted under different tested temperatures in 200 μL of elution solvent (10% mobile phase C in water with 0.05% rat plasma) spiked with 25 ng/mL ASO-IS. Sample extraction procedures were performed on a KingFisher APEX (Thermo Fisher Scientific, Waltham, MA). The detailed sample extraction procedures are listed in Figure .

1.

1

Open in a new tab

Overall workflow of the hybrid LC–MS/MS methodology for the quantification of antibody-siRNA conjugates.

LC–MS/MS Conditions

Mass spectrometric analyses were performed on a QTRAP 6500+ (SCIEX, Framingham, MA) using multiple reaction monitoring (MRM) in electrospray negative ionization mode. Detailed MRM transitions and parameters of the analyte and IS were summarized in Table S1. Chromatographic separation was achieved on a Clarity Oligo-XT column (50 × 2.0 mm, 1.7 μm particle size, Phenomenex, Terrace, CA) installed on a Nexera X2 HPLC equipped with three LC-30 AD pumps (Shimadzu, Columbia, MD). A ternary mobile phase system was used, with 100% water as mobile phase A, 100% acetonitrile as mobile phase B, and 150 mM DMCHA and 250 mM HFIP in 100% acetonitrile as mobile phase C. Column temperature was set at 70 °C. Detailed flow rates and gradients of binary phases A and B can be found in Table S2. The injection volume was 10 μL.

Pharmacokinetic Study in Mice

A single-dose PK study was conducted for Ab-siRNA01 in C57BL/6 mice. Mice (n = 3) were dosed with 6.7 mg/kg of Ab-siRNA01 by intravenous (IV) injection. For PK evaluation, blood samples were collected at 0.25, 2, 6, 24, 48, and 72 h postdose. Collected blood samples were allowed to clot for a minimum of 30 min followed by centrifugation at 1500g at 4 °C for 15 min. The supernatant serum was collected and stored in a −70 °C freezer until extraction and analysis.

Results and Discussion

Selection of Surrogate Analyte for the Quantification of ARCs

It is challenging to use LC–MS/MS to directly measure ARCs because of their unique biophysical properties, e.g., large molecular weight (>100 kDa), highly negatively charged siRNA payload, and potential molecular heterogeneous antibody. For LC–MS/MS quantification of siRNA, the antisense strand of siRNA is usually selected as the surrogate analyte since it is the pharmacologically active strand. Similarly, for ARCs, the antisense strand of the conjugated siRNA payload is the pharmacologically active component. Moreover, often the 3′ end of the siRNA sense strand is covalently conjugated to the antibody with a linker. ARCs may undergo cleavage, metabolism, or degradation in vivo and generate various sense strand-related products (free sense strand, sense strand with linker, etc.), making the sense strand not suitable as the surrogate analyte for a targeted LC–MS/MS assay. Therefore, we selected the antisense strand of ARCs as the surrogate analyte. In the current study, AS01, the antisense strand of Ab-siRNA01, was used as the surrogate analyte and its standard compound used for optimizing LC–MS/MS parameters. All the calibration standards and QC samples were prepared with Ab-siRNA01, the intact ARC analyte.

LC–MS/MS Method Development

A full scan was carried out to obtain the relative abundance distribution of multiply charged ion species of AS01 (Figure S1a). The ion at m/z 775 was the most abundant ion and thus was selected as the precursor ion and fragmented to generate the product ions. As shown in Figure S1b, product ions at m/z 79 (phosphate ion) and 95 (phosphorothioate ion) are the most dominant fragments due to their stoichiometry inherited from the backbone of the oligonucleotide. Though m/z 79 had a higher intensity, m/z 95 was selected as the quantifier product ion to avoid potential interference from the endogenous oligonucleotides. Another precursor ion (m/z 704) with different charge state coupled with its phosphorothioate product ion was also monitored and used as the qualifier transition for ensuring analyte identity (Table S1).

Ion-pair reversed-phase chromatography using a combination of alkylamine and fluoroalcohol has become the gold-standard for improving the chromatography and MS sensitivity of oligonucleotides. We applied this well-established approach for AS01 and adopted the LC conditions from previously published methods with slight modification. − Specifically, DMCHA was used as the alkylamine ion-pair reagent for the improved analyte retention, separation, and peak shape; and HFIP (fluoroalcohol) was used for the best chromatographic separation and mass spectrometric response. To ensure that the composition of the sample extract for LC injection is consistent and compatible with the LC starting gradient, the eluted extract was dissolved in 25 mM HFIP and 15 mM DMCHA in acetonitrile and water (v/v 1:9) with 0.05% rat plasma. 10% acetonitrile and 0.05% rat plasma were included to prevent potential nonspecific binding loss of the antisense strand analyte.

Analog IS has been commonly used as the IS for LC–MS/MS bioanalysis of oligonucleotides. , Similarly, we used an analogue oligonucleotide (a 20-mer oligonucleotide with different sequence) as the IS for this method. The IS was added at the end of the sample extraction to compensate for variations in subsequent operation and LC–MS/MS analysis. The method using the analogue IS showed satisfactory performance with good accuracy, precision, and robustness.

Sample Extraction Strategy and Method Optimization

Since the antisense strand was used as the surrogate analyte for the quantification of ARC, any free siRNA or other compounds containing the siRNA component in the samples may cause interference to the assay. To overcome this challenge, a sample preparation method that can selectively capture the ARC analyte, and not the free siRNA or other interfering compounds, is required for the success of the method. The antibody component of Ab-siRNA01 is a humanized IgG antibody. The assay developed in the current work aimed to mainly support animal studies for early stage projects. Therefore, we chose an antihuman IgG antibody to selectively extract the ARC analyte from the samples. AS01, the antisense strand of Ab-siRNA01, was eluted from the captured ARC by heat denaturation and subsequently analyzed by specific and sensitive LC–MS/MS. The hybrid LC–MS/MS workflow for the quantification of ARCs is schematized in Figure .

For early-stage or discovery projects such as the current study, generic commercial antihuman antibodies can be used as the capture antibody for convenience and availability to enable fast method development and study support. When the project moves to late or development stage, more specific antibodies (e.g., anti-id antibody specific to the antibody component) can be generated and used as the capture antibody, which would further improve the selectivity, sensitivity, and robustness of the assay.

Hybrid LC–MS/MS methods have been commonly used in ADC bioanalysis for the quantification of an antibody-conjugated payload. For these methods, usually enzymatic digestion (e.g., cathepsin B, papain) or chemical cleavage is used to cleave the linker and release a specific conjugated payload (payload attached with part of the linker) for analysis. , For ADCs with noncleavable linkers, nonspecific proteolysis of the antibody portion of the ADC may be used to generate payload attached with a peptide fragment for analysis. However, multiple forms of conjugated payload (e.g., payload attached with different peptide fragments) may be generated depending on the position and number of conjugation sites, which will add significant complexity and challenges for direct quantification of the conjugated payload. Therefore, for ADCs with noncleavable linkers, ADC concentrations are often measured by ligand binding assays that cannot generate direct information on the drug to antibody ratio (DAR). One advantage of the method developed here is that elevated temperature (heat denaturation) was used to break the siRNA duplex and release the antisense strand for analysis. This allows easy and convenient generation of the surrogate analyte for LC–MS detection. There is no limitation on the linker used for ARCs, and the method can be easily applied to both ARCs with cleavable linkers and ARCs with noncleavable linkers.

Amount of Capture Antibody on Recovery

We first evaluated the effect of the amount of capture antibody on the extraction recovery of the Ab-siRNA01 conjugate. Different volumes (20, 30, and 50 μL, which correspond to 0.2, 0.3, and 0.5 mg of beads, respectively) of capture antibody-coated streptavidin beads were compared for the extraction of Ab-siRNA01 high QC (5360 ng/mL) samples in serum. Based on the vendor’s specification, the maximum loading capacity of streptavidin beads is approximately 20 μg biotinylated IgG per mg of streptavidin beads. Thus, the estimated capture antibodies added into each sample were 4, 6, and 10 μg, respectively, which are approximately 15-, 22-, and 37-fold of the amount of the ARC analyte. As shown in Figure , the extraction recovery of Ab-siRNA01 with 20 μL of beads is around 66%. Increasing the volume of the beads to 30 and 50 μL improved the recovery to around 80%, but no significant difference in recovery was seen between 30 and 50 μL of beads. This suggests that the optimal recovery for Ab-siRNA01 was achieved by using approximately 30 μL of beads. Therefore, 30 μL of beads (22-fold of capture antibody) was used for the subsequent optimization and assessment experiments.

2.

2

Open in a new tab

Amount of capture antibody on the recovery of Ab-siRNA01.

We also evaluated the recovery of Ab-siRNA01 at different concentrations across the assay range. Overall, the immunocapture method achieved consistent recoveries between low- and high-concentration samples. As shown in Figure , similar recoveries of around 70% were achieved for low and mid QCs at 20.1 and 3350 ng/mL. For the high QCs (5360 ng/mL), the recovery decreased slightly to around 62%, which may be due to the saturation of the binding between the capture antibody and the ARC analyte at high concentrations. To avoid the potential saturation and ensure the linearity of the assay, we reduced the curve range to 500-fold (6.70–3350 ng/mL) from the originally planned 1000-fold. This also helped to reduce the amount of capture antibody used and reduced the cost of the assay.

3.

3

Open in a new tab

Assessment of recovery consistency across different concentrations of Ab-siRNA01 (QCL: 20.1 ng/mL, QCM: 3350 ng/mL, QCH: 5360 ng/mL).

Elution Temperature on Recovery

Since the antisense strand of the captured Ab-siRNA01 conjugate is used as the surrogate analyte, complete release of the antisense strand from the captured conjugate is critical for achieving maximal recovery. Heating at a temperature higher than the melting temperature (T m) of the siRNA is a commonly used approach to unwind the siRNA duplex and elute the antisense strand. , Heating was also used to elute the antisense strand in this work, and the effect of elution temperature on recovery was evaluated. Four temperatures (65, 75, 85, and 95 °C) that are above the estimated T m of siRNA01 (38 °C) were tested. As shown in Figure , the recovery at 65 °C is the lowest, at around 40%, which may be because the temperature was not high enough to completely break the siRNA duplex and elute the antisense strand. Optimal recovery was obtained when the elution temperature was increased to 75–85 °C. The recovery at 75 °C is slightly higher than that at 85 °C (60% vs 56%), but the difference is not statistically significant. The recovery decreased to 50% when the elution temperature was further increased to 95 °C. The reason for the decreased recovery at 95 °C is unknown. It may be related to the denaturation or degradation of the captured Ab-siRNA01 conjugate at high temperature. Therefore, 75 °C is selected as the elution temperature for the following assessments of assay performance. In general, the optimal elution temperature is correlated to the T m of the siRNA payload or the binding affinity between two strands of the siRNA duplex: the higher the T m of the siRNA, the higher the elution temperature needed to ensure complete elution of the antisense strand.

4.

4

Open in a new tab

Effect of elution temperature on the recovery of Ab-siRNA01.

Lot-to-Lot Variation of Capture Antibody

As mentioned above, the assay developed in this work was aimed to support early-stage animal studies that focus more on the speed and efficiency of method development and study support. Therefore, a generic polyclonal antihuman IgG antibody that recognizes and binds to the Fc region of a humanized antibody was selected as the capture reagent because of its relatively low cost and easy availability (commercially available from vendors). However, one major concern is the lot-to-lot variation of the reagents. For polyclonal antibodies, each individual lot is a highly complex population of various antibody molecules that may have different epitopes with different binding affinities. Thus, different lots of capture reagents may have different extraction recoveries of the ARC analyte, which may affect the performance of the assay. We assessed the lot-to-lot variation of the capture antibody using three different lots purchased from the same vendor. As expected, different extraction recoveries were observed when using different lots of capture antibodies (Figure ). One lot (Lot 1) has significantly higher recovery compared to the other two lots (82% vs 51% and 56%), which may be due to its stronger binding affinity to the target ARC analyte. Though the other two lots showed a relatively low recovery, they were able to provide satisfactory performance for the hybrid LC–MS/MS assay. Later, when Lot 1 was depleted of stock, Lot 3 was used as the capture antibody to complete the assay assessment and qualification and support the animal studies.

5.

5

Open in a new tab

Lot-to-lot variation of capture antibody on the recovery of Ab-siRNA01.

To ensure the consistent performance of the assay between runs, the same lot of reagents are preferred to be used as the capture antibody. When required to support multiple studies or perform the assay over a long period of time, it is recommended that an adequate amount of the same lot be stocked to avoid lot-to-lot variation. When the switch is required to a new lot, the new lot should be assessed and qualified before use to ensure acceptable assay performance. Later when the projects move to the development stage or if resources allow, a recombinant monoclonal antihuman antibody with good binding affinity and selectivity is recommended to be generated as the capture reagent to ensure the consistency and performance of the assay. For human assays, the generic antihuman antibody cannot be used as the capture reagent. An anti-idiotypic antibody specific to the target ARC can be generated and used as the capture antibody for the assay.

Interference Assessment of Free siRNA Payload to the Ab-siRNA Conjugate

In real study samples, there may be free siRNA that coexists with the ARC analyte due to linker instability or siRNA payload cleavage from the Ab-siRNA conjugate in vivo. The potential nonspecific binding of the free siRNA to the capture antibodies may cause interference to the assay. Therefore, we evaluated if the presence of free siRNA in the samples may cause any interference and jeopardize the specificity of the assay. 200 ng/mL of siRNA01 was spiked into blank serum and Ab-siRNA01 serum QCs at different concentration levels (20.1, 3350, and 5360 ng/mL Ab-siRNA01, corresponding to 3.00, 500, and 800 ng/mL of siRNA01 on the conjugate). The samples were extracted and analyzed by the developed hybrid LC–MS/MS method. The accuracy and precision of the measured QCs were used to assess the interference and assay specificity. As shown in Table , all the QCs met the acceptance criteria with the average accuracy (% nominal) at 97.3–104%, and the precision (% CV) within 8.7%. Blank serum spiked with 200 ng/mL siRNA01 was also analyzed, and no peak was observed at the expected retention time of the analyte. All of these results indicated the absence of interference from the coexisting free siRNA in the samples.

1. Accuracy and Precision of Ab-siRNA01 QCs in Mouse Serum with the Presence of 200 ng/mL of siRNA01.

nominal conc. 20.1 ng/mL 3350 ng/mL 5360 ng/mL
measured conc. (ng/mL) 20.9 3491 5842
  19.3 3410 5172
  20.8 3276 5856
  17.3 3430 5394
mean 19.6 3402 5566
S.D. 1.7 90.5 339.1
n 4 4 4
% CV 8.7 2.7 6.1
% nominal 97.3 102 104
Open in a new tab

Assay Qualification and Performance

With the optimized conditions, the method was qualified for the quantification of Ab-siRNA01 in mouse serum. Ten levels of calibration standards ranging from 6.70 to 3350 ng/mL and four levels of QCs (6.70, 20.1, 536, and 2512.5 ng/mL) in mouse serum were prepared and used for the evaluation. Figure S2 shows a representative calibration curve of Ab-siRNA01 in mouse serum. A 1/x 2 weighted quadratic regression model provided the best fit of Ab-siRNA01 ranging from 6.70 to 3350 ng/mL (500-fold concentration range), with a coefficient of correlation (r) of 0.9948. All standards met the acceptable criteria of within ±15% (±20% at LLOQ) of the nominal concentration across the assay range. Figure shows the representative chromatograms of blank, LLOQ, and high QC samples of Ab-siRNA01 in mouse serum. There was no endogenous interference observed around the retention time of the analyte in the blank serum, suggesting good specificity of the optimized assay. The signal-to-noise ratio of the LLOQ sample (6.70 ng/mL) was larger than 5 (Figure ), demonstrating good assay sensitivity. Table summarizes the accuracy and precision of Ab-siRNA01 QCs in mouse serum. Within-batch precision (% CV) of all four QC levels were all within 10.0%; and the mean accuracies ranged from 91.1% to 101% of the nominal concentrations of the analyte. The benchtop stability of Ab-siRNA01 in mouse serum was assessed using low and high QC placed on ice for 4 h. The evaluation results were summarized in Table S3. All QCs were within the 15% acceptance criteria, which demonstrated the stability of Ab-siRNA01 under the tested condition. These results showed acceptable performance of the developed assay using the optimized assay parameters.

6.

6

Open in a new tab

Representative LC–MS/MS chromatograms of Ab-siRNA01 (left) and the internal standard, ASO-IS (right) in a blank serum spiked with 25 ng/mL ASO-IS (A), a mouse serum spiked with Ab-siRNA01 at the LLOQ concentration of 6.70 ng/mL (B), and the high QC of 2512.5 ng/mL (C).

2. Accuracy and Precision of Ab-siRNA01 QC Samples in Mouse Serum.

  LLOQ QC low QC mid QC high QC
  (6.70 ng/mL) (20.1 ng/mL) (536 ng/mL) (2512.5 ng/mL)
measured conc. (ng/mL) 6.19 18.8 533 2472
  6.90 20.6 544 2492
  6.63 16.4 480 2747
  6.49 17.4 544 2459
mean 6.55 18.3 525 2543
S.D. 0.30 1.83 30.7 137
n 4 4 4 4
% CV 4.5 10.0 5.9 5.4
% nominal 97.8 91.1 98.0 101.2
Open in a new tab

Application to a Mouse Study

The qualified assay was applied to a single-dose study in mice to assess the serum pharmacokinetics of Ab-siRNA01. One run was conducted for the analysis of Ab-siRNA01 in mouse serum, and the run met the acceptance criteria for standards and QCs (% CV ≤ 15%; % bias ≤ 15%, 20% for LLOQ). For the three levels of QCs (low, mid, and high), the mean % bias was 4.9–7.0%, and the % CV was 4.4–5.2%, showing good assay performance for analyzing real PK study samples. Figure shows the PK profile of Ab-siRNA01 in mouse serum following a single IV administration of 6.7 mg/kg of Ab-siRNA01. In this study, the total antibody concentrations in serum were also measured using an LBA method. The total antibody concentrations matched well with the Ab-siRNA conjugate concentrations at the early time points, which further confirmed the good accuracy of the hybrid LC–MS method for the quantification of Ab-siRNA01. At later time points, the concentrations of the Ab-siRNA conjugate became lower than the total antibody concentrations, indicating releasing of the siRNA payload from the ARC in vivo in circulation.

7.

7

Open in a new tab

Serum concentration (mean ± SD) vs time profiles of Ab-siRNA01 in mice after single IV dose of Ab-siRNA01 at 6.7 mg/kg.

Conclusion

A hybrid LC–MS/MS methodology was successfully developed for the quantification of antibody-siRNA conjugates in biological samples. For Ab-siRNA01, the test ARC analyte in this work, an assay was developed and qualified over the range of 6.70–3350 ng/mL in mouse serum. The method achieved satisfactory performance in terms of specificity, sensitivity, accuracy, and precision. The qualified method has been successfully applied to support animal pharmacokinetic studies of Ab-siRNA01. This is the first report of the use of hybrid LC–MS/MS for the quantitative bioanalysis of ARCs. The methodology can also be applied to the bioanalysis of other ARCs and greatly facilitates their research and development.

Compared to qPCR-based assays, this new hybrid LC–MS/MS method achieved highly specific quantification of ARCs with much tighter accuracy and precision (met the 15%/20% acceptance criteria, which was typically used for small-molecule bioanalysis). The clean sample extract resulted from the immunoaffinity sample preparation helped to obtain a good sensitivity (LLOQ of 6.70 ng/mL, equivalent to 1.00 ng/mL siRNA) that is comparable to LBA methods. Additionally, the developed methodology can be applied to both cleavable and noncleavable linkers for ARCs. With these combined advantages, the hybrid LC–MS/MS methodology has great potential to be more widely used and become the method of choice for the quantitative bioanalysis of ARCs.

Supplementary Material

ac5c02727_si_001.pdf (152.6KB, pdf)

Acknowledgments

The authors thank Dr. Patrick Trapa for his review of the manuscript.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.5c02727.

  • Mass spectrometry parameters for analyte and internal standard; LC gradient and flow rate condition for the LC separation; bench-top stability of Ab-siRNA01 in mouse serum; mass spectra of AS01, the antisense strand of siRNA01; and representative calibration curve of Ab-siRNA01 in mouse serum (PDF)

The authors declare no competing financial interest.

References

  1. Moazzam M., Zhang M., Hussain A., Yu X., Huang J., Huang Y.. The landscape of nanoparticle-based siRNA delivery and therapeutic development. Mol. Ther. 2024;32(2):284–312. doi: 10.1016/j.ymthe.2024.01.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Caplen N. J., Parrish S., Imani F., Fire A., Morgan R. A.. Specific inhibition of gene expression by small double-stranded RNAs in invertebrate and vertebrate systems. Proc. Natl. Acad. Sci. U.S.A. 2001;98(17):9742–9747. doi: 10.1073/pnas.171251798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Elbashir S. M., Harborth J., Lendeckel W., Yalcin A., Weber K., Tuschl T.. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature. 2001;411(6836):494–498. doi: 10.1038/35078107. [DOI] [PubMed] [Google Scholar]
  4. Jadhav V., Vaishnaw A., Fitzgerald K., Maier M. A.. RNA interference in the era of nucleic acid therapeutics. Nat. Biotechnol. 2024;42(3):394–405. doi: 10.1038/s41587-023-02105-y. [DOI] [PubMed] [Google Scholar]
  5. Tang Q., Khvorova A.. RNAi-based drug design: considerations and future directions. Nat. Rev. Drug Discovery. 2024;23:341–364. doi: 10.1038/s41573-024-00912-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Schroeder A., Levins C. G., Cortez C., Langer R., Anderson D. G.. Lipid-based nanotherapeutics for siRNA delivery. J. Intern. Med. 2010;267(1):9–21. doi: 10.1111/j.1365-2796.2009.02189.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. McDougall R., Ramsden D., Agarwal S., Agarwal S., Aluri K., Arciprete M., Brown C., Castellanos-Rizaldos E., Charisse K., Chong S., Cichocki J., Fitzgerald K., Goel V., Gu Y., Guenther D., Habtemariam B., Jadhav V., Janas M., Jayaraman M., Kurz J.. et al. The Nonclinical Disposition and Pharmacokinetic/Pharmacodynamic Properties of N-Acetylgalactosamine–Conjugated Small Interfering RNA Are Highly Predictable and Build Confidence in Translation to Human. Drug Metab. Dispos. 2022;50(6):781. doi: 10.1124/dmd.121.000428. [DOI] [PubMed] [Google Scholar]
  8. Dugal-Tessier J., Thirumalairajan S., Jain N.. Antibody-Oligonucleotide Conjugates: A Twist to Antibody-Drug Conjugates. J. Clin. Med. 2021;10(4):838. doi: 10.3390/jcm10040838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Malecova B., Burke R. S., Cochran M., Hood M. D., Johns R., Kovach P. R., Doppalapudi V. R., Erdogan G., Arias J. D., Darimont B., Miller C. D., Huang H., Geall A., Younis H. S., Levin A. A.. Targeted tissue delivery of RNA therapeutics using antibody-oligonucleotide conjugates (AOCs) Nucleic Acids Res. 2023;51(12):5901–5910. doi: 10.1093/nar/gkad415. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cuellar T. L., Barnes D., Nelson C., Tanguay J., Yu S. F., Wen X., Scales S. J., Gesch J., Davis D., van Brabant Smith A., Leake D., Vandlen R., Siebel C. W.. Systematic evaluation of antibody-mediated siRNA delivery using an industrial platform of THIOMAB-siRNA conjugates. Nucleic Acids Res. 2015;43(2):1189–1203. doi: 10.1093/nar/gku1362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kaur S., Xu K., Saad O. M., Dere R. C., Carrasco-Triguero M.. Bioanalytical Assay Strategies for The Development of Antibody–Drug Conjugate Biotherapeutics. Bioanalysis. 2013;5(2):201–226. doi: 10.4155/bio.12.299. [DOI] [PubMed] [Google Scholar]
  12. Wang J., Gu H., Liu A., Kozhich A., Rangan V., Myler H., Luo L., Wong R., Sun H., Wang B., Vezina H. E., Deshpande S., Zhang Y., Yang Z., Olah T. V., Aubry A.-F., Arnold M. E., Pillutla R., DeSilva B.. Antibody–drug conjugate bioanalysis using LB-LC–MS/MS hybrid assays: strategies, methodology and correlation to ligand-binding assays. Bioanalysis. 2016;8(13):1383–1401. doi: 10.4155/bio-2016-0017. [DOI] [PubMed] [Google Scholar]
  13. Mou S., Huang Y., Rosenbaum A. I.. ADME Considerations and Bioanalytical Strategies for Pharmacokinetic Assessments of Antibody-Drug Conjugates. Antibodies. 2018;7:41. doi: 10.3390/antib7040041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Mu R., Yuan J., Huang Y., Meissen J. K., Mou S., Liang M., Rosenbaum A. I.. Bioanalytical Methods and Strategic Perspectives Addressing the Rising Complexity of Novel Bioconjugates and Delivery Routes for Biotherapeutics. BioDrugs. 2022;36(2):181–196. doi: 10.1007/s40259-022-00518-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Qin Q., Gong L.. Current Analytical Strategies for Antibody–Drug Conjugates in Biomatrices. Molecules. 2022;27:6299. doi: 10.3390/molecules27196299. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Wang L., Ji C.. Advances in quantitative bioanalysis of oligonucleotide biomarkers and therapeutics. Bioanalysis. 2016;8(2):143–155. doi: 10.4155/bio.15.234. [DOI] [PubMed] [Google Scholar]
  17. Tremblay G. A., Oldfield P. R.. Bioanalysis of siRNA and oligonucleotide therapeutics in biological fluids and tissues. Bioanalysis. 2009;1(3):595–609. doi: 10.4155/bio.09.66. [DOI] [PubMed] [Google Scholar]
  18. Yuan L., Dupuis J.-F., Mekhssian K.. A Novel Hybridization LC-MS/MS Methodology for Quantification of siRNA in Plasma, CSF and Tissue Samples. Molecules. 2023;28(4):1618. doi: 10.3390/molecules28041618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Song Z., Lu A., Yuan L.. Analyte and probe melting temperature guided method development strategy for hybridization LC-MS/MS quantification of siRNAs. J. Pharm. Biomed. Anal. 2025;253:116556. doi: 10.1016/j.jpba.2024.116556. [DOI] [PubMed] [Google Scholar]
  20. Humphreys S. C., Thayer M. B., Campuzano I. D. G., Netirojjanakul C., Rock B. M.. Quantification of siRNA-Antibody Conjugates in Biological Matrices by Triplex-Forming Oligonucleotide ELISA. Nucleic Acid Ther. 2019;29(3):161–166. doi: 10.1089/nat.2018.0770. [DOI] [PubMed] [Google Scholar]
  21. Tan M., Vernes J.-M., Chan J., Cuellar T. L., Asundi A., Nelson C., Yip V., Shen B., Vandlen R., Siebel C., Meng Y. G.. Real-time quantification of antibody–short interfering RNA conjugate in serum by antigen capture reverse transcription–polymerase chain reaction. Anal. Biochem. 2012;430(2):171–178. doi: 10.1016/j.ab.2012.08.018. [DOI] [PubMed] [Google Scholar]
  22. Toshima A., Shiraishi Y., Shinmi D., Kagawa Y., Enokizono J.. Comprehensive Analyses of the Intracellular and in Vivo Disposition of Fab– Small Interfering RNA Conjugate to Identify Key Issues to Improve Its in Vivo Activity. Drug Metab. Dispos. 2023;51(3):338–347. doi: 10.1124/dmd.122.001098. [DOI] [PubMed] [Google Scholar]
  23. Murphy A., Hill R., Berna M.. Bioanalytical Approaches to Support the Development of Antibody-Oligonucleotide Conjugate (AOC) Therapeutic Proteins. Xenobiotica. 2024;54(8):552–562. doi: 10.1080/00498254.2024.2339983. [DOI] [PubMed] [Google Scholar]
  24. van Dongen W. D., Niessen W. M. A.. Bioanalytical LC–MS of therapeutic oligonucleotides. Bioanalysis. 2011;3(5):541–564. doi: 10.4155/bio.11.8. [DOI] [PubMed] [Google Scholar]
  25. Basiri B., Bartlett M.. LC–MS of oligonucleotides: applications in biomedical research. Bioanalysis. 2014;6(11):1525–1542. doi: 10.4155/bio.14.94. [DOI] [PubMed] [Google Scholar]
  26. Sutton J. M., Kim J., El Zahar N. M., Bartlett M. G.. Bioanalysis and Biotransformation of Oligonucleotide Therapeutics by Liquid Chromatography-Mass Spectrometry. Mass Spectrom. Rev. 2021;40(4):334–358. doi: 10.1002/mas.21641. [DOI] [PubMed] [Google Scholar]
  27. Liu A., Cheng M., Zhou Y., Deng P.. Bioanalysis of Oligonucleotide by LC-MS: Effects of Ion Pairing Regents and Recent Advances in Ion-Pairing-Free Analytical Strategies. Int. J. Mol. Sci. 2022;23(24):15474. doi: 10.3390/ijms232415474. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Li P., Dupuis J.-F., Vrionis V., Mekhssian K., Magee T., Yuan L.. Validation and application of hybridization liquid chromatography-tandem mass spectrometry methods for quantitative bioanalysis of antisense oligonucleotides. Bioanalysis. 2022;14(9):589–601. doi: 10.4155/bio-2022-0015. [DOI] [PubMed] [Google Scholar]
  29. Yuan L., Dupuis J.-F., Vrionis V., Mekhssian K., Magee T.. A validated capillary microsampling liquid chromatography–tandem mass spectrometry method for quantification of antisense oligonucleotides in mouse serum. Bioanalysis. 2023;15(12):683–694. doi: 10.4155/bio-2023-0096. [DOI] [PubMed] [Google Scholar]
  30. Jiang D., Cook B. E., Pickel T., Yuan L.. Quantitative Determination of Click-Reactive Antisense Oligonucleotide and Its Reactivity in Rat Brains by Hybridization LC-MS/MS to Support Pretargeted PET Imaging. Anal. Chem. 2025;97(23):12155–12163. doi: 10.1021/acs.analchem.5c00644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Jiang D., Yuan L.. Microflow LC–MS/MS to improve sensitivity for antisense oligonucleotides bioanalysis: critical role of sample cleanness. Bioanalysis. 2022;14(21):1365–1376. doi: 10.4155/bio-2022-0201. [DOI] [PubMed] [Google Scholar]
  32. Jiang D., Li P., Yuan L.. Bioanalysis of free antisense oligonucleotide payload from antibody–oligonucleotide conjugate by hybridization LC-MS/MS. Bioanalysis. 2024;16(15):1–10. doi: 10.1080/17576180.2024.2368339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Guimaraes G., Yuan L., Li P.. Antisense Oligonucleotide In Vitro Protein Binding Determination in Plasma, Brain, and Cerebral Spinal Fluid Using Hybridization LC-MS/MS. Drug Metab. Dispos. 2022;50(3):268–276. doi: 10.1124/dmd.121.000751. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ac5c02727_si_001.pdf (152.6KB, pdf)

Articles from Analytical Chemistry are provided here courtesy of American Chemical Society

RESOURCES